Pub Date : 2015-03-02DOI: 10.1109/ICAPR.2015.7050649
H. Aggarwal, A. Majumdar
This work proposes techniques for demosaicing multi-spectral images obtained from a single sensor architecture. This is a new problem. Compressed Sensing (CS) based formulations can recover images by exploiting the sparsity of the images in the wavelet domain. In this work, we improve upon existing techniques by accounting for the hierarchical (tree-structured) correlation that exists among the wavelet coefficients of piecewise smooth signals. For a single image, this turns out to be an elastic -net problem. Since our problem involves multi-spectral images, the proposed formulation leads to a joint-sparse elastic-net optimization problem which is solved via Split Bregman type algorithm. Our proposed improvement yields considerably better recovery results compared to existing techniques.
{"title":"Multi-spectral demosaicing: A joint-sparse elastic-net formulation","authors":"H. Aggarwal, A. Majumdar","doi":"10.1109/ICAPR.2015.7050649","DOIUrl":"https://doi.org/10.1109/ICAPR.2015.7050649","url":null,"abstract":"This work proposes techniques for demosaicing multi-spectral images obtained from a single sensor architecture. This is a new problem. Compressed Sensing (CS) based formulations can recover images by exploiting the sparsity of the images in the wavelet domain. In this work, we improve upon existing techniques by accounting for the hierarchical (tree-structured) correlation that exists among the wavelet coefficients of piecewise smooth signals. For a single image, this turns out to be an elastic -net problem. Since our problem involves multi-spectral images, the proposed formulation leads to a joint-sparse elastic-net optimization problem which is solved via Split Bregman type algorithm. Our proposed improvement yields considerably better recovery results compared to existing techniques.","PeriodicalId":379764,"journal":{"name":"2015 Eighth International Conference on Advances in Pattern Recognition (ICAPR)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123075071","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-03-02DOI: 10.1109/ICAPR.2015.7050707
M. Mandal, S. Maity, A. Mukhopadhyay
Normally, statistical methods are used to generate rankings for genes in terms of their ability to distinguish between normal and malignant tumors from a gene expression dataset. However, different statistical methods yield different ranks for same gene and there is no universally accepted method for ranking. Therefore rank aggregation is required to find the overall ranking of the set of genes. There are various rank aggregation methods in the existing literature to integrate the rankings produced by various statistical tests. Moreover, the problem of integration of some partial rankings, containing unequal numbers of genes, is more challenging. In this article, a multiobjective genetic algorithm based rank aggregation method is proposed to integrate some partial rankings in an unbiased way. The first objective is to minimize the total distance from the reference ranking to the input rankings. For distance calculation, the Scaled Footrule Distance is used. The second objective is to minimize the standard deviation among those distances in order to avoid bias toward a particular input ranking. The proposed method is applied on some real-life microarray gene expression datasets, and the performance of it is compared with that of several existing rank aggregation techniques with respect to accuracy and the AUC (Area under ROC curve) value. Again, for real-life datasets, accuracy is plotted for visual comparison.
{"title":"Partial rank aggregation using multiobjective genetic algorithm: Application in ranking genes","authors":"M. Mandal, S. Maity, A. Mukhopadhyay","doi":"10.1109/ICAPR.2015.7050707","DOIUrl":"https://doi.org/10.1109/ICAPR.2015.7050707","url":null,"abstract":"Normally, statistical methods are used to generate rankings for genes in terms of their ability to distinguish between normal and malignant tumors from a gene expression dataset. However, different statistical methods yield different ranks for same gene and there is no universally accepted method for ranking. Therefore rank aggregation is required to find the overall ranking of the set of genes. There are various rank aggregation methods in the existing literature to integrate the rankings produced by various statistical tests. Moreover, the problem of integration of some partial rankings, containing unequal numbers of genes, is more challenging. In this article, a multiobjective genetic algorithm based rank aggregation method is proposed to integrate some partial rankings in an unbiased way. The first objective is to minimize the total distance from the reference ranking to the input rankings. For distance calculation, the Scaled Footrule Distance is used. The second objective is to minimize the standard deviation among those distances in order to avoid bias toward a particular input ranking. The proposed method is applied on some real-life microarray gene expression datasets, and the performance of it is compared with that of several existing rank aggregation techniques with respect to accuracy and the AUC (Area under ROC curve) value. Again, for real-life datasets, accuracy is plotted for visual comparison.","PeriodicalId":379764,"journal":{"name":"2015 Eighth International Conference on Advances in Pattern Recognition (ICAPR)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124962822","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-03-02DOI: 10.1109/ICAPR.2015.7050697
Debajyoti Sinha, Utpal Garain, S. Bandyopadhyay
This paper attempts to employ learning based pattern classification technique to extract events from biological literature. Although various approaches to extract events have been explored, none is suitable for designing a practical system of event extraction. Extracting events more precisely is still an ongoing process. In this paper, new features that seem to be relevant for the given task are investigated. Two syntactic patterns namely phrase structure and dependency structure are explored to produce improved results with respect to the Cancer Genetics Data provided in the BioNLP'13 Shared Task. A stacked model based on conditional probability scores are also considered as features. The patterns and the probability scores along with some other linguistic features are fed to SVMs to train it for the task of bio-event extraction from natural language articles. The results are compared with the performance of the best extraction system in Cancer Genetics Task.
{"title":"Event extraction from cancer genetics literature","authors":"Debajyoti Sinha, Utpal Garain, S. Bandyopadhyay","doi":"10.1109/ICAPR.2015.7050697","DOIUrl":"https://doi.org/10.1109/ICAPR.2015.7050697","url":null,"abstract":"This paper attempts to employ learning based pattern classification technique to extract events from biological literature. Although various approaches to extract events have been explored, none is suitable for designing a practical system of event extraction. Extracting events more precisely is still an ongoing process. In this paper, new features that seem to be relevant for the given task are investigated. Two syntactic patterns namely phrase structure and dependency structure are explored to produce improved results with respect to the Cancer Genetics Data provided in the BioNLP'13 Shared Task. A stacked model based on conditional probability scores are also considered as features. The patterns and the probability scores along with some other linguistic features are fed to SVMs to train it for the task of bio-event extraction from natural language articles. The results are compared with the performance of the best extraction system in Cancer Genetics Task.","PeriodicalId":379764,"journal":{"name":"2015 Eighth International Conference on Advances in Pattern Recognition (ICAPR)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122013385","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-03-02DOI: 10.1109/ICAPR.2015.7050677
A. Chandran, A. Loh, P. Vadakkepat
A Non-recursive Motion Similarity Clustering (NMSC) algorithm is proposed to identify pedestrians traveling together in social groups. The clustering algorithm is unsupervised and can automatically identify social groups within a region of interest in a video. Social groups are identified using only pedestrian motion information by imposing motion parameter thresholds defined by social psychological principles. Social groups are identified without any prior training. In addition to detecting small social groups, NMSC also detects short-term groups (occurring for a few seconds) and social groups with sparsely distributed pedestrians. The real-time performance and group identification accuracy reveal that the proposed clustering algorithm performs better compared to existing algorithms even for scenes with a large number of pedestrians.
{"title":"Identifying social groups in pedestrian crowd videos","authors":"A. Chandran, A. Loh, P. Vadakkepat","doi":"10.1109/ICAPR.2015.7050677","DOIUrl":"https://doi.org/10.1109/ICAPR.2015.7050677","url":null,"abstract":"A Non-recursive Motion Similarity Clustering (NMSC) algorithm is proposed to identify pedestrians traveling together in social groups. The clustering algorithm is unsupervised and can automatically identify social groups within a region of interest in a video. Social groups are identified using only pedestrian motion information by imposing motion parameter thresholds defined by social psychological principles. Social groups are identified without any prior training. In addition to detecting small social groups, NMSC also detects short-term groups (occurring for a few seconds) and social groups with sparsely distributed pedestrians. The real-time performance and group identification accuracy reveal that the proposed clustering algorithm performs better compared to existing algorithms even for scenes with a large number of pedestrians.","PeriodicalId":379764,"journal":{"name":"2015 Eighth International Conference on Advances in Pattern Recognition (ICAPR)","volume":"392 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124710790","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-03-02DOI: 10.1109/ICAPR.2015.7050682
Ankita Shukla, A. Majumdar, R. Ward
In Wireless Body Area Networks (WBAN) the energy consumption is dominated by sensing, processing and communication. Previous Compressed Sensing (CS) based solutions to EEG tele-monitoring over WBAN's could only reduce the communication cost. In this work, we propose to reduce the sensing and processing energy costs as well, by randomly under-sampling the signal. We formulate a theoretically sound framework based on Kronecker Compressed Sensing (KCS) for recovering signals acquired via random under-sampling. We have shown experimentally that when the signals are acquired via under-sampling, all previous CS based techniques fail; only our proposed formulation succeeds. We have also carried out a discussion on the power savings provided by our method; the analysis indicate significant reduction in energy cost.
{"title":"A Kronecker Compressed Sensing formulation for energy efficient EEG sensing","authors":"Ankita Shukla, A. Majumdar, R. Ward","doi":"10.1109/ICAPR.2015.7050682","DOIUrl":"https://doi.org/10.1109/ICAPR.2015.7050682","url":null,"abstract":"In Wireless Body Area Networks (WBAN) the energy consumption is dominated by sensing, processing and communication. Previous Compressed Sensing (CS) based solutions to EEG tele-monitoring over WBAN's could only reduce the communication cost. In this work, we propose to reduce the sensing and processing energy costs as well, by randomly under-sampling the signal. We formulate a theoretically sound framework based on Kronecker Compressed Sensing (KCS) for recovering signals acquired via random under-sampling. We have shown experimentally that when the signals are acquired via under-sampling, all previous CS based techniques fail; only our proposed formulation succeeds. We have also carried out a discussion on the power savings provided by our method; the analysis indicate significant reduction in energy cost.","PeriodicalId":379764,"journal":{"name":"2015 Eighth International Conference on Advances in Pattern Recognition (ICAPR)","volume":"753 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123871571","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-03-02DOI: 10.1109/ICAPR.2015.7050676
R. Ghosh, Dipanjan Ghosh, Sreemoyee Roy, A. Mukherjee
Air quality information has assumed much importance over the years due to the increase in air pollution. One major hindrance in monitoring of air pollutants is the dearth of spatial availability of aerosol concentration measurements due to the cost involved in deployment of sensors. In this respect, self similarity analysis of data can be very useful. This work is based on standard grid based pollutant dispersion models in a simulated environment over different scales of grid size. The fractal dimension is considered as a scale invariant metric which gives an idea about the variation in pollutant concentration across different scales. A method is detailed for measuring the fractal dimension properties. Results indicate that it is possible to apply the dispersion models across different scales and also the air quality monitored in one region can be compared with other regions.
{"title":"Exploring the self similar properties for monitoring of air quality information","authors":"R. Ghosh, Dipanjan Ghosh, Sreemoyee Roy, A. Mukherjee","doi":"10.1109/ICAPR.2015.7050676","DOIUrl":"https://doi.org/10.1109/ICAPR.2015.7050676","url":null,"abstract":"Air quality information has assumed much importance over the years due to the increase in air pollution. One major hindrance in monitoring of air pollutants is the dearth of spatial availability of aerosol concentration measurements due to the cost involved in deployment of sensors. In this respect, self similarity analysis of data can be very useful. This work is based on standard grid based pollutant dispersion models in a simulated environment over different scales of grid size. The fractal dimension is considered as a scale invariant metric which gives an idea about the variation in pollutant concentration across different scales. A method is detailed for measuring the fractal dimension properties. Results indicate that it is possible to apply the dispersion models across different scales and also the air quality monitored in one region can be compared with other regions.","PeriodicalId":379764,"journal":{"name":"2015 Eighth International Conference on Advances in Pattern Recognition (ICAPR)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128139713","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-03-02DOI: 10.1109/ICAPR.2015.7050673
Soumik Mondal, Patrick A. H. Bours
Continuous Authentication by analysing the user's behaviour profile on the computer input devices is challenging due to limited information, variability of data and the sparse nature of the information. As a result, most of the previous research was done as a periodic authentication, where the analysis was made based on a fixed number of actions or fixed time period. Also, the experimental data was obtained for most of the previous research in a very controlled condition, where the task and environment were fixed. In this paper, we will focus on actual continuous authentication that reacts on every single action performed by the user. The experimental data was collected in a complete uncontrolled condition from 52 users by using our data collection software. In our analysis, we have considered both keystroke and mouse usages behaviour pattern to avoid a situation where an attacker avoids detection by restricting to one input device because the continuous authentication system only checks the other input device. The result we have obtained from this research is satisfactory enough for further investigation on this domain.
{"title":"Continuous Authentication in a real world settings","authors":"Soumik Mondal, Patrick A. H. Bours","doi":"10.1109/ICAPR.2015.7050673","DOIUrl":"https://doi.org/10.1109/ICAPR.2015.7050673","url":null,"abstract":"Continuous Authentication by analysing the user's behaviour profile on the computer input devices is challenging due to limited information, variability of data and the sparse nature of the information. As a result, most of the previous research was done as a periodic authentication, where the analysis was made based on a fixed number of actions or fixed time period. Also, the experimental data was obtained for most of the previous research in a very controlled condition, where the task and environment were fixed. In this paper, we will focus on actual continuous authentication that reacts on every single action performed by the user. The experimental data was collected in a complete uncontrolled condition from 52 users by using our data collection software. In our analysis, we have considered both keystroke and mouse usages behaviour pattern to avoid a situation where an attacker avoids detection by restricting to one input device because the continuous authentication system only checks the other input device. The result we have obtained from this research is satisfactory enough for further investigation on this domain.","PeriodicalId":379764,"journal":{"name":"2015 Eighth International Conference on Advances in Pattern Recognition (ICAPR)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115904171","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-03-02DOI: 10.1109/ICAPR.2015.7050691
S. Adhikari, J. Sing, D. K. Basu, M. Nasipuri
The standard fuzzy C-means (FCM) algorithm does not fully utilize the spatial information for image segmentation and is sensitive to noise especially in the presence of intensity inhomogeneity in magnetic resonance imaging (MRI) images. The underlying reason is that a single fuzzy membership function in FCM algorithm cannot properly represent pattern associations to all clusters. In this paper, we present a spatial fuzzy C-means (SpFCM) algorithm for the segmentation of MRI images. The algorithm utilizes spatial information from the neighbourhood of each pixel under consideration and is realized by defining a probability function. A new membership function is introduced using this spatial information to generate local membership values for each pixel. Finally, new clustering centers and weighted joint membership functions are presented based on the local and global membership functions. The resulting SpFCM algorithm solves the problem of sensitivity to noise and intensity inhomogeneity in MRI data and thereby improves the segmentation results. The experimental results on several simulated and real-patient MRI brain images show that the SpFCM algorithm has superior performance on image segmentation when compared to some FCM-based algorithms.
{"title":"A spatial fuzzy C-means algorithm with application to MRI image segmentation","authors":"S. Adhikari, J. Sing, D. K. Basu, M. Nasipuri","doi":"10.1109/ICAPR.2015.7050691","DOIUrl":"https://doi.org/10.1109/ICAPR.2015.7050691","url":null,"abstract":"The standard fuzzy C-means (FCM) algorithm does not fully utilize the spatial information for image segmentation and is sensitive to noise especially in the presence of intensity inhomogeneity in magnetic resonance imaging (MRI) images. The underlying reason is that a single fuzzy membership function in FCM algorithm cannot properly represent pattern associations to all clusters. In this paper, we present a spatial fuzzy C-means (SpFCM) algorithm for the segmentation of MRI images. The algorithm utilizes spatial information from the neighbourhood of each pixel under consideration and is realized by defining a probability function. A new membership function is introduced using this spatial information to generate local membership values for each pixel. Finally, new clustering centers and weighted joint membership functions are presented based on the local and global membership functions. The resulting SpFCM algorithm solves the problem of sensitivity to noise and intensity inhomogeneity in MRI data and thereby improves the segmentation results. The experimental results on several simulated and real-patient MRI brain images show that the SpFCM algorithm has superior performance on image segmentation when compared to some FCM-based algorithms.","PeriodicalId":379764,"journal":{"name":"2015 Eighth International Conference on Advances in Pattern Recognition (ICAPR)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134353035","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-03-02DOI: 10.1109/ICAPR.2015.7050713
Suman K. Ghosh, Raunak Saha
Over time, old films and pictures are eroded and often get corrupted with physical scratches. Digitization of these physically corrupted films lead to the presence of various scratch marks and aberrations oriented along multiple directions in the digitized image. Detection of these scratches for subsequent restoration is a rather difficult task because of the sensitiveness to noise and interference of background contour textures. To address this problem, we propose a highly robust and real time spatial scratch detection algorithm for static images. We deal with the more potent problem of detecting scratches in images regardless of orientation, color or shape by coupling binary detection with Hough Transformation and image rotation. Unavailability of temporal information makes such detection even more challenging. Experimental results suggest the effectiveness of our proposed method keeping in consideration computational time complexity constraints.
{"title":"A simple and robust algorithm for the detection of multidirectional scratch from digital images","authors":"Suman K. Ghosh, Raunak Saha","doi":"10.1109/ICAPR.2015.7050713","DOIUrl":"https://doi.org/10.1109/ICAPR.2015.7050713","url":null,"abstract":"Over time, old films and pictures are eroded and often get corrupted with physical scratches. Digitization of these physically corrupted films lead to the presence of various scratch marks and aberrations oriented along multiple directions in the digitized image. Detection of these scratches for subsequent restoration is a rather difficult task because of the sensitiveness to noise and interference of background contour textures. To address this problem, we propose a highly robust and real time spatial scratch detection algorithm for static images. We deal with the more potent problem of detecting scratches in images regardless of orientation, color or shape by coupling binary detection with Hough Transformation and image rotation. Unavailability of temporal information makes such detection even more challenging. Experimental results suggest the effectiveness of our proposed method keeping in consideration computational time complexity constraints.","PeriodicalId":379764,"journal":{"name":"2015 Eighth International Conference on Advances in Pattern Recognition (ICAPR)","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124071911","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2015-03-02DOI: 10.1109/ICAPR.2015.7050689
S. Kar, S. Maity
Automatic extraction of retinal blood vessels is an important issue for the diagnosis and the treatment of different retinal disorders. Most of the retinal images are of low contrast due to non-uniform illumination during acquisition process. Therefore, vessel extraction from unevenly illuminated retinal background is really a challenging task. To extract the vessels which lie in the optic disc region, the removal of the optic disc is also important. This paper proposes an algorithm for automatic blood vessel extraction and optic disc removal on poorly illuminated retinal images using curvelet transform, morphological operation, matched filtering and fuzzy entropy maximization. Curvelet transform is used to extract the finest details along the vessels since it can represent the lines, the edges, the curvatures, the missing and the imprecise boundary details efficiently. To remove the optic disc, the curvelet based edge enhanced image is first opened by a disk shaped structuring element which is then subtracted from the inverted histogram equalized image. Matched filtering intensifies the blood vessels' response in the enhanced image. The multiple threshold values for the maximum matched filter response that maximize the fuzzy entropy are considered to be the optimal thresholds to extract the different types of vessel silhouettes from the background. Differential Evolution algorithm is used to obtain the optimal combination of the fuzzy parameters. Performance evaluated on publicly available DRIVE database demonstrate that the present work outperforms the existing works for various types of vessels extraction and optic disc removal even from poorly illuminated retinal images.
{"title":"Blood vessel extraction with optic disc removal in retinal images","authors":"S. Kar, S. Maity","doi":"10.1109/ICAPR.2015.7050689","DOIUrl":"https://doi.org/10.1109/ICAPR.2015.7050689","url":null,"abstract":"Automatic extraction of retinal blood vessels is an important issue for the diagnosis and the treatment of different retinal disorders. Most of the retinal images are of low contrast due to non-uniform illumination during acquisition process. Therefore, vessel extraction from unevenly illuminated retinal background is really a challenging task. To extract the vessels which lie in the optic disc region, the removal of the optic disc is also important. This paper proposes an algorithm for automatic blood vessel extraction and optic disc removal on poorly illuminated retinal images using curvelet transform, morphological operation, matched filtering and fuzzy entropy maximization. Curvelet transform is used to extract the finest details along the vessels since it can represent the lines, the edges, the curvatures, the missing and the imprecise boundary details efficiently. To remove the optic disc, the curvelet based edge enhanced image is first opened by a disk shaped structuring element which is then subtracted from the inverted histogram equalized image. Matched filtering intensifies the blood vessels' response in the enhanced image. The multiple threshold values for the maximum matched filter response that maximize the fuzzy entropy are considered to be the optimal thresholds to extract the different types of vessel silhouettes from the background. Differential Evolution algorithm is used to obtain the optimal combination of the fuzzy parameters. Performance evaluated on publicly available DRIVE database demonstrate that the present work outperforms the existing works for various types of vessels extraction and optic disc removal even from poorly illuminated retinal images.","PeriodicalId":379764,"journal":{"name":"2015 Eighth International Conference on Advances in Pattern Recognition (ICAPR)","volume":"143 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131949800","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：