Language：English   Publishing type：Research paper (scientific journal)   Publisher：OPTICAL SOC AMER  

In general, viewers are more attracted to local features in images at a shorter viewing distance and to global features in images at a longer viewing distance. However, numerical analysis of the effect of viewing distance on human texture perception and how the perception of global and local changes under certain conditions are still undetermined. In this paper, we present statistical prediction of the relationship between the domination ratio of global and local features and the viewing distances under the control of several factors, using the logistic regression model. We synthesized textures by separately controlling global and local textural features using a texture model based on mathematical morphology, namely the primitive, grain, and point configuration texture model. Visual sensory tests were carried out on 80 subjects during two sets of experiments. The collected data were statistically analyzed using logistic regression and Akaike information criteria. Besides the main factor of viewing distance, the factors including gender, changing the order of viewing positions, and prior knowledge were also shown quantitatively to have significant influence on human texture perception. Our results showed that (1) local features of a texture were more attractive to females than males, (2) the first impression might have affected subsequent decisions in texture perception, and (3) subjects who had prior knowledge (supervised) were more sensitive to the changes in global and local dominance. (4) Regarding the interactions of the factors, prior knowledge reduced the effects of individual differences and perception condition differences on human texture perception. This study is dedicated to the construction of numerical relationships between viewing distance and human texture perception as well as to cognitive investigation of biases in global and local perceptions. (C) 2013 Optical Society of America

DOI： 10.1364/JOSAA.30.001394

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Institute of Electronics, Information and Communications Engineers (IEICE)  

DOI： 10.1587/transfun.2024sml0002

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)   Publisher：Japan Society of Kansei Engineering  

DOI： 10.5057/ijae.ijae-d-24-00023

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Institute of Electrical and Electronics Engineers Inc.  

Calcification regions are sometimes observed on dental panoramic radiographs and it is known that these regions are a sign of vascular disease. It has been pointed out that the detection of calcification regions in dental panoramic radiographs can encourage patients to undergo medical checkups by a physician. Medical checkups can prevent the sudden onset of vascular disease. For this purpose, a method of automatically detecting calcification regions using an object detector based on deep learning has been proposed. Although this method significantly reduces the number of false positives compared with conventional methods based on image features, its detection accuracy is still insufficient. In this paper, we propose a method of detecting calcification regions in dental panoramic radiographs using a novel object detector based on deep learning. On the basis of PraNet, which has been increasingly applied to medical image processing in recent years, we introduce the Double U-Net structure and enhance the prediction accuracy for the initial guidance region. The proposed method can improve the detection accuracy for calcification regions. The experimental results show that the proposed method improves the detection performance compared with other methods.

DOI： 10.1109/ISCIT52804.2021.9590604

Scopus

researchmap

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Society for Imaging Science and Technology  

We found a similarity between the effect of temporal chord transition in music and temporal color transition in visual perception. Human visual impressions between a color arrangement and a temporal transition from a modified arrangement to the original arrangement were compared in experiments with sensory tests. The experimental results showed that the consonance of color arrangements can be enhanced by a "pivoted" transition, in which some appearance parameters like hues and color tones are preserved, from a dissonant arrangement in comparison with a static presentation of the original arrangement. The results suggest that an enhancement effect of consonance similar to resolution of chords in music exists in a higher order mechanism of human color vision.

Scopus

researchmap

researchmap

Language：Japanese   Publisher：Kyoto University  

CiNii Books

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：BRITISH INST RADIOLOGY  

Objectives: This study proposed a new automated screening system based on a hybrid genetic swarm fuzzy (GSF) classifier using digital dental panoramic radiographs to diagnose females with a low bone mineral density (BMD) or osteoporosis.
Methods: The geometrical attributes of both the mandibular cortical bone and trabecular bone were acquired using previously developed software. Designing an automated system for osteoporosis screening involved partitioning of the input attributes to generate an initial membership function (MF) and a rule set (RS), classification using a fuzzy inference system and optimization of the generated MF and RS using the genetic swarm algorithm. Fivefold cross-validation (5-FCV) was used to estimate the classification accuracy of the hybrid GSF classifier. The performance of the hybrid GSF classifier has been further compared with that of individual genetic algorithm and particle swarm optimization fuzzy classifiers.
Results: Proposed hybrid GSF classifier in identifying low BMD or osteoporosis at the lumbar spine and femoral neck BMD was evaluated. The sensitivity, specificity and accuracy of the hybrid GSF with optimized MF and RS in identifying females with a low BMD were 95.3%, 94.7% and 96.01%, respectively, at the lumbar spine and 99.1%, 98.4% and 98.9%, respectively, at the femoral neck BMD. The diagnostic performance of the proposed system with femoral neck BMD was 0.986 with a confidence interval of 0.942-0.998. The highest mean accuracy using 5-FCV was 97.9%. with femoral neck BMD.
Conclusions: The combination of high accuracy along with its interpretation ability makes this proposed automatic system using hybrid GSF classifier capable of identifying a large proportion of undetected low BMD or osteoporosis at its early stage.

DOI： 10.1259/dmfr.20160076

Web of Science

researchmap

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：JOURNAL HARD TISSUE BIOLOGY  

Among patients who visited our hospital with a chief complaint of disorders involving the oromandibular region, with suspected carotid artery calcification based on panoramic radiography, and who underwent computed tomography for a detailed examination of these disorders, there were four patients in whom a definitive diagnosis of carotid artery calcification was made. We herein provide an overview of these cases, along with a brief review of the relevant literature. When imaging findings suggestive of carotid artery calcification are obtained on panoramic radiographs, dentists can request detailed examinations and treatment at specialized medical institutions in the early stage by providing patients with information about the risks of atherosclerotic diseases. Early detection and treatment may reduce the severity or risks of complications and damage caused by atherosclerotic diseases, as well as possibly contributing to improved quality of life. In this study, we evaluated carotid artery calcification visible on panoramic radiographs. However, it seems that further detailed analysis is needed to examine the association between carotid artery calcification and a past history of risk factors for arteriosclerosis.

Web of Science

researchmap

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In this paper, we propose a framework to assess visual complexity of paintings. This framework provides a machine learning scheme for investigating the relationship between human visual complexity perception and low-level image features. Since the global and local characteristics of paintings affect human's holistic impression and detail perception, we design a set of methods to extract the features that represent the global and local characteristics of paintings. By feature selection, we look into the role that each image feature plays in assessing visual complexity. Then the selected features are combined by a Support Vector Machine for classification. Experimental results indicate that the proposed work can predict the visual complexity perception of paintings with the accuracy of 88.13%, which is highly close to the assessments given by humans. Compared with the conventional measure of complexity, our approach considers human visual perception and performs more efficiently in assessing visual complexity of painting images.

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

Purpose: To prevent low bone mineral density (BMD), that is, osteoporosis, in postmenopausal women, it is essential to diagnose osteoporosis more precisely. This study presented an automatic approach utilizing a histogrambased automatic clustering (HAC) algorithm with a support vector machine (SVM) to analyse dental panoramic radiographs (DPRs) and thus improve diagnostic accuracy by identifying postmenopausal women with low BMD or osteoporosis. Materials and Methods: We integrated our newly-proposed histogram-based automatic clustering (HAC) algorithm with our previously-designed computer-aided diagnosis system. The extracted moment-based features (mean, variance, skewness, and kurtosis) of the mandibular cortical width for the radial basis function (RBF) SVM classifier were employed. We also compared the diagnostic efficacy of the SVM model with the back propagation (BP) neural network model. In this study, DPRs and BMD measurements of 100 postmenopausal women patients (aged &gt
50 years), with no previous record of osteoporosis, were randomly selected for inclusion. Results: The accuracy, sensitivity, and specificity of the BMD measurements using our HAC-SVM model to identify women with low BMD were 93.0% (88.0%-98.0%), 95.8% (91.9%-99.7%) and 86.6% (79.9%-93.3%), respectively, at the lumbar spine
and 89.0% (82.9%-95.1%), 96.0% (92.2%-99.8%) and 84.0% (76.8%-91.2%), respectively, at the femoral neck. Conclusion: Our experimental results predict that the proposed HAC-SVM model combination applied on DPRs could be useful to assist dentists in early diagnosis and help to reduce the morbidity and mortality associated with low BMD and osteoporosis. © 2013 by Korean Academy of Oral and Maxillofacial Radiology.

DOI： 10.5624/isd.2013.43.3.153

Scopus

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

In this paper, we propose a novel texture analysis method capable of estimating multiple primitives, which are elements repetitively arranged to compose a texture, in multi-structured textures. The approach is based on a texture description model that uses mathematical morphology, called the "Primitive, Grain, and Point Configuration (PGPC)" texture model. The estimation of primitives based on the PGPC texture model involves searching the optimal structuring element for primitives according to a size distribution function and a magnification. The proposed method achieves the following two improvements: (1) the simultaneous estimation of a multiple number of primitives in multi-structured textures with a ranking of primitives on the basis of their significance. and (2) the introduction of flexibility in the process of magnification to obtain a higher degree of fitness of large grains. With a computational combination of different primitives, the method provides an ordered priority to denote the significance of elements. The promising performance of the proposed method is experimentally shown by a comparison with conventional methods.

DOI： 10.1587/transfun.E95.A.1759

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：OPTICAL SOC JAPAN  

In our previous work we determined that five important characteristics affect the perception of visual complexity of a texture: regularity, roughness, directionality, density, and understandability. In this paper, a set of objective methods for measuring these characteristics is proposed: regularity is estimated by an autocorrelation function; roughness is computed based on local changes; directionality is measured by the maximum line-likeness of edges in different directions; and density is calculated from the edge density. Our analysis shows a significant correlation between the objective measures and subjective evaluations. In addition, for the estimation of understandability, a new approach is proposed. We asked the respondents to name each texture, and then we sorted all these names into different types, including names that were similar. We discovered that understandability is affected by two factors of a texture: the maximum number of similar names assigned to a specific type and the total number of types.

DOI： 10.1007/s10043-012-0047-1

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：BIOMED CENTRAL LTD  

Background: Early diagnosis of osteoporosis can potentially decrease the risk of fractures and improve the quality of life. Detection of thin inferior cortices of the mandible on dental panoramic radiographs could be useful for identifying postmenopausal women with low bone mineral density (BMD) or osteoporosis. The aim of our study was to assess the diagnostic efficacy of using kernel based support vector machine (SVM) learning regarding the cortical width of the mandible on dental panoramic radiographs to identify postmenopausal women with low BMD.
Methods: We employed our newly adopted SVM method for continuous measurement of the cortical width of the mandible on dental panoramic radiographs to identify women with low BMD or osteoporosis. The original X-ray image was enhanced, cortical boundaries were determined, distances among the upper and lower boundaries were evaluated and discrimination was performed by a radial basis function. We evaluated the diagnostic efficacy of this newly developed method for identifying women with low BMD (BMD T-score of -1.0 or less) at the lumbar spine and femoral neck in 100 postmenopausal women (>= 50 years old) with no previous diagnosis of osteoporosis. Sixty women were used for system training, and 40 were used in testing.
Results: The sensitivity and specificity using RBF kernel-SVM method for identifying women with low BMD were 90.9% [95% confidence interval (CI), 85.3-96.5] and 83.8% (95% CI, 76.6-91.0), respectively at the lumbar spine and 90.0% (95% CI, 84.1-95.9) and 69.1% (95% CI, 60.1-78.6), respectively at the femoral neck. The sensitivity and specificity for identifying women with low BMD at either the lumbar spine or femoral neck were 90.6% (95% CI, 92.0-100) and 80.9% (95% CI, 71.0-86.9), respectively.
Conclusion: Our results suggest that the newly developed system with the SVM method would be useful for identifying postmenopausal women with low skeletal BMD.

DOI： 10.1186/1471-2342-12-1

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

Estimation of a suitable window shape and appropriate weights in weighted median filters is one of important problems. In this study, we formulate the design of weighted median filter as an optimization problem, and estimate optimal filters directly from degraded images. In our previous work, we estimated optimal window shapes and weights by using a Genetic Algorithm (GA) with a fixed size of window as a constraint in optimization. To determine an appropriate window size is difficult but essential since it depends on both a type of texture and a noise rate. Here, we optimize weighted median filters without the constraint in the size of filters. Numerical experiments show that our method design a filter with a suitable size to both a size of pattern in textures and the noise rates. In addition, we compare the designed filters with the filters obtained by conventional supervised design and another unsupervised design methods.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

This study aimed to realize and compare the effectiveness of cortical and trabecular bone measures in a computer-aided system based on support vector machine (SVM) for identifying postmenopausal women with low BMD or osteoporosis. Dental panoramic radiographs of 69 postmenopausal women, as well as bone mineral density (BMD) assessments at the lumbar spine and the femoral neck were used. Average width and variance of continuous measurements of mandibular cortical bone and distribution of average length and angle (direction) of trabecular bone segments on panoramic radiographs were measured by computer-aided systems and used as inputs. The accuracy of the cortical measures using average width and variance was 87% and of the trabecular measures using average length and angle was 65% with RBF kernel-SVM method for diagnosing women with low BMD at the lumbar spine. Likewise, the accuracy of combined cortical and trabecular bone measures was 88%. Our results suggest that the cortical bone measurement is more appropriate than the trabecular bone measurement for triage screening of osteoporosis. © 2012 IEEE.

DOI： 10.1109/ICSMC.2012.6377702

Scopus

researchmap

Language：Japanese  

CiNii Books

researchmap

DOI： 10.1111/j.2041-1626.2011.00095.x

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

A method of estimating dual primitives in a textural image is proposed. This method is based on the Primitive, Grain, and Point Configuration (PGPC) texture model, which regards a texture as an arrangement of grains derived from one or a few primitives. Appropriate primitives can be represented by morphological structuring elements estimated from a texture. Conventional primitive estimation methods estimate only one primitive from each textural image. However, they do not work well on textural images that contain more than one basic structure, since two or more types of grain cannot be generated from only one primitive. The proposed method simultaneously estimates two optimal structuring elements of a texture. The experimental results show that the proposed method provides more representative estimations than the conventional method.

DOI： 10.1587/transfun.E94.A.1165

Web of Science

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Books

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：SPIE-INT SOC OPTICAL ENGINEERING  

The aim of this study is to develop a computer-aided osteoporosis diagnosis system that automatically determines the inferior cortical width of the mandible continuously on dental panoramic radiographs to realize statistically more robust measurements than the conventional one-point measurements. The cortical width was continuously measured on dental panoramic radiographs by enhancing the original image, determining cortical boundaries, and finally evaluating the distance between boundaries continuously throughout the region of interest. The diagnostic performance using the average width calculated from the continuous measurement was compared with BMD at lumbar spine and femoral neck in 100 postmenopausal women of whom 50 to the development of the tool and 50 to its validation with no history of osteoporosis was evaluated. We experimentally showed the superiority of our method with improved sensitivity and specificity of identifying the development subjects were 90.0% and 75.0% in women with low spinal BMD and 81.8% and 69.2% in those with low femoral BMD, respectively. The corresponding values in the validation subjects were 93.3% and 82.9% at the lumbar spine and 92.3% and 75.7% at the femoral neck, respectively in terms of efficacy for diagnosing osteoporosis. We also assessed the diagnosis and classification of women with osteoporosis using support vector machine employing the average and variance of the continuous measurements gave excellent discrimination ability. It yields sensitivity and specificity of 90.9% and 83.8%, respectively with lumbar spine and 90.0% and 69.1%, respectively with femoral neck BMD. Performance comparison and simplicity of this method indicate that our computer-aided system is readily applicable to clinical practice.

DOI： 10.1117/12.878101

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：SPRINGER-VERLAG BERLIN  

An improved morphological estimation method of textural elements based on the Primitive, Grain, and Point Configuration (PGPC) texture model is proposed. The PGPC texture model has shown promising applications such as noise removal, texture modification, and texture synthesis. However, the estimation is not always successful since the magnification process of the primitive to ensure the assumption that the grains are homothetic does not always fit to each image. We propose in this paper a novel estimation method introducing more flexibility into homotheticity's conventional assumption of the grains, and exploring a suitable structuring element for the homothetic magnification process of the primitive. Experimental results show that the proposed method provides more representative grains than the conventional method.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

Visual complexity perception is an important issue in the fields of psychology and computer vision because it leads to the better understanding of the nature of human perception as well as the properties of the objects being perceived. In this study, five important characteristics of texture images that affect visual complexity perception are identified: regularity, understandability, roughness, directionality, and density. Among these, understandability is a deterministic characteristic, which reflects the viewer's prior knowledge and experience. These characteristics significantly affect the visual complexity perception of texture images. In order to achieve our objective, we carried out two experiments involving visual complexity assessment and paired comparison evaluation with 30 respondents. We applied correlation analysis, factor analysis, and multidimensional scaling to analyze the collected data. The experimental results showed that most of the human impressions of visual complexity can be explained by the perceived characteristics of texture images. © 2011 IEEE.

DOI： 10.1109/ICBAKE.2011.13

Scopus

researchmap

Language：English   Publisher：Japan Society of Kansei Engineering  

Since the surface texture of materials often affects human visual impressions as much as or more than the design, shape, or color properties, texture characteristics have been studied as features of object identification. We have been investigating the effect of texture on visual impression and objective identification using black fabrics that do not exhibit any effects of color. Our studies showed that visual impressions of texture correspond to complex micro-components and global structures of image features of those textures. Our results also showed that some important elements influence human visual impressions and identification of textures. Because of a variety of fibrous structures, it is not easy to provide a systematic analysis of clothing materials. Nevertheless, developing the method and collecting data on these elements and their effects using these image features will be important. To make this research applicable for wider use, we have been studying precisely what it is about an arbitrary texture that influences human visual impressions and sensibility. As a new step, in this paper, a texture is altered and transformed using the parameter estimation method of texture based on mathematical morphology, which is often used for extracting image components that are useful for representation and description. A texture is decomposed into a primitive and grain arrangement which correspond to local and global characteristics, respectively. Different textures are created by modifying the primitive and the arrangement to investigate the effects of modifications of local and global features. The relationship between the parameters and visual impressions of the modified textures were evaluated. This study shows that influence of both local and global structures of the texture along with their combinations and mutual interactions are important for identification of human visual impression.

DOI： 10.5057/kei.10.11

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

In this paper we propose an unsupervised method of optimizing structuring elements (SEs) used for impulse noise reduction in texture images through the opening operation which is one of the morphological operations. In this method a genetic algorithm (GA) which can effectively search wide search spaces is applied and the size of the shape of the SE is included in the design variables. Through experiments it is shown that our new approach generally outperforms the conventional method.

DOI： 10.1587/transfun.E93.A.2196

Web of Science

researchmap

Language：English  

CiNii Books

researchmap

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：OPTICAL SOC JAPAN  

A method of evaluating human visual impressions of gray scale textures using morphological manipulation is proposed. To study the effects of textural features on human Kansei, we introduced a texture analysis method based on mathematical morphology. Kansei is a Japanese word for sensibility or emotion. Kansei engineering is an approach to connect human sensibility with engineering applications. The proposed method allows us to manipulate global and local properties of a texture separately. Variations of textures were generated by repetitively modifying arranged objects and configurations of the arrangements of original textures. The manipulated textures were presented to human respondents and the similarity of those textures based on human impressions was evaluated. Hierarchical clustering was applied to the similarity matrix generated from respondents' observations. The results of the human evaluation were compared with that of the objective similarity evaluation adopting six global textural features. The global features such as density, regularity, and directionality of the point configurations were shown to have significant effects on human visual impressions and identification of textures. In the case of a texture without significant characteristics in its point configuration, local features such as grain shape have an effect on visual impressions.

DOI： 10.1007/s10043-010-0016-5

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

This study aimed to realize a newly developed method of continuous measurements of the cortical width of the lower border of the mandible, between the upper and lower margins of the cortical bone to identify women with low bone mineral density (BMD) or osteoporosis. An automatic clustering algorithm is applied to obtain a robust estimate of the cortical width. This continuous measurement method provides more accurate cortical width than the conventional one-point method. The system's efficacy in identifying low BMD at the lumbar spine and femoral neck in 100 postmenopausal women (≥50 years) with no history of osteoporosis was evaluated. The mandibular cortical width below the mental foramen was measured by enhancing the original image of the panoramic radiograph, determining cortical boundaries and evaluating the distance between boundaries continuously, by applying the clustering algorithm to a significant portion of the histogram. It is experimentally shown that the improved sensitivity and specificity provided more stable and significant diagnostic accuracy than the conventional one-point method.

Scopus

researchmap

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

A method of estimating dual primitives in a textural image is proposed. This method is based on the Primitive, Grain, and Point Configuration (PGPC) texture model, which regards a texture as an arrangement of grains derived from one or a few primitives. Appropriate primitives can be represented by morphological structuring elements estimated from a texture. Conventional primitive estimation methods estimate only one primitive from each textural image. However, they do not work well on textural images that contain more than one basic structure, since two or more types of grain cannot be generated from only one primitive. The proposed method simultaneously estimates two optimal structuring elements of a texture. The experimental results show that the proposed method provides more representative estimations than the conventional method. ©2010 IEEE.

DOI： 10.1109/ISCIT.2010.5664896

Scopus

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

Measurement of an individual molar provides rich information for forensic personal identification. We propose a computer-based system for extracting an individual molar from dental panoramic radiographs. A molar is obtained by extracting the region-of-interest, separating the maxilla and mandible, and extracting the boundaries between teeth. The proposed system is almost fully automatic; all that the user has to do is clicking three points on the boundary between the maxilla and the mandible.

DOI： 10.1587/transinf.E92.D.2287

Web of Science

researchmap

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：BRITISH INST RADIOLOGY  

Objectives: An eroded inferior cortex of the mandible detected on dental panoramic radiographs is useful for identifying post-menopausal won-ten with low skeletal bone mineral density (BMD) or osteoporosis. The purposes of this study were to develop a computer-aided diagnosis (CAD) system that automatically determines cortical erosion of the mandible on denial panoramic radiographs and to assess the validation of this CAD system.
Methods: A CAD system was developed based on mathematical morphology for identifying post-menopausal women with low skeletal BMD or osteoporosis, based on World Health Organization criteria, by identifying whether the endosteal margin of mandibular cortical bone was eroded. The sensitivity, specificity, predictive value, accuracy and likelihood for a positive risk result were calculated using dichotomous 2 x 2 tables using 100 panoramic radiographs.
Results: For identifying women with low skeletal BMD, the sensitivity and specificity were 76.8% and 61.1%, respectively, the positive and negative predictive values were 90.0% and 36.7%, accuracy was 74.0% and the likelihood for positive risk was 1.96. The respective values for identifying women with osteoporosis were 94.4% and 43.8%, the positive and negative predictive values were 48.6% and 93.3%, respectively, the accuracy was 62.0% and the likelihood for positive risk was 1.68.
Conclusion: Our results suggest that a CAD system applied to dental panoramic radiographs may be useful for identifying post-menopausal women with low skeletal BNID or osteoporosis.

DOI： 10.1259/dmfr/68621207

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：ELSEVIER SCIENCE INC  

Mandibular cortical erosion detected on dental panoramic radiographs (DPRs) may be useful for identifying women with osteoporosis, but little is known about the variation in diagnostic efficacy of observers worldwide. The purpose of this study was to measure the accuracy in identifying women at risk for osteoporosis in a worldwide group of observers using DPRs. We constructed a website that included background information about osteoporosis screening and instructions regarding the interpretation of mandibular cortical erosion. DPRs of 100 Japanese postmenopausal women aged 50 years or older who had completed skeletal bone mineral measurements by dual energy X-ray absorptiometry were digitized at 300 dpi. These were displayed on the website and used for the evaluation of diagnostic efficacy. Sixty observers aged 25 to 66 years recruited from 16 countries participated in this study. These observers classified cortical erosion into one of three groups (none, mild to moderate, and severe) on the website via the Internet, twice with an approximately 2-week interval. The diagnostic efficacy of the Osteoporosis Self-Assessment Tool (CST), a simple clinical decision rule based on age and weight, was also calculated and compared with that of cortical erosion. The overall mean sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the 60 observers in identifying women with osteoporosis by cortical erosion on DPRs were 82.5, 46.2, 46.7, and 84.0%, respectively. Those same values by the OST index were 82.9, 43.1, 43.9, and 82.4%, respectively. The intra-observer agreement in classifying cortical erosion on DPRs was sufficient (weighted kappa values > 0.6) in 36 (60%) observers. This was significantly increased in observers who specialized in oral radiology (P < 0.05). In the 36 observers with sufficient intra-observer agreement, the overall mean sensitivity, specificity, PPV, and NPV in identifying women with osteoporosis by any cortical erosion were 83.5, 48.7, 48.3, and 85.7%, respectively. The mean PPV and NPV were significantly higher in the 36 observers with sufficient intra-observer agreement than in the 24 observers with insufficient intra-observer agreement. Our results reconfirm the efficacy of cortical erosion findings in identifying postmenopausal women at risk for osteoporosis, among observers with sufficient intra-observer agreement Information gathered from radiographic examination is at least as useful as that gathered from the OST index. (C) 2008 Elsevier Inc. All rights reserved.

DOI： 10.1016/j.bone.2008.03.014

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

Estimation of a suitable window shape and appropriate weights in weighted median filters is one of important problems. In this paper, we propose a new unsupervised design method for the window shapes and weights of weighted median filters. This technique is applied to texture images corrupted by impulse noise. As the performance measure for optimization by simulated annealing, rank-ordered absolute differences (ROAD) statistic is used for the estimation of the window shapes and weights, respectively. The simulation results show that the proposed method can design the weighted median filters with almost same performance compared to the one by the supervised technique.

Web of Science

researchmap

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：ELSEVIER SCIENCE BV  

This paper proposes a new method of image thresholding by using cluster organization from the histogram of an image. A new similarity measure proposed is based on inter-class variance of the clusters to be merged and the intra-class variance of the new merged cluster. Experiments on practical images illustrate the effectiveness of the new method. (c) 2006 Elsevier B.V. All rights reserved.

DOI： 10.1016/j.patrec.2006.02.022

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：SPRINGER LONDON LTD  

Introduction: Mandibular inferior cortical width manually measured on dental panoramic radiographs may be useful for identifying postmenopausal women with low skeletal bone mineral density (BMD). Automatic measurement of cortical width may enable us to identify a large number of postmenopausal women with suspected low skeletal BMD. The purposes of this study were to develop a computer-aided system for measuring mandibular cortical width on dental panoramic radiographs and clarify the diagnostic efficacy of this system. Methods: Panoramic radiographs of 100 postmenopausal women who had had BMD assessments of the lumbar spine and the femoral neck were used in this study. Experienced oral radiologist determined the position of the mental foramen on 100 digitized dental panoramic radiographs. After determination of the mental foramen, mandibular cortical width below the mental foramen was measured automatically with a computer-aided system by identifying the area of interest, enhancing the original image, determining inner and outer margins of the cortex, and selecting an appropriate point. Cortical width measured by this system was compared with BMD of the lumbar spine and the femoral neck. Results: There were statistically significant correlation between cortical width measured by the computer-aided system and spinal BMD (r=0.50) and femoral neck BMD (r=0.54). These correlations were similar with those between cortical width by manual measurement and skeletal BMD. Sensitivity and specificity for identifying postmenopausal women with low spinal BMD by the computer-aided system were about 88.0% and about 58.7%, respectively. Those for identifying postmenopausal women with low femoral neck BMD by this system were about 87.5% and about 56.3%, respectively. Conclusion: Our results suggest that our computer-aided system may be useful for identifying postmenopausal women with low skeletal BMD.

DOI： 10.1007/s00198-005-0045-2

Web of Science

researchmap

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：SPRINGER-VERLAG BERLIN  

This paper proposes a method of texture analysis using morphological size distribution. Our framework is based on the concept that a texture is described by estimation of primitive, size distribution of grains derived from the primitive, and spatial distribution of the grains. We concentrate on estimation of primitive using an assumption on grain size distribution. We assume a model that grains are derived from one primitive, and a uniform size distribution since we consider target textures containing grains of various sizes. Thus the structuring element used for the measurement of size distribution is optimized to obtain the most uniform size density function. The optimized structuring element is an estimate of the primitive under the assumption. Simulated annealing algorithm is employed for the optimization.

Web of Science

researchmap

researchmap

Language：English   Publisher：IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

A novel method for the primitive description of the multiprimitive texture is proposed. This method segments a texture by the watershed algorithm into fragments each of which contains one grain. The similar fragments are grouped by the cluster analysis in the feature space whose basis is the morphological size density. Each primitive is extracted as the grain of the central fragment in each cluster.

Web of Science

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Books

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

It is proposed a novel method that optimizes nonlinear filters by unsupervised learning using a novel definition of morphological pattern spectrum, called "morphological opening/closing spectrum (MOCS)." The MOCS can separate smaller portions of image objects from approximate shapes even if the shapes are degraded by noisy pixels. Our optimization method analogizes the linear low-pass filtering and Fourier spectrum: filter parameters are adjusted to reduce the portions of smaller sizes in MOCS, since they are regarded as the contributions of noises like high-frequency components. This method has an advantage that it uses only target noisy images and requires no example of ideal outputs. Experimental results of applications of this method to optimization of morphological open-closing filter for binary images are presented.

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

The pattern spectrum has been proposed to represent morphological size distribution of an image. However, the conventional pattern spectrum cannot extract approximate shape information from image objects spotted by noisy pixels since this is based only on opening. In this paper, a novel definition of the pattern spectrum, morphological multiresolution pattern spectrum (MPS), involving both opening and closing is proposed. MPS is capable of distinguishing details from approximate information of the image.

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：OPTICAL SOC JAPAN  

We described recently the multiresolution pattern spectrum (MPS), a method to extract the size distribution of image objects based on visual impression. In this paper, an optical implementation method of MPS is proposed using the visual area coding technique, an encoding method to operate gray scale images optically while preserving visual impressions oi these.

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：SOC PHOTO-OPT INSTRUM ENG  

Mathematical morphology with gray scale structuring elements has attracted much attention, since combinations of the operations in this class can realize almost all noise-removing filters. However, the optimization method for the combination is still uncertain. In this paper, an optimization method for a mathematical morphological filter with gray scale structuring elements is proposed. This method is based on the concept of a neural network with morphological operations and on learning using simulated annealing. The method is also applied to gray scale bipolar morphological filters for image differentiation. (C) 1996 Society of Photo-Optical Instrumentation Engineers.

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

The threshold logic filter is an important nonlinear filter class, which is defined by a threshold logic function of binary input values. It is proved in this paper that not all positive self-dual logical functions are threshold functions if the number of input variables is 5 or more. The positive self-dual logical filter is a limited class but includes almost all filters for noise removal. Our result means that not all positive self-dual filters can be expressed by one operation of the threshold logic filters. In the sense of the filter expression by neural networks, the two-layer network cannot always optimize even this limited class. © 1995, Japan Society of Applied Physics. All rights reserved.

DOI： 10.1007/s10043-995-0327-0

Scopus

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

A simulator to aid in the design of nonlinear image processing filters is developed. This simulator assists the filter design by determining parameters manually by trial-and-error. The method is also useful for improving the results of the learning optimization methods. An example of the application of the technique to improve a resultant filter by a learning method of a mathematical morphological filter is shown. © 1995, Japan Society of Applied Physics. All rights reserved.

DOI： 10.1007/s10043-995-0163-2

Scopus

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

An efficient algorithm based on the simulated annealing for the learning optimization of morphological filters is proposed. The learning stage is divided into two consecutive parts
the initial-learning stage finds and fixes the most important parts of the structuring elements, and the precise-learning stage determines details of the rest. This method significantly reduces the number of trials for the modification of structuring elements. The proposed algorithm is applied to the learning optimization of the bipolar morphological operation, whose optimization problem has not yet been investigated. It is shown experimentally that the algorithm optimizes the operator as efficiently as the conventional one and reduces the amount of calculation. © 1995, Japan Society of Applied Physics. All rights reserved.

DOI： 10.1007/s10043-995-0159-y

Scopus

researchmap

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：ELSEVIER SCIENCE BV  

An optimization method of the mathematical morphological filter is proposed. This method is based on the concept of learning by the simulated annealing. A novel neural network architecture based on mathematical morphology is utilized.

Web of Science

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

It is reported that the first partial derivative of the phase distribution for a transparent object can be measured automatically by a personal computer with an image processor. The moire pattern is formed by superimposing the Fourier image distorted by the transparent object on the master grating. But if only the moire patterns are observed, it is not shown if the fringe order numbers increase or decrease. The fringe scanning method is presented to overcome the fringe order numbers. The four-step method is used to calculate the first partial derivative. By a simple experiment, the first partial derivative can be shown in the three dimensional graphics. © 1994, Japan Society of Applied Physics. All rights reserved.

DOI： 10.1007/BF03254883

Scopus

researchmap

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：PERGAMON-ELSEVIER SCIENCE LTD  

A class of new differential filter that is a generalized version of the weighted median filter is proposed. The weighted median filter is a generalization of the median filter that has weight coefficients. In this paper, we propose an extended version of the weighted median filter that has negative weight coefficients. We call it the rank-order based nonlinear differential operator (RONDO). We show that RONDO acts as an effective differential operator. It detects edges in images degraded by impulsive noise effectively. It also has clear selectivity of edge orientation.

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：ELSEVIER SCIENCE BV  

The effects of iterations of the RANK filter are studied experimentally. The RANK filters are a new family of nonlinear filters. The effects of the RANK filters are significant because of their local adaptivity, i.e. flexibility of the shape of the filter window. Images of random values that are uniformly independently distributed are processed by the RANK filters iteratively, and the fixed points, which are defined as the images invariant to additional filtering are observed. It is found that various RANK filters generate some structures in the fixed point images and distinct modes in their histograms. The generation of these modes is dependent upon the filter parameters.

Web of Science

researchmap

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：OPTICAL SOC AMER  

The weighted median filter (WMF) is a generalization of the median filter. The WMF is more effective for image processing than the conventional median filter. However, the design of the parameters of the WMF is a difficult problem. A novel method of optimizing the WMF is proposed that utilizes the close relation between the nonrecursive WMF and the feed-forward neural network with shift-invariant weight coefficients. The optimization problem of the WMF results in the learning of the interconnection weights of the network.

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：JAPAN SOC APPLIED PHYSICS  

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

It is demonstrated that all signals of arbitrary dimension converge to the root signals by iterative operations of the recursive rank-order filter under certain conditions. We utilize the relation between rank-order filters and Hopfield's neural network model. © 1990.

DOI： 10.1016/0167-8655(90)90026-X

Scopus

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：PERGAMON-ELSEVIER SCIENCE LTD  

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Publ by Int Soc for Optical Engineering  

Recently, mathematical morphology has been applied to image processing and analysis. From the point of view of optical computing, mathematical morphology is an efficient tool of optical image processing because of its simplicity and shift-invariance. The authors propose a morphological operation named bipolar morphology. It is a fundamental operation of image differentiation, like erosion and dilation for mathematical smoothing. Morphological edge detecting operators, can be defined by the combination of bipolar morphological operations. Bipolar morphology is related to a differential operator recently proposed, RONDO (rank-order-based nonlinear differential operator). RONDO is a generalized weighted median filter with positive and negative weight coefficients. It detects edges in images degraded by impulsive noise effectively. It also has selectivity of edge orientation.

Scopus

researchmap

Language：Japanese  

CiNii Books

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Information on feels of materials is mainly obtained by tactile sense for human beings. It is a very important sense for evaluation of materials. Human beings, however, can sense feels of materials by visual impressions, without touching the materials. It is rather usual that the feels of materials are visually estimated before actual touch on the materials. Actions of human beings are based on the evaluation, for example a sheet of towel is recognized as a high-grade one based on the estimation that the sheet looks soft. The feels are called in this article "visual feels of materials," and our methods of quantitative analysis on the visual feels of material are introduced.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

In this paper, we propose an improvement of an automatic detection method for carotid artery calcification in dental panoramic radiographs based on local features. To reduce misdetections for the candidate region of the calcification, the proposed method is focused on the estimated shape of the calcification. Machine learning is also applied to distinguish the misdetections. A new removal method for the mandibular bone and the area below mandible is also proposed. The proposed method can reduce the misdetection of the carotid artery calcification. The experimental result shows that the improvement of the detection performance is obtained.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

In this paper, we propose an unsupervised design method for optimal structuring element (SE) of morphological filters. We formulate the design of SE as an optimization problem, and estimate an optimal shape and brightness of SE directly from degraded images. We estimated an optimal shape and brightness by using a Genetic Algorithm (GA). Estimation of SE that can remove noise while preserving original signals is achieved by objective function that use Rank-Ordered Logarithmic Differences (ROLD) statistic for the impulse noise detection.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

The existence of the carotid artery calcification as an index for arteriosclerosis has attracted a great deal of attention. This paper proposes an improvement of the automatic detection method for carotid artery calcification in dental panoramic radiographs based on intensity gradients. This paper focuses on the local variation of the intensities in possible regions of the calcification in the radiograph. The local region is modified to enhance the difference between the calcification and its neighborhood. The proposed method can improve the number of the detected calcification regions. Experimental results show that a good detection performance is obtained.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Estimation of a suitable window shape and appropriate weights in weighted median filters is one of important problems. In this study, we formulate the design of weighted median filter as an optimization problem, and estimate optimal filters directly from degraded images. In our previous work, we estimated optimal window shapes and weights by using a Genetic Algorithm (GA) with a fixed size of window as a constraint in optimization. To determine an appropriate window size is difficult but essential since it depends on both a type of texture and a noise rate. Here, we optimize weighted median niters without the constraint in the size of filters. Numerical experiments show that designed filters outperform the niters obtained by our previous approach.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

The existence of carotid artery calcification as an index for arteriosclerosis attracts a great deal of attention. Thus the system for detecting the carotid artery calcification from dental panoramic radiographs has been developed to prompt the patients to check. This paper proposes an improvement for the automatic detection method for carotid artery calcification in digital dental panoramic radiographs based on intensity gradients. This paper reconsiders the parameters setting in the conventional method which is based on the intensities gradients of the carotid artery calcification in the radiograph. The proposed method can also reduce the miss-detection of the carotid artery calcification by elimination of mandible and spine regions. Experimental results show that the good detection performance is obtained.

CiNii Books

researchmap

Language：English   Publisher：情報処理学会  

researchmap

Language：English   Publisher：The Institute of Electronics, Information and Communication Engineers  

In this paper, an analysis method of the effect of several factors such as viewing distance in texture perception using morphological and statistical model is proposed. We synthesized textures by controlling local and global textural features separately using a texture model based on mathematical morphology. Forty subjects estimated visual impression of the texture stimuli with changing the viewing distance. The collected data was analyzed using the logistic regression model and Akaike Information Criterion (AIC). The experimental results indicate that the mutual interactions of local and global features of a texture are significantly affected by the viewing distance on human perception. The analysis numerically indicates that other factors such as gender, prior knowledge, and the changing order of viewing positions also have great influences on human texture perception.

CiNii Books

researchmap

Language：English   Publisher：映像情報メディア学会  

CiNii Books

researchmap

Language：English   Publisher：The Institute of Electronics, Information and Communication Engineers  

In this paper, an analysis method of the effect of several factors such as viewing distance in texture perception using morphological and statistical model is proposed. We synthesized textures by controlling local and global textural features separately using a texture model based on mathematical morphology. Forty subjects estimated visual impression of the texture stimuli with changing the viewing distance. The collected data was analyzed using the logistic regression model and Akaike Information Criterion (AIC). The experimental results indicate that the mutual interactions of local and global features of a texture are significantly affected by the viewing distance on human perception. The analysis numerically indicates that other factors such as gender, prior knowledge, and the changing order of viewing positions also have great influences on human texture perception.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

This study aims to perform continuous measurements of the width of the cortical bone on dental panoramic radiographs for the identification of osteoporotic patients. This continuous measurement provides more accurate cortical width than the conventional one-point method. The mandibular cortical width below the mental foramen is measured by determining cortical boundaries and evaluating the distance between boundaries continuously. The estimate of the width is then obtained by extracting the most significant portion of the histogram of the measurements using the clustering algorithm. It is experimentally shown that the improved sensitivity and specificity provided more stable and significant diagnostic efficacy than the conventional one-point method.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

It has been reported that the carotid artery calcification is closely related to the progress of arteriosclerosis in elderly people. We have investigated an automatic detection system of carotid artery calcifications on dental panoramic radiographs in order to incidentally find arteriosclerosis and refer the patients to medical professionals. This paper proposes a new detection method based on intensity gradients. It is observed that the intensities of the calcifications in radiographs are higher than those in the surrounding regions. The proposed method utilizes this to realize an automatic detection with smaller number of prefixed parameters than the conventional ones. The detection ability of the proposed method is experimentally shown to be superior to the conventional ones.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

The existence of carotid artery calcification as an index for arteriosclerosis attracts a great deal of attention. Thus the system for detecting the carotid artery calcification from dental panoramic radiographs has been developed to prompt the patients to check. This paper proposes a new automatic detection method for carotid artery calcification in dental panoramic radiographs based on intensity gradients. This paper focuses on the fact that the intensities of the carotid artery calcification in the radiograph are higher than that in neighborhood. The proposed method can reduce the number of parameters adjusted in advance and the automatic detection method from the entire image can be developed. Experimental results show that good detection performance is obtained.

CiNii Books

researchmap

researchmap

Language：English   Publisher：IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

Measurement of an individual molar provides rich information for forensic personal identification. We propose a computer-based system for extracting an individual molar from dental panoramic radiographs. A molar is obtained by extracting the region-of-interest, separating the maxilla and mandible, and extracting the boundaries between teeth. The proposed system is almost fully automatic; all that the user has to do is clicking three points on the boundary between the maxilla and the mandible.

DOI： 10.1587/transinf.E92.D.2287

Web of Science

researchmap

Language：English   Publisher：The Institute of Electronics, Information and Communication Engineers  

An analysis method of the effect of the viewing distance in texture perception using morphological and statistical model is proposed. Texture perception has been an area of interest for many researchers. However, the effect of the viewing distance which is an essential issue in texture discrimination is still an open problem. In this paper, we synthesized textures by controlling local and global textural features separately using a texture model based on mathematical morphology, namely the PGPC texture model. Visual sensory tests were carried out on thirty-two respondents in four experiments. The collected data was analyzed using the logistic regression model. The experimental results indicate that the viewing distance and the mutual interactions of local and global features of a texture have significant effects on human perception. Other factors such as prior knowledge and the order of the viewing positions influence human perception in texture discrimination. This study contributes to the construction of the numerical relation between the viewing distance and human texture perception.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

It is difficult to search the name of a flower by the image of a botanical database system. This paper presents a novel botanical database system queried by images using the similarity search. The employed image feature in the color distribution of corollae, since the shape of corollae, which has been used on a feature in conventional researches, is not stable and dependent on photographic directions. The color distribution is depicted as a hue histogram of the corollae obtained by an automatic extraction. The similarity between a key flower and a flower in the database is defined by the cross-correlation. The experimental results show that the system achieves the precision 72%.

CiNii Books

researchmap

Publisher：社団法人電子情報通信学会  

本研究では、テクスチャ認識における視距離の影響をモルフォロジカルなモデルと統計的モデルを用いて解析する手法を提案している。この手法では、まず、マセマティカル・モルフォロジに基づくテクスチャモデルを用い、テクスチャのグローバルな特徴とローカルな特徴を個別にコントロールしてテクスチャ画像を合成し、そのテクスチャを視距離を変えながら被験者に提示して視覚的印象の評価を行った。次に、ロジスティック回帰モデルに基づく判別関数を用いて評価データを解析し、テクスチャ認識における視距離の影響を定量的に示した。その結果、視距離とテクスチャの識別との関係において、テクスチャのローカルな特徴とグローバルな特徴の相互作用や視距離を変化させる順序、テクスチャに対する予備知識などが与える影響を定量的に解析できることが明らかになった。

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Osteoporosis is a bone disease that leads to an increased risk of fracture. Osteoporosis itself has no subjective symptoms. Automatic diagnosis system using dental panoramic radiographs of mandible has attracted a lot of attention. In this paper, we propose an automatic diagnosis method, in which the parallel and vertical trabecular structures to the tooth roots in mandible are measured and evaluated. It shows that the parallel trabecular structures of osteoporosis patients have decreased more than vertical ones. This results can be used to classify the osteoporosis patients and healthy persons, and to diagnose osteoporosis automatically.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Estimation of a suitable window shape and appropriate weights in weighted median filters is one of important problems. In this study, we formulate the design of weighted median filter as an optimization problem, and optimal window shapes and weights are estimated directly from degraded images. In the previous work, we estimated them by using a Genetic Algorithm (GA) on texture images corrupted by positive impulse noises. Here, we extends our proposed unsupervised design to the positive and negative impulse removal. Through experiments, it is shown that designed filters outperform compared to conventional impulse noise filters from the perspective of both high noise removal performance and preservative quality.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

It is difficult to search the name of a flower by the image of a botanical database system. This paper presents a novel botanical database system queried by images using the similarity search. The employed image feature in the color distribution of corollae, since the shape of corollae, which has been used on a feature in conventional researches, is not stable and dependent on photographic directions. The color distribution is depicted as a hue histogram of the corollae obtained by an automatic extraction. The similarity between a key flower and a flower in the database is defined by the cross-correlation. The experimental results show that the system achieves the precision 72%.

CiNii Books

researchmap

Publisher：社団法人電子情報通信学会  

荷重メジアンフィルタにおいて,画像に対して適切な荷重および窓形状を推定することは重要な問題である.本研究ではこれらの設計を最適化問題として定式化し,劣化画像から直接,最適な窓形状と荷重を推定する手法を提案している.前回の報告では,最適化手法に遺伝的アルゴリズム(GA)を採用し,正のインパルス性雑音が重畳した画像を対象として最適な窓形状・荷重を推定した.本研究では,目的関数およびGAを改良し,対象画像を正負のインパルス性雑音が重畳した画像に拡張する.ここでは処理対象画像としてテクスチャ画像を用い,数値実験を通して従来のインパルス性雑音の除去フィルタと比較して,雑音除去性能と原信号保存性に優れたフィルタ設計が可能であることを示す.

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Osteoporosis is a bone disease that leads to an increased risk of fracture. Osteoporosis itself has no subjective symptoms. Automatic diagnosis system using dental panoramic radiographs of mandible has attracted a lot of attention. In this paper, we propose an automatic diagnosis method, in which the parallel and vertical trabecular structures to the tooth roots in mandible are measured and evaluated. It shows that the parallel trabecular structures of osteoporosis patients have decreased more than vertical ones. This results can be used to classify the osteoporosis patients and healthy persons, and to diagnose osteoporosis automatically.

CiNii Books

researchmap

Language：Japanese   Publisher：社団法人電子情報通信学会  

CiNii Books

researchmap

Publisher：社団法人電子情報通信学会  

本稿では,歯科パノラマX線写真に写る頸動脈石灰化領域を自動抽出する手法について提案する.頸動脈上の石灰化の有無は,動脈硬化の危険性を表す指標として注目されている.歯科パノラマX線写真はコントラストが悪く,通常のエッジ抽出法では強い輝度を持つ骨の領域上に多くのエッジが抽出されてしまう.石灰化部位の輝度値が骨と類似しているという問題もある.今までいくつかの石灰化部位を抽出する手法が提案されている.しかし従来法はあらかじめ検索領域を指定しなくてはならず,画像全体から石灰化部位を抽出することは困難であった.そこで,顎骨と脊椎の領域の推定を行い除去することで,石灰化部位が存在する領域を限定し,その領域に領域指定型の従来法を一定間隔で適用することにより,画像全体から石灰化部位の抽出を行う.実際の歯科パノラマX線写真に適用した結果,良好な抽出結果が得られた.

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

To recover texture images from impulse noise by the opening operation which is one of morphological operations, an suitable structure element (SE) has to be estimated. In this study, we proposed an unsupervised design method for the SEs using Genetic Algorithm (GA). In previous work, it has been shown that our proposal method outperforms a conventional method of Simulated Annealing (SA), under a restricted noise rate. Here we examine effectiveness of our approach against several noise conditions.

CiNii Books

researchmap

Language：English   Publisher：IEEE  

This paper proposes improvement for the detection method for carotid artery calcification in dental panoramic radiographs. The existence of the carotid artery calcification as an index for arteriosclerosis attracts a great deal of attention. The detection method for carotid artery calcification has been proposed. However the detection rate is about 50% for real data and its improvement should be required. In this paper, the area of the processing window is narrowed to be 300 x 300 pixels. Then we focus on the shape and area of the target region and improve the detection and misdetection rates. From experimental results, the carotid artery calcification region can extract more precisely compared to the conventional method.

DOI： 10.1109/ISPACS.2009.5383888

Web of Science

researchmap

Language：English   Publisher：IEEE  

Deterministic Multi-step Crossover Fusion (dMSXF) is a promising crossover method in combinatorial problems, which performs multi-step searches focusing on inheritance of parents' characteristics using a neighborhood structure and a distance measure. DMSXF shows good search ability in various kinds of combinatorial problem. In this paper, we apply dMSXF to unsupervised design of morphological filters based on the opening operation for noise removal. We estimate suitable structuring elements (SE) for texture images corrupted by impulse noise. Here we fix the shape of SE and optimize pixel values of each element of SE. Through the experiments, it is shown that dMSXF works very well on estimating optimal SEs just like typical combinatorial problems.

DOI： 10.1109/ISPACS.2009.5383775

Web of Science

researchmap

Language：English   Publisher：IEEE  

Design of both a suitable window shape and appropriate weights in weighted median filters is one of important problems. Hitherto, unsupervised design methods of the filters by using Simulated Annealing (SA) or Genetic Algorithm (GA) have been proposed for texture images corrupted by impulse noise. These techniques estimate the optimal window shape and the optimal filter weights separately, and they have been shown to perform as well as a supervised method. In this paper, we propose a new approach which optimize both the window shape and the weight at the same time to design more sophisticated filters. We apply GA and adopt Rank-Ordered Logarithmic Difference (ROLD) statistics as objective function to design a weighted median filter. Through experiments, it was shown that our new approach outperformed compared to conventional methods.

DOI： 10.1109/ISPACS.2009.5383865

Web of Science

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Set granulation is an approximation method of a set by grouping elements to an eqivalence class based on approximate identity of the attributes, and it is the main issue of rough set theory. This research introduces the adjacency class instead of the equivalence class, and performs the granulation based on the morphological opening. It defines the granulation and the measurement of granurality based on similarity, which is an extension of the identity.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

To recover texture images from impulse noise by the opening operation which is one of morphological operations, an suitable structure element (SE) has to be estimated. Hitherto, an unsupervised design method for the SEs has been proposed, and it have adopted Simulated Annealing (SA) as an optimization method. In this approach, SEs are designed under a neighborhood structure which keeps the size of shape, and the size is gradually reduced in the search process. Due to this, the search space is restricted. In this paper, Genetic Algorithm (GA) which can search effectively on wide search spaces is applied. Through experiments, it is shown that our new approach outperformes conventional methods.

CiNii Books

researchmap

Language：Japanese   Publisher：Information Processing Society of Japan (IPSJ)  

It has been recently shown that the cell nucleus is a highly compartmentalized organelle with chromosome and chromatin domains formed by DNA. Although nuclear proteins also have been speculated to form higher order structure in the compartment between chromatin domains, how the formation of interchromatin compartment is organized is still unclear. We have previously shown that RAD51, a recombinational repair protein, forms nuclear foci, typically during S phase, and accumulates at sites of DNA double-strand breaks. We found that over-expressed RAD51 forms bundle-like structures in the nucleus. Since the structures are formed in the interchromatin compartment, we decided to study the influence of chromatin structure to the organization of interchromatin compartment. For this purpose, we examined the change of the RAD51 bundle-like structure shape after de-condensation of chromatin by the treatment with trichostatin A (TSA), a histone deacetylase inhibitor. The complexity of the GFP-RAD51 bundle-like structures was analyzed quantitatively using digital image processing methods.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

It has been recently shown that the cell nucleus is a highly compartmentalized organelle with chromosome and chromatin domains formed by DNA. Although nuclear proteins also have been speculated to form higher order structure in the compartment between chromatin domains, how the formation of interchromatin compartment is organized is still unclear. We have previously shown that RAD51, a recombinational repair protein, forms nuclear foci, typically during S phase, and accumulates at sites of DNA double-strand breaks. We found that over-expressed RAD51 forms bundle-like structures in the nucleus. Since the structures are formed in the interchromatin compartment, we decided to study the influence of chromatin structure to the organization of interchromatin compartment. For this purpose, we examined the change of the RAD51 bundle-like structure shape after de-condensation of chromatin by the treatment with trichostatin A (TSA), a histone deacetylase inhibitor. The complexity of the GFP-RAD51 bundle-like structures was analyzed quantitatively using digital image processing methods.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

It has been recently shown that the cell nucleus is a highly compartmentalized organelle with chromosome and chromatin domains formed by DNA. Although nuclear proteins also have been speculated to form higher order structure in the compartment between chromatin domains, how the formation of interchromatin compartment is organized is still unclear. We have previously shown that RAD51, a recombinational repair protein, forms nuclear foci, typically during S phase, and accumulates at sites of DNA double-strand breaks. We found that over-expressed RAD51 forms bundle-like structures in the nucleus. Since the structures are formed in the interchromatin compartment, we decided to study the influence of chromatin structure to the organization of interchromatin compartment. For this purpose, we examined the change of the RAD51 bundle-like structure shape after de-condensation of chromatin by the treatment with trichostatin A (TSA), a histone deacetylase inhibitor. The complexity of the GFP-RAD51 bundle-like structures was analyzed quantitatively using digital image processing methods.

CiNii Books

researchmap

Language：English   Publisher：BRITISH INST RADIOLOGY  

Objectives: An eroded inferior cortex of the mandible detected on dental panoramic radiographs is useful for identifying post-menopausal won-ten with low skeletal bone mineral density (BMD) or osteoporosis. The purposes of this study were to develop a computer-aided diagnosis (CAD) system that automatically determines cortical erosion of the mandible on denial panoramic radiographs and to assess the validation of this CAD system.
Methods: A CAD system was developed based on mathematical morphology for identifying post-menopausal women with low skeletal BMD or osteoporosis, based on World Health Organization criteria, by identifying whether the endosteal margin of mandibular cortical bone was eroded. The sensitivity, specificity, predictive value, accuracy and likelihood for a positive risk result were calculated using dichotomous 2 x 2 tables using 100 panoramic radiographs.
Results: For identifying women with low skeletal BMD, the sensitivity and specificity were 76.8% and 61.1%, respectively, the positive and negative predictive values were 90.0% and 36.7%, accuracy was 74.0% and the likelihood for positive risk was 1.96. The respective values for identifying women with osteoporosis were 94.4% and 43.8%, the positive and negative predictive values were 48.6% and 93.3%, respectively, the accuracy was 62.0% and the likelihood for positive risk was 1.68.
Conclusion: Our results suggest that a CAD system applied to dental panoramic radiographs may be useful for identifying post-menopausal women with low skeletal BNID or osteoporosis.

DOI： 10.1259/dmfr/68621207

Web of Science

researchmap

Language：English   Publisher：ELSEVIER SCIENCE INC  

Mandibular cortical erosion detected on dental panoramic radiographs (DPRs) may be useful for identifying women with osteoporosis, but little is known about the variation in diagnostic efficacy of observers worldwide. The purpose of this study was to measure the accuracy in identifying women at risk for osteoporosis in a worldwide group of observers using DPRs. We constructed a website that included background information about osteoporosis screening and instructions regarding the interpretation of mandibular cortical erosion. DPRs of 100 Japanese postmenopausal women aged 50 years or older who had completed skeletal bone mineral measurements by dual energy X-ray absorptiometry were digitized at 300 dpi. These were displayed on the website and used for the evaluation of diagnostic efficacy. Sixty observers aged 25 to 66 years recruited from 16 countries participated in this study. These observers classified cortical erosion into one of three groups (none, mild to moderate, and severe) on the website via the Internet, twice with an approximately 2-week interval. The diagnostic efficacy of the Osteoporosis Self-Assessment Tool (CST), a simple clinical decision rule based on age and weight, was also calculated and compared with that of cortical erosion. The overall mean sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the 60 observers in identifying women with osteoporosis by cortical erosion on DPRs were 82.5, 46.2, 46.7, and 84.0%, respectively. Those same values by the OST index were 82.9, 43.1, 43.9, and 82.4%, respectively. The intra-observer agreement in classifying cortical erosion on DPRs was sufficient (weighted kappa values > 0.6) in 36 (60%) observers. This was significantly increased in observers who specialized in oral radiology (P < 0.05). In the 36 observers with sufficient intra-observer agreement, the overall mean sensitivity, specificity, PPV, and NPV in identifying women with osteoporosis by any cortical erosion were 83.5, 48.7, 48.3, and 85.7%, respectively. The mean PPV and NPV were significantly higher in the 36 observers with sufficient intra-observer agreement than in the 24 observers with insufficient intra-observer agreement. Our results reconfirm the efficacy of cortical erosion findings in identifying postmenopausal women at risk for osteoporosis, among observers with sufficient intra-observer agreement Information gathered from radiographic examination is at least as useful as that gathered from the OST index. (C) 2008 Elsevier Inc. All rights reserved.

DOI： 10.1016/j.bone.2008.03.014

Web of Science

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Estimation of a suitable window shape and appropriate weights in weighted median filters is one of important problems. Hitherto, an unsupervised design method for the filters has been proposed. In this method, a simulated annealing is applied to design the optimal filter for texture images corrupted by impulse noise, and it ahs been shown to perform as equally well as a supervised technique. In this paper, we apply a genetic algorithm and adopt Rank-Ordered Logarithmic Difference (ROLD) statistics as objective function to design a weighted median filter. Through experiments, it was shown that our new approach outperformed compared to conventional methods.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Estimation of a suitable window shape and appropriate weights in weighted median filters is one of important problems. Hitherto, an unsupervised design method for the filters has been proposed. In this method, a simulated annealing is applied to design the optimal filter for texture images corrupted by impulse noise, and it ahs been shown to perform as equally well as a supervised technique. In this paper, we apply a genetic algorithm and adopt Rank-Ordered Logarithmic Difference (ROLD) statistics as objective function to design a weighted median filter. Through experiments, it was shown that our new approach outperformed compared to conventional methods.

CiNii Books

researchmap

Language：English  

An evaluation method of human visual impressions in gray scale textures using morphological morphology is proposed. Variations of textures are generated by modifying repetitively arranged objects and configurations of the arrangements of original textures. The variations are presented to human respondents, and similarity of modified textures based on human impressions is evaluated. The results of the human evaluation are compared with the results of similarity evaluation based on image features. It shows that global features such as density, regularity and directionality of the point configurations have significant effects on human visual impressions and identification of textures. In case of a texture without significant characteristics on its point configuration, local features such as the shape of the objects have some effects on the visual impressions. © 2008 IEEE.

DOI： 10.1109/HIS.2008.80

Scopus

researchmap

researchmap

researchmap

researchmap

Publisher：社団法人電子情報通信学会  

本稿では,歯科パノラマX線写真における頸動脈石灰化領域を検出する一手法について提案する.この石灰化領域の有無は,動脈硬化の危険性を示す指標として注目を集めている.石灰化部位は周囲より輝度値が高いため,領域内の平均輝度値より輝度値が高い領域を抽出することを考える.しかし,歯科パノラマX線写真は全体的にコントラストが悪く,Sobe1フィルタやラプラシアンフィルタなどのエッジ検出フィルタでエッジ特徴を用いた抽出を行っても,正確に頸動脈石灰化部位を検出することは難しい.そこで,ファジィ濃度強調法と代数的画像オペレータを併用した検出手法について検討した.その結果,いくつかの歯科パノラマX線写真から石灰化部位を抽出することに成功したので報告する.

researchmap

Publisher：社団法人電子情報通信学会  

ラフ集合論は,集合の近似を行なう考え方のひとつで,人間の認知のモデル化の基盤に適していることから,感性工学など各種の応用が行なわれている.本稿では,ラフ集合論に,本来画像処理の手法であるマセマティカル・モルフォロジにおけるopeningに相当する演算を導入することで,ラフ集合論において同値関係のみならず近傍関係を取り扱う手法について検討する.

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Rough set theory is a framework of approximating sets. It has recently been applied to kansei engineering and other application researches, since it is suitable to a basement for modelization of human cognition. In this paper, a method of handling not only the equivalence relationship but also the neighborhood relationship is considered by introducing an operation equivalent to opening in mathematical morphology, which is originally an image processing framework, into the rough set theory.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

This paper proposes a detection method for carotid artery calcification in dental panoramic radiographs. The existence of the carotid artery calcification as an index for arteriosclerosis attracts a great deal of attention. The carotid artery calcification has brighter intensity compared to the neighborhood region and its fact is utilized in the proposed method. To compensate the bad contrast of the dental panoramic radiographs, we also employ the fuzzy image contrast enhancement and algebraic image operator to the proposed method. From experimental results, the carotid artery calcification region can extract from several dental panoramic radiographs.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Rough set theory is a framework of approximating sets. It has recently been applied to kansei engineering and other application researches, since it is suitable to a basement for modelization of human cognition. In this paper, a method of handling not only the equivalence relationship but also the neighborhood relationship is considered by introducing an operation equivalent to opening in mathematical morphology, which is originally an image processing framework, into the rough set theory.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

This paper proposes a detection method for carotid artery calcification in dental panoramic radiographs. The existence of the carotid artery calcification as an index for arteriosclerosis attracts a great deal of attention. The carotid artery calcification has brighter intensity compared to the neighborhood region and its fact is utilized in the proposed method. To compensate the bad contrast of the dental panoramic radiographs, we also employ the fuzzy image contrast enhancement and algebraic image operator to the proposed method. From experimental results, the carotid artery calcification region can extract from several dental panoramic radiographs.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Estimation of a suitable window shape and appropriate weights in weighted median filters is one of important problems. In this paper, we propose a new unsupervised design method for window shapes and weights of weighted median filters. This technique is applied to texture images corrupted by impulse noise. As the performance measure for optimization by the simulated annealing method, the rank-ordered absolute differences statistic is used for the estimation of the window shapes and weights, respectively. The simulation results show that the proposed method can design the weighted median filters with almost same performance compared to the one by the supervised technique.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Estimation of a suitable window shape and appropriate weights in weighted median filters is one of important problems. In this paper, we propose a new unsupervised design method for window shapes and weights of weighted median filters. This technique is applied to texture images corrupted by impulse noise. As the performance measure for optimization by the simulated annealing method, the rank-ordered absolute differences statistic is used for the estimation of the window shapes and weights, respectively. The simulation results show that the proposed method can design the weighted median filters with almost same performance compared to the one by the supervised technique.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Estimation of a suitable window shape and appropriate weights in weighted median filters is one of important problems. In this paper, we propose a new unsupervised design method for window shapes and weights of weighted median filters. This technique is applied to texture images corrupted by impulse noise. As the performance measure for optimization by the simulated annealing method, the rank-ordered absolute differences statistic is used for the estimation of the window shapes and weights, respectively. The simulation results show that the proposed method can design the weighted median filters with almost same performance compared to the one by the supervised technique.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

For noise reduction in texture images by the opening operation which is one of morphology operations, the optimization technique for estimating the structure element by some texture model called "PGPC model" has been proposed. In this model, we consider the texture images to be composed of an arrangement of similar small objects. The similar texture image and its corrupted one is used for estimation of the structure elements in this technique. However, in real application, the estimation of the structure element from the corrupted image is desirable. In this paper, we propose the technique to estimate the optimal structure elements from corrupted images. This technique consists of the simulated annealing with changing of several constraints. The proposed technique provides almost same structures as given by the supervised training. The proposed method is also extended to the optimal design of closing filters.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

For noise reduction in texture images by the opening operation which is one of morphology operations, the optimization technique for estimating the structure element by some texture model called "PGPC model" has been proposed. In this model, we consider the texture images to be composed of an arrangement of similar small objects. The similar texture image and its corrupted one is used for estimation of the structure elements in this technique. However, in real application, the estimation of the structure element from the corrupted image is desirable. In this paper, we propose the technique to estimate the optimal structure elements from corrupted images. This technique consists of the simulated annealing with changing of several constraints. The proposed technique provides almost same structures as given by the supervised training. The proposed method is also extended to the optimal design of closing filters.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

For noise reduction in texture images by the opening operation which is one of morphology operations, the optimization technique for estimating the structure element by some texture model called "PGPC model" has been proposed. In this model, we consider the texture images to be composed of an arrangement of similar small objects. The similar texture image and its corrupted one is used for estimation of the structure elements in this technique. However, in real application, the estimation of the structure element from the corrupted image is desirable. In this paper, we propose the technique to estimate the optimal structure elements from corrupted images. This technique consists of the simulated annealing with changing of several constraints. The proposed technique provides almost same structures as given by the supervised training. The proposed method is also extended to the optimal design of closing filters.

CiNii Books

researchmap

Language：Japanese  

CiNii Books

researchmap

researchmap

researchmap

researchmap

researchmap

Language：Japanese   Publisher：ライフサイエンス出版  

CiNii Books

researchmap

Other Link： http://search.jamas.or.jp/link/ui/2008048804

researchmap

researchmap

researchmap

Language：English   Publisher：The Institute of Electronics, Information and Communication Engineers  

In this paper we present an automated system for extracting region of interest from raw dental panoramic radiographs for forensic personal identification. The region of interest is acquired by segmenting and separating teeth in raw panoramic radiograph and then extracting a set of signatures for each tooth by applying a filter. The proposed system performs well and successfully processed all images.

CiNii Books

researchmap

Language：English   Publisher：ELSEVIER SCIENCE BV  

This paper proposes a new method of image thresholding by using cluster organization from the histogram of an image. A new similarity measure proposed is based on inter-class variance of the clusters to be merged and the intra-class variance of the new merged cluster. Experiments on practical images illustrate the effectiveness of the new method. (c) 2006 Elsevier B.V. All rights reserved.

DOI： 10.1016/j.patrec.2006.02.022

Web of Science

researchmap

Language：English   Publishing type：Research paper, summary (international conference)  

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Estimation of a suitable window shape and appropriate weights in weighted median filters is one of important problems. In this paper, we propose a new design method for window shapes and weights of weighted median filters. This technique is applied to texture images corrupted by impulse noise. To improve mean square errors (MSE), the window shapes and weights are estimated by using the simulated annealing method, respectively. From simulation examples, the designed filters by the proposed method give better results than that of comparable methods.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Estimation of a suitable window shape and appropriate weights in weighted median filters is one of important problems. In this paper, we propose a new design method for window shapes and weights of weighted median filters. This technique is applied to texture images corrupted by impulse noise. To improve mean square errors (MSE), the window shapes and weights are estimated by using the simulated annealing method, respectively. From simulation examples, the designed filters by the proposed method give better results than that of comparable methods.

CiNii Books

researchmap

Language：English   Publisher：SPRINGER LONDON LTD  

Introduction: Mandibular inferior cortical width manually measured on dental panoramic radiographs may be useful for identifying postmenopausal women with low skeletal bone mineral density (BMD). Automatic measurement of cortical width may enable us to identify a large number of postmenopausal women with suspected low skeletal BMD. The purposes of this study were to develop a computer-aided system for measuring mandibular cortical width on dental panoramic radiographs and clarify the diagnostic efficacy of this system. Methods: Panoramic radiographs of 100 postmenopausal women who had had BMD assessments of the lumbar spine and the femoral neck were used in this study. Experienced oral radiologist determined the position of the mental foramen on 100 digitized dental panoramic radiographs. After determination of the mental foramen, mandibular cortical width below the mental foramen was measured automatically with a computer-aided system by identifying the area of interest, enhancing the original image, determining inner and outer margins of the cortex, and selecting an appropriate point. Cortical width measured by this system was compared with BMD of the lumbar spine and the femoral neck. Results: There were statistically significant correlation between cortical width measured by the computer-aided system and spinal BMD (r=0.50) and femoral neck BMD (r=0.54). These correlations were similar with those between cortical width by manual measurement and skeletal BMD. Sensitivity and specificity for identifying postmenopausal women with low spinal BMD by the computer-aided system were about 88.0% and about 58.7%, respectively. Those for identifying postmenopausal women with low femoral neck BMD by this system were about 87.5% and about 56.3%, respectively. Conclusion: Our results suggest that our computer-aided system may be useful for identifying postmenopausal women with low skeletal BMD.

DOI： 10.1007/s00198-005-0045-2

Web of Science

researchmap

Language：English   Publishing type：Research paper, summary (international conference)   Publisher：SPRINGER LONDON LTD  

Web of Science

researchmap

Language：English   Publisher：関西大学学術フロンティア・センター  

CiNii Books

researchmap

researchmap

Language：English   Publisher：関西大学学術フロンティア・センター  

CiNii Books

researchmap

Language：English   Publisher：IEEE COMPUTER SOC  

In this paper, an extraction method of trabecular structures from dental panoramic radiographs using mathematical morphological operations is proposed. It can extract trabeculae excluding roots of teeth and enables the assessment of tooth extraction for trabecular pattern. A measurement method of the lengths and directions of trabecular segments is also proposed in this paper It is suggested that the measurements of healthy and osteoporotic examples support our hypothesis that the trabeculae parallel to the roots are reduced more than those perpendicular to the roots by osteoporosis.

DOI： 10.1109/ICPR.2006.522

Web of Science

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

Language：English   Publisher：SPRINGER  

A novel method of texture characterization, called intersize correlation of grain occurrences, is proposed. This idea is based on a model of texture description, called "Primitive, Grain and Point Configuration (PGPC)" texture model. This model assumes that a texture is composed by arranging grains, which are locally extended objects appearing actually in a texture. The grains in the PGPC model are regarded to be derived from one primitive by the homothetic magnification, and the size of grain is defined as the degree of magnification. The intersize correlation is the correlation between the occurrences of grains of different sizes located closely to each other. This is introduced since homothetic grains of different sizes often appear repetitively and the appearance of smaller grains depends on that of larger grains. Estimation methods of the primitive and grain arrangement of a texture are presented. A method of estimating the intersize correlation and its application to texture regeneration are presented with experimental results. The regenerated texture has the same intersize correlation as the original while the global arrangement of large-size grains are completely different. Although the appearance of the resultant texture is globally different from the original, the semi-local appearance in the neighborhood of each large-size grain is preserved.

Web of Science

researchmap

Language：English   Publisher：IEEE COMPUTER SOC  

A hybridized classification system of the logistic discriminant analysis and the three-layer neural network is proposed. This system is basically a linear discrimination and is assisted by the neural network only for the cases that are difficult to be classified by linear methods. This system presents a simple discrimination structure given by linear methods, and its computational cost is much lower than the exclusive use of the neural network while the misclassification rate is as low as the neural network. The ability of this system is shown experimentally in the case of applying it to image identification problems. The computation time for the learning process is reduced to one-fifth by this method in this experiment, while the misclassification rate remains almost the same.

Web of Science

researchmap

Language：Japanese   Publisher：関西大学学術フロンティア・センター  

CiNii Books

researchmap

Language：English   Publisher：IEEE  

In this paper, we propose a new design method for window shapes and weights of weighted median filters. This technique is applied to texture images corrupted by impulse noise. To improve mean square errors (MSE), the window shapes and weights are estimated by using the simulated annealing method, respectively. From simulation examples, the proposed method gives better results compared to the ordinary weighted median filters.

Web of Science

researchmap

Language：English   Publisher：IEEE COMPUTER SOC  

This paper proposes a novel dewarping technique for document images of bound volumes. This technique is a kind of model fitting techniques for estimating the warp of each text line by fitting some elastic curve model to the text line. Differing from conventional techniques, the proposed technique is applicable to document images including local irregularities such as formulae, short text lines, and figures, since the proposed technique dewarps whole document images by fitting splines while considering the global optimality that specifies the desirable relationship among the splines. The experimental results on several document images including the local irregularities indicated the effectiveness of the proposed technique. The experimental result also indicated the effectiveness of the vertical division of a document image into some partial document images for more accurate dewarping.

DOI： 10.1109/ICDAR.2005.87

Web of Science

researchmap

researchmap

researchmap

researchmap

Language：English   Publisher：SPIE-INT SOC OPTICAL ENGINEERING  

Osteoporotic fractures are associated with substantial morbidity, increased medical cost and high mortality risk. Several equipments of bone assessment have been developed to identify individuals, especially postmenopausal women, with hi-h risk of osteoporotic fracture; however, a large segment of women with low skeletal bone mineral density (BMD), namely women with high risk of osteoporotic fractures, cannot be identified sufficiently because osteoporosis is asymptomatic.
Recent studies have been demonstrating that mandibular inferior cortical width manually measured on panoramic radiographs may be useful for the identification of women with low BNID. Automatic measurement of cortical width may enable us to identify a large number of asymptomatic women with low BNID. The purpose of this study was to develop a computer-aided system for measuring the mandibular cortical width on panoramic radiographs.
Initially, oral radiologists determined the region of interest based on the position of mental foramen. Some enhancing image techniques were applied so as to measure the cortical width at the best point. Panoramic radiographs of 100 women who had BNID assessments of the lumbar spine and femoral neck were used to confirm the efficacy of our new system. Cortical width measured with our system was compared with skeletal BNID. There were significant correlation between cortical width measured with our system and skeletal BNID. These correlations were similar with those between cortical width manually measured by the dentist and skeletal BNID. Our results suggest that our new system may be useful for mass screening of osteoporosis.

DOI： 10.1117/12.594458

Web of Science

researchmap

Language：English   Publisher：The Institute of Electronics, Information and Communication Engineers  

Image thresholding is usually used to distinguish the foreground from the background. The image thresholding problem is closely related to the clustering problem, which segments the image into several classes. This paper proposes a novel method of image thresholding using the optimal histogram segmentation by the cluster organization based on the similarity between adjacent clusters. The similarity measure proposed in this paper is based on the mean distance of the clusters to be marged and the compactness of the new merged cluster. Since this method is not based on the minimization of a function, the problem of selecting the threshold at the local minima is avoided. This approach overcomes the local minima that affect most of the conventional methods by maximizing the interclass variance and minimizing the intraclass variance. It is straightforward to extend the method to multi-level thresholding problem by stopping the grouping as the expected segment number is achieved. Results obtained from automatic thresholding of the experimental images are showing the validity of the method.

CiNii Books

researchmap

researchmap

researchmap

researchmap

Language：English   Publisher：SPRINGER-VERLAG BERLIN  

Quantitative evaluation of human sensibility, or "Kansei," has recently attracted much attention. Evaluation of human visual impression is one of the important topics of this area. It is proposed in this paper that, an objective evaluation method of human impressions based on the logistic discriminant analysis. This method applies the logistic discriminant analysis using various image features for sample images that have been already discriminated by sensory tests. An image feature yielding the logistic discrimination model of relatively high fitness to the human discrimination is regarded as that expressing human sensibility more appropriately than other features yielding low fitness. This method employs no. subjective words to express human impressions but enables objective,,comparison of image features with respect to the ability of expressing human sensibility. This method was applied experimentally for exploring image features to describe human visual impressions of black fabrics. Furthermore, the comparison with the result of the discriminant analysis employing some contracted scores of the features by Principal Components Analysis (PCA) shows superiority of our proposed method.

Web of Science

researchmap

researchmap

Language：English   Publisher：IEEE  

Application of opening, which is a basic operation of mathematical morphology, often yields undesired results like the occurrence of artifacts. This problem can be avoided by using a gray scale structuring element resembling the objects in the target image. Since a texture image is composed of an arrangement of similar small objects, it is possible to derive such a gray scale structuring element by estimating one representative gray scale object from those arranged in the texture image. In this paper, an estimation method of the optimal gray scale structuring element using "Primitive, Grain, and Point Configuration (PGPC)" texture model is proposed.

Web of Science

researchmap

researchmap

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Quantitative evaluation of human sensibility, or "Kansei" has recently attracted much attention. Evaluation of human visual impression is one of the important topics of this area. Most of the conventional studies have applied multivariate statistical analysis for extracting relationships between results of sensory tests and image features. These studies, however, employ subjective words of observers to express human impressions. It is proposed in this paper an objective evaluation method of human impressions based on the logistic discriminant analysis. This method applies the logistic discriminant analysis using various image features for sample images that are already discriminated by sensory tests. An image feature yielding the logistic discrimination model of relatively high fitness to the human discrimination is regarded as that expressing human sensibility more appropriately than other features yielding low fitness. This method employs no subjective words to express human impressions but enables objective comparison of image features with respect to the ability of expressing human sensibility. This method is applied experimentally for exploring image features to describe human visual impressions of black fabrics, which are important resources of formal dress.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

We have proposed "Primitive and Point Configuration Texture Model," which modelizes a texture by a primitive and its location configuration. This model enables treating a texture as a set of grains derived from a primitive by homothetic modification, and describing a texture by three characteristics -primitive shape, grain size distribution, and grain location configuration. We have proposed several estimation method of the primitive shape under some assumptions on the size distribution. We propose in this paper a primitive shape estimation method even when both the grain size distribution and the grain location configuration are unknown.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Fourier descriptor, which extracts shape characteristics of plain curves, has an ability of approximating an original curve by reconstruction using low-frequency components. Although the Fourier descriptor is originally defined forcontinuous curves, several methods of its discretization have been proposed since practical image processing treats quantized curves. The influence of the quantization is large for small objects. In this paper, the influence of the discretizations investigated by comparing the quantization methods experimentally, in the case of describing elementary small objects in a texture. The result sggests thatggests that the P- and Z-descriptors, which consider the curvature, are superior to the G-descriptor, which regards a discretized curve as a point sequence.

CiNii Books

researchmap

Publisher：社団法人電子情報通信学会  

われわれは,テクスチャの"Primitive and Point Configuration (PPC)テクスチャモデル"による記述を研究している. PPCテクスチャモデルでは,ある要素形状から派生した粒子が多数配置されることでテクスチャが構成されていると考え,要素形状と粒子の空間配置によってテクスチャを記述する.本研究では,PPCテクスチャモデルについて,対象のテクスチャにおける粒子の空間分布を操作するいくつかの手法を提案する.これらの手法では,粒子の空間配置をランダム点過程で表現する.これらの手法により,テクスチャのローカルな特徴である要素形状は保持したまま,それとは独立にグローバルな特徴を操作することができる.

researchmap

Language：English   Publisher：The Institute of Electronics, Information and Communication Engineers  

Effective features for discrimination of low-contrast images are proposed for identification of black cloths in kansei engineering approach. These features are obtained by segmentation of an image based on clustering of its histogram. They are applied to an image identification problem using the logistic discrimination, and their effectiveness is shown using an information criterion in our experiments.

CiNii Books

researchmap

Language：English   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

For diagnosys of osteoporosis from mandibles, it is necessary to extract and evaluate trabeculae from a lot of panoramic X-ray images. A method of trabeculae extraction using high-quality X-ray images taken by CR (Computed Radiography) has been proposed. However, it is easier to collect panoramic X-ray images than CR images, since panoramic X-ray imaging equipments are much more popular. The quality of panoramic X-ray images is lower than that of CR, and trabeculae and roots of teeth are overlapped in X-ray images of mandibles. We succeeded extracting trabeculae from the overlapped images, and evaluating the structure by converting trabecular images into line segments.

CiNii Books

researchmap

Publisher：システム制御情報学会  

DOI： 10.11509/isciesci.47.1_18

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

There are about 10 million or more osteoporosis patients in Japan now, early detection and early treatment are considered to be important. Recent studies suggested that the appearance (normal or eroded) of the mandibular inferior cortex on dental panoramic radiographs may be useful to detect osteoporosis. The purposes of our study were to develop CAD which offers diagnostic support for osteoporosis by discriminating such form changes of the mandible on panoramic radiographs, and evaluating the performance. As a result, sensitivity was 82.2%, specificity was 60.0%, and accuracy was 70.0%. The diagnostic performance of CAD was superior to that of dental practitioners (sensitivity : 78.2%, specificity : 39.9%, accuracy : 58.0%).

CiNii Books

researchmap

Language：English   Publisher：SPRINGER-VERLAG BERLIN  

This paper proposes a method of texture analysis using morphological size distribution. Our framework is based on the concept that a texture is described by estimation of primitive, size distribution of grains derived from the primitive, and spatial distribution of the grains. We concentrate on estimation of primitive using an assumption on grain size distribution. We assume a model that grains are derived from one primitive, and a uniform size distribution since we consider target textures containing grains of various sizes. Thus the structuring element used for the measurement of size distribution is optimized to obtain the most uniform size density function. The optimized structuring element is an estimate of the primitive under the assumption. Simulated annealing algorithm is employed for the optimization.

Web of Science

researchmap

DOI： 10.11509/isciesci.47.1_18

researchmap

Language：English   Publisher：I O S PRESS  

Quantitative evaluation of human sensibility, or "Kansei," has recently attracted much attention. Evaluation of human visual impression is one of the important topics of this area. Most of the conventional studies have applied multivariate statistical analysis for extracting relationships between results of sensory tests and image features. These studies, however, employ subjective words of observers to express human impressions. It is proposed in this paper an objective evaluation method of human impressions based on the logistic discriminant analysis. This method applies the logistic discriminant analysis using various image features for sample images that are already discriminated by sensory tests. An image feature yielding the logistic discrimination model of relatively high fitness to the human discrimination is regarded as that expressing human sensibility more appropriately than other features yielding low fitness. This method employs no subjective words to express human impressions but enables objective comparison of image features with respect to the ability of expressing human sensibility. This method is applied experimentally for exploring image features to describe human visual impressions of black fabrics, which are important resources of formal dress.

Web of Science

researchmap

researchmap

Language：English   Publisher：SPRINGER-VERLAG BERLIN  

A model for texture description, called "Primitive and Point Configuration (PPC) texture model," and an estimation method of the primitive, which is an elementary object for configuring a texture, are proposed in this paper. The PPC texture model regards that a texture is composed by arranging grains that are derived from one or a few primitives by some modification. The primitive shape is estimated by the principle that the primitive resembling the grains best should be the optimal estimation. This estimation is achieved by finding the structuring element that minimizes the integral of the size distribution function of a target texture.

DOI： 10.1007/3-540-45103-X_25

Web of Science

researchmap

Language：English   Publisher：SPRINGER-VERLAG BERLIN  

We have investigated "Primitive and Point Configuration (PPC) texture model," which regards a texture to be composed by arranging grains that are derived from one or a few primitives by some modification. The PPC texture model is able to characterize primitive shapes and grain location configuration independently. We propose in this paper methods of manipulating for the grain location configuration. These methods enable to manipulate global characteristics of a texture independently on local ones. Our methods modify the grain configuration using random point process and generate different textures while preserving the original primitive shape.

Web of Science

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：English   Publishing type：Rapid communication, short report, research note, etc. (scientific journal)   Publisher：IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

A novel method for the primitive description of the multiprimitive texture is proposed. This method segments a texture by the watershed algorithm into fragments each of which contains one grain. The similar fragments are grouped by the cluster analysis in the feature space whose basis is the morphological size density. Each primitive is extracted as the grain of the central fragment in each cluster.

Web of Science

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Object identification is a statistical method to discriminate whether two images are obtained from an identical category of objects or from two distinct categories, based on feature distances between the images. We apply logistic discrimination analysis and estimate the confidence degree of discrimination. We combine it with the feed-forward neural network to improve the discrimination ability near the discrimination boundary, where the confidence degree of discrimination is low. It is shown that our method exceeds the simple logistic discrimination in the discrimination ability, and exceeds the simple neural network in the computational complexity of learning.

CiNii Books

researchmap

Publisher：電子情報通信学会  

researchmap

researchmap

researchmap

researchmap

Language：English   Publisher：C S I R O  

A novel method of the multiprimitive texture analysis is proposed. This method segments a texture using the watershed algorithm into fragments each of which contains one grain. The size density function of each fragment is calculated, and the fragments are located in the feature space each of whose basis is the size density of a size. The shape of each grain is distorted by the segmentation if the grains overlap, and the watershed algorithm may cause the over-segmentation. Thus the fragments containing grains corresponding to one primitive scatter in the feature space. However, the following cluster analysis collects neighborhood fragments in the feature space into a cluster. The grains in a cluster are regarded as corresponding to one primitive. The number of distinctive primitives shapes is obtained as the number of distinctive clusters, and each primitive is obtained as the central fragment of each cluster.

Web of Science

researchmap

Language：English   Publisher：C S I R O  

A novel method of texture primitive description using morphological skeleton is proposed. The skeleton of an object has the property that it is reduced to one point when the structuring element used for the skeletonization is exactly hornothetic to the object. This method applies this property. This method minimizes the number of pixels contained in the skeleton. If we assume that-the texture is composed of one primitive, the structuring element minimizing the number of pixels is homothetic to the primitive of the texture because of the above property of the skeleton. Simulated annealing is employed for the minimization. This method has an advantage that it requires no assumption on the sizing distribution of grains in the texture.

Web of Science

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Image Information and Television Engineers  

A novel texture analysis method is proposed. This method is applicable to the description of texture that is a mixture of two or more distinctive sets of elementary shapes. This method classifies elementary shapes into hierarchical clusters based on the pattern spectrum, and detects typical elementary shapes of each cluster.

DOI： 10.11485/iteac.2001.0_222

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

This paper proposes a novel method of texture images analysis based on the morphology operation. The pattern spectrum which is calculated by morphological method, describes size distributions of objects contained in an image. Our method can be applied texture images which are constructed by only one element. The size and shape of the element are described by the structuring element of morphology operation.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

The efficiency of the pattern spectrum analysis method for texture classification is investigated in this paper. The pattern spectrum is a mathematical morphological method to describe the size distribution of objects in an image. We have proposed a method to extract the shape of an elementary particle of textures by optimizing the structuring element to reduce the variance of pattern spectrum that is calculated using this structuring element. In this paper we showed the texture discrimination efficiency of our method by comparing the distances between structuring elements extracted from similar as well as distinctive pairs of textures. Other evaluation functions than reducing the variance of pattern spectrum are also compared.

CiNii Books

researchmap

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Books

researchmap

Language：English   Publisher：IEEE COMPUTER SOC  

This paper proposes a novel texture modelling method based on the pattern spectrum. The pattern spectrum is a mathematical morphological method to describe the size distribution of objects contained in an image. Our method is based on the idea of obtaining a model of the elementary particles that form a texture by optimizing a gray scale structuring element to fit the shape of elementary particles. The optimization method is applied in two stages: The first stage optimizes the extent of structuring element and the second optimizes the pixel values in the extent.

Web of Science

researchmap

researchmap

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Books

researchmap

Language：English  

This paper proposes a novel method of texture analysis based on mathematical morphology. The pattern spectrum is a mathematical morphological method to describe size distributions of objects contained in an image. Our method is based on the idea of optimizing the shape of structuring elements to fit the shape of elementary particles that form a texture. Since the variance of size distribution described by the pattern spectrum is small if the structuring element for calculating the pattern spectrum fits the particles, the shape of an elementary object is obtained by the structuring element optimized to reduce the variance of size distribution. Experimental results are shown when the structuring element is restricted to binary ones of 5×5-pixels. © 1999 IEEE.

DOI： 10.1109/ICIAP.1999.797596

Scopus

researchmap

Language：Japanese  

CiNii Books

researchmap

Publisher：社団法人電子情報通信学会  

本研究は, メディアン系フィルタリングではインパルス性雑音を除去しながらも画像の微妙な輝度変化が失われやすいことについて, その度合いを評価する方法を提案する.画像中で同じ値の画素が集まっている部分をクラスタとよぶことにし, さらに1つのクラスタにある同じ値の画素の数を連結度と定義し, 画素値と連結度の組に対して頻度を与える「連結度付ヒストグラム」を定義することで, 各種フィルタリングにおけるクラスタの出来具合の違いを表現する.さらに, フィルタリングによる連結度方向のヒストグラムの分布の変化を表す指標を定義し, 1つの画像例について, メディアンおよび中央値フィルタでの輝度変化の失われる度合いを数量的に示す.

researchmap

Publisher：社団法人電子情報通信学会  

本研究は, メディアン系フィルタリングではインパルス性雑音を除去しながらも画像の微妙な輝度変化が失われやすいことについて, その度合いを評価する方法を提案する.画像中で同じ値の画素が集まっている部分をクラスタとよぶことにし, さらに1つのクラスタにある同じ値の画素の数を連結度と定義し, 画素値と連結度の組に対して頻度を与える「連結度付ヒストグラム」を定義することで, 各種フィルタリングにおけるクラスタの出来具合の違いを表現する.さらに, フィルタリングによる連結度方向のヒストグラムの分布の変化を表す指標を定義し, 1つの画像例について, メディアンおよび中央値フィルタでの輝度変化の失われる度合いを数量的に示す.

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

The median filter family removes impulsive noises effectively, however, it tends to eliminate small variations of pixel values contained in the original image. This paper proposes a method to evaluate the degree of this elimination of variations. The area where pixels of the same value are gathering is called "cluster" in this paper. The term "connectivity" is defined as the number of pixels contained in a cluster. The "histogram with connectivity", which assigns the number of pixels to each pair of pixel value and connectivity, is proposed in this paper, and the differences of the degree of cluster generation are illustrated. The value to describe the variation of the distribution of the histogram along the axis of connectivity is also defined, and the degree of elimination of the variation of pixel values are evaluated for a sample image in the case of the median and the average filters.

CiNii Books

researchmap

researchmap

Language：English   Publisher：IEEE COMPUTER SOC  

A novel method of the active contour model is proposed by introducing the mathematical morphological operations on the "contour object," which is a binary image object whose outline is the contour. The contours are shrunk and expanded by simply applying the mathematical morphological operations on the contour objects. The split and fusion. of contour are also carried out by the mathematical morphological operations, and no additional procedure for split or fusion is required.

Web of Science

researchmap

Language：English   Publisher：IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

The pattern spectrum has been proposed to represent morphological size distribution of an image. However, the conventional pattern spectrum cannot extract approximate shape information from image objects spotted by noisy pixels since this is based only on opening. In this paper, a novel definition of the pattern spectrum, morphological multiresolution pattern spectrum (MPS), involving both opening and closing is proposed. MPS is capable of distinguishing details from approximate information of the image.

Web of Science

researchmap

Language：English   Publisher：OPTICAL SOC JAPAN  

We described recently the multiresolution pattern spectrum (MPS), a method to extract the size distribution of image objects based on visual impression. In this paper, an optical implementation method of MPS is proposed using the visual area coding technique, an encoding method to operate gray scale images optically while preserving visual impressions oi these.

DOI： 10.1007/BF02935946

Web of Science

researchmap

Language：English   Publisher：SOC PHOTO-OPT INSTRUM ENG  

Mathematical morphology with gray scale structuring elements has attracted much attention, since combinations of the operations in this class can realize almost all noise-removing filters. However, the optimization method for the combination is still uncertain. In this paper, an optimization method for a mathematical morphological filter with gray scale structuring elements is proposed. This method is based on the concept of a neural network with morphological operations and on learning using simulated annealing. The method is also applied to gray scale bipolar morphological filters for image differentiation. (C) 1996 Society of Photo-Optical Instrumentation Engineers.

DOI： 10.1117/1.600827

Web of Science

researchmap

researchmap

Language：English   Publisher：I E E E  

Web of Science

researchmap

researchmap

Language：English   Publisher：SPIE - INT SOC OPTICAL ENGINEERING  

Web of Science

researchmap

Language：English   Publisher：OPTICAL SOC JAPAN  

The threshold logic filter is an important nonlinear filter class, which is defined by a threshold logic function of binary input values. It is proved in this paper that not all positive self-dual logical functions are threshold functions if the number of input variables is 5 or more. The positive self-dual logical filter is a limited class but includes almost all filters for noise removal. Our result means that not all positive self-dual filters can be expressed by one operation of the threshold logic filters. In the sense of the filter expression by neural networks, the two-layer network cannot always optimize even this limited class.

DOI： 10.1007/s10043-995-0327-0

Web of Science

researchmap

Language：English   Publisher：OPTICAL SOC JAPAN  

An efficient algorithm based on the simulated annealing for the learning optimization of morphological filters is proposed. The learning stage is divided into two consecutive parts; the initial-learning stage finds and fixes the most important parts of the structuring elements, and the precise-learning stage determines details of the rest. This method significantly reduces the number of trials for the modification of structuring elements. The proposed algorithm is applied to the learning optimization of the bipolar morphological operation, whose optimization problem has not yet been investigated. It is shown experimentally that the algorithm optimizes the operator as efficiently as the conventional one and reduces the amount of calculation.

Web of Science

researchmap

Language：English   Publisher：OPTICAL SOC JAPAN  

A simulator to aid in the design of nonlinear image processing filters is developed. This simulator assists the filter design by determining parameters manually by trial-and-error. The method is also useful for improving the results of the learning optimization methods. An example of the application of the technique to improve a resultant filter by a learning method of a mathematical morphological filter is shown.

Web of Science

researchmap

researchmap

researchmap

researchmap

Language：English   Publisher：ELSEVIER SCIENCE BV  

An optimization method of the mathematical morphological filter is proposed. This method is based on the concept of learning by the simulated annealing. A novel neural network architecture based on mathematical morphology is utilized.

DOI： 10.1016/0030-4018(94)90629-7

Web of Science

researchmap

Language：English  

It is reported that the first partial derivative of the phase distribution for a transparent object can be measured automatically by a personal computer with an image processor. The moire pattern is formed by superimposing the Fourier image distorted by the transparent object on the master grating. But if only the moire patterns are observed, it is not shown if the fringe order numbers increase or decrease. The fringe scanning method is presented to overcome the fringe order numbers. The four-step method is used to calculate the first partial derivative. By a simple experiment, the first partial derivative can be shown in the three dimensional graphics. © 1994, Japan Society of Applied Physics. All rights reserved.

DOI： 10.1007/BF03254883

Scopus

researchmap

Language：English  

It is reported that the first partial derivative of the phase distribution for a transparent object can be measured automatically by a personal computer with an image processor. The moire pattern is formed by superimposing the Fourier image distorted by the transparent object on the master grating. But if only the moire patterns are observed, it is not shown if the fringe order numbers increase or decrease. The fringe scanning method is presented to overcome the fringe order numbers. The four-step method is used to calculate the first partial derivative. By a simple experiment, the first partial derivative can be shown in the three dimensional graphics. © 1994, Japan Society of Applied Physics. All rights reserved.

DOI： 10.1007/BF03254883

Scopus

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

Language：English   Publisher：PERGAMON-ELSEVIER SCIENCE LTD  

A class of new differential filter that is a generalized version of the weighted median filter is proposed. The weighted median filter is a generalization of the median filter that has weight coefficients. In this paper, we propose an extended version of the weighted median filter that has negative weight coefficients. We call it the rank-order based nonlinear differential operator (RONDO). We show that RONDO acts as an effective differential operator. It detects edges in images degraded by impulsive noise effectively. It also has clear selectivity of edge orientation.

DOI： 10.1016/0031-3203(92)90067-S

Web of Science

researchmap

Language：English   Publisher：ELSEVIER SCIENCE BV  

Threshold logic filter is a broad class of the nonlinear filters. A designing method of cascaded threshold logic filters is proposed. The designing method is based on the learning rule for the multilayer neural network. The results of the optimization of the cascaded threshold logic filters are presented. Optimization includes those of the weighted median filter and the rank-order based nonlinear differential operator (RONDO) for specific applications.

Web of Science

researchmap

Language：English   Publisher：ELSEVIER SCIENCE BV  

The effects of iterations of the RANK filter are studied experimentally. The RANK filters are a new family of nonlinear filters. The effects of the RANK filters are significant because of their local adaptivity, i.e. flexibility of the shape of the filter window. Images of random values that are uniformly independently distributed are processed by the RANK filters iteratively, and the fixed points, which are defined as the images invariant to additional filtering are observed. It is found that various RANK filters generate some structures in the fixed point images and distinct modes in their histograms. The generation of these modes is dependent upon the filter parameters.

Web of Science

researchmap

researchmap

Language：English   Publisher：OPTICAL SOC AMER  

The weighted median filter (WMF) is a generalization of the median filter. The WMF is more effective for image processing than the conventional median filter. However, the design of the parameters of the WMF is a difficult problem. A novel method of optimizing the WMF is proposed that utilizes the close relation between the nonrecursive WMF and the feed-forward neural network with shift-invariant weight coefficients. The optimization problem of the WMF results in the learning of the interconnection weights of the network.

Web of Science

researchmap

researchmap

researchmap

researchmap

Language：English   Publisher：ELSEVIER SCIENCE BV  

DOI： 10.1016/0167-8655(90)90026-X

Web of Science

researchmap

Language：English   Publisher：JAPAN SOC APPLIED PHYSICS  

Web of Science

researchmap

Language：English   Publisher：JAPAN SOC APPLIED PHYSICS  

Web of Science

researchmap

Language：English   Publisher：PERGAMON-ELSEVIER SCIENCE LTD  

DOI： 10.1016/0031-3203(90)90002-3

Web of Science

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

Event date： 2023.8

Language：English   Presentation type：Oral presentation (general)  

researchmap

Event date： 2023.8

Language：English   Presentation type：Oral presentation (general)  

researchmap

Venue：Barcelona, Spain  

researchmap

researchmap

researchmap

researchmap

researchmap

invited talk

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

Venue：Lillehammer, Norway  

researchmap

Venue：Potsdam, Germany  

researchmap

Venue：Berlin, Germany  

researchmap

Venue：Berlin, Germany  

researchmap

researchmap

Venue：Fontainebleau, France  

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

Venue：Takamatsu, Japan  

researchmap

Venue：Verbania, Italy  

researchmap

Venue：Gyeongju, Korea  

researchmap

Venue：Lake Buena Vista, FL USA  

researchmap

Venue：Singapore  

researchmap

Venue：Singapore  

researchmap

researchmap

Application no：2004-304855  Date applied：2004.10

Patent/Registration no：4956745  Date registered：2012.3  Date issued：2012.3

PCT application for US patent

researchmap

Application no：2003-001395  Date applied：2003.1

Announcement no：2004-209089  Date announced：2004.7

Patent/Registration no：3964795  Date registered：2007.6  Date issued：2007.6

supported by JST

researchmap