Pub Date : 2012-10-01DOI: 10.1109/IPTA.2012.6469553
Muhammad Asad, W. Ikram
In order to facilitate the visually impaired person in navigation, we have developed a prototype guidance system. The main assumption of this guidance system is that there are many straight paths in different real world scenarios. These straight paths have parallel edges, which when captured as an image seem to converge to a single point called the vanishing point. Proper feature extraction and mathematical modelling of the captured frame leads to the detection of these parallel edges. The vanishing point is then calculated and a decision system is formed which notifies the blind person about his/her deviation from a straight path. The scope of this system is limited to a straight path and has been tested in different lighting conditions and with different level of occlusion. A laptop mounted on a 2D robotic platform is used to develop and verify the robustness of the algorithm. Finally, a smartphone based real-time application has been implemented for this visual guidance system, in which the decision system returns an audio output to guide the visually impaired person. This application has an average execution rate of 20 frames per second, with each frame being of 320 by 240 pixel size. The system has an accuracy of 84.1% in a scenario with pedestrians and other objects, while without pedestrians it produces an accuracy of over 90%.
{"title":"Smartphone based guidance system for visually impaired person","authors":"Muhammad Asad, W. Ikram","doi":"10.1109/IPTA.2012.6469553","DOIUrl":"https://doi.org/10.1109/IPTA.2012.6469553","url":null,"abstract":"In order to facilitate the visually impaired person in navigation, we have developed a prototype guidance system. The main assumption of this guidance system is that there are many straight paths in different real world scenarios. These straight paths have parallel edges, which when captured as an image seem to converge to a single point called the vanishing point. Proper feature extraction and mathematical modelling of the captured frame leads to the detection of these parallel edges. The vanishing point is then calculated and a decision system is formed which notifies the blind person about his/her deviation from a straight path. The scope of this system is limited to a straight path and has been tested in different lighting conditions and with different level of occlusion. A laptop mounted on a 2D robotic platform is used to develop and verify the robustness of the algorithm. Finally, a smartphone based real-time application has been implemented for this visual guidance system, in which the decision system returns an audio output to guide the visually impaired person. This application has an average execution rate of 20 frames per second, with each frame being of 320 by 240 pixel size. The system has an accuracy of 84.1% in a scenario with pedestrians and other objects, while without pedestrians it produces an accuracy of over 90%.","PeriodicalId":267290,"journal":{"name":"2012 3rd International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115604403","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 : 2012-10-01DOI: 10.1109/IPTA.2012.6469574
M. Oussalah, S. Wang
Automatic facial expression recognition from video clips is a challenging task due to computational complexity, limitations of image analysis and subjectivity. This paper advocates a fuzzy based approach for emotion classification. On the other hand, several proposals have been put forward to enhance the pre-processing stage prior to the classification. This includes a combination of a boundary elliptical model for skin detection, adaptive thresholding, principal component analysis and use of cam-shift for face tracking. The performances of the developed system have been evaluated using TFEID and video clips and compared with Bayes' classifier.
{"title":"Fuzzy emotion recognition model for video sequences","authors":"M. Oussalah, S. Wang","doi":"10.1109/IPTA.2012.6469574","DOIUrl":"https://doi.org/10.1109/IPTA.2012.6469574","url":null,"abstract":"Automatic facial expression recognition from video clips is a challenging task due to computational complexity, limitations of image analysis and subjectivity. This paper advocates a fuzzy based approach for emotion classification. On the other hand, several proposals have been put forward to enhance the pre-processing stage prior to the classification. This includes a combination of a boundary elliptical model for skin detection, adaptive thresholding, principal component analysis and use of cam-shift for face tracking. The performances of the developed system have been evaluated using TFEID and video clips and compared with Bayes' classifier.","PeriodicalId":267290,"journal":{"name":"2012 3rd International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"143 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114231868","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 : 2012-10-01DOI: 10.1109/IPTA.2012.6469525
V. Cantoni, A. Ferone, O. Ozbudak, A. Petrosino
This paper presents an approach for protein motif retrieval founded on protein secondary structures (SSs) in 3D. This is a new way to analyze the protein 3D structure. In this approach, based on the Generalized Hough Transform (GHT), the primitives are the triangles defined by the midpoints of three SSs. The three distances between each SSs couple are used in searching and in the voting process. The barycenter of the motif is assigned as the Reference Point (RP). All motif triangles are compared with all possible triangles in the macromolecule. The lengths of triangle edges are used as selective parameters. For every correspondence a vote is given to the point which is figured out as motif barycenter with a special mapping rule and the point having most votes is determined as candidate RP. In this paper we made some experiments for retrieval of four- and five-SSs motif from the macromolecule. Experimental results showed that the RP is determined with precision and this new approach to retrieve the motif is simple to implement, computationally efficient and fast.
{"title":"Search of protein structural blocks through secondary structure triplets","authors":"V. Cantoni, A. Ferone, O. Ozbudak, A. Petrosino","doi":"10.1109/IPTA.2012.6469525","DOIUrl":"https://doi.org/10.1109/IPTA.2012.6469525","url":null,"abstract":"This paper presents an approach for protein motif retrieval founded on protein secondary structures (SSs) in 3D. This is a new way to analyze the protein 3D structure. In this approach, based on the Generalized Hough Transform (GHT), the primitives are the triangles defined by the midpoints of three SSs. The three distances between each SSs couple are used in searching and in the voting process. The barycenter of the motif is assigned as the Reference Point (RP). All motif triangles are compared with all possible triangles in the macromolecule. The lengths of triangle edges are used as selective parameters. For every correspondence a vote is given to the point which is figured out as motif barycenter with a special mapping rule and the point having most votes is determined as candidate RP. In this paper we made some experiments for retrieval of four- and five-SSs motif from the macromolecule. Experimental results showed that the RP is determined with precision and this new approach to retrieve the motif is simple to implement, computationally efficient and fast.","PeriodicalId":267290,"journal":{"name":"2012 3rd International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"120 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116739337","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 : 2012-10-01DOI: 10.1109/IPTA.2012.6469547
Qiong Nie, S. Bouchafa, A. Mérigot
This paper deals with 3D scene reconstruction from an on-board moving camera in the context of automatic driver assistance systems. The aim of our study is to detect any kind of parameterized surface from a moving camera without camera calibration or any prior knowledge about the vehicle egomotion. We assume that the 3D scene is a set of 3D planes that are classified into three categories according to their orientation: lateral planes (buildings), horizontal planes (the road) and frontal planes (moving cars or crossing pedestrians). We propose an iterative voting process that takes advantages of some specific iso-velocity curves properties in order to build a set of appropriate voting spaces. Each of them facilitates the detection of a specific plane model. A tough problem as the detection of a parameterized surface from a moving camera is reduced to an easy maxima finding in several voting spaces. We focus in this paper on the iterative scheme that allows to deal with several spaces at the same time. We choose to adapt an histogram splitting approach in order to achieve a complete plane detection process.
{"title":"Voting spaces cooperation for 3D plane detection from monocular image sequences","authors":"Qiong Nie, S. Bouchafa, A. Mérigot","doi":"10.1109/IPTA.2012.6469547","DOIUrl":"https://doi.org/10.1109/IPTA.2012.6469547","url":null,"abstract":"This paper deals with 3D scene reconstruction from an on-board moving camera in the context of automatic driver assistance systems. The aim of our study is to detect any kind of parameterized surface from a moving camera without camera calibration or any prior knowledge about the vehicle egomotion. We assume that the 3D scene is a set of 3D planes that are classified into three categories according to their orientation: lateral planes (buildings), horizontal planes (the road) and frontal planes (moving cars or crossing pedestrians). We propose an iterative voting process that takes advantages of some specific iso-velocity curves properties in order to build a set of appropriate voting spaces. Each of them facilitates the detection of a specific plane model. A tough problem as the detection of a parameterized surface from a moving camera is reduced to an easy maxima finding in several voting spaces. We focus in this paper on the iterative scheme that allows to deal with several spaces at the same time. We choose to adapt an histogram splitting approach in order to achieve a complete plane detection process.","PeriodicalId":267290,"journal":{"name":"2012 3rd International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"83 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116229737","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 : 2012-10-01DOI: 10.1109/IPTA.2012.6469538
R. Bakken, Lars Moland Eliassen
Human pose estimation is the process of approximating the configuration of the body's underlying skeletal articulation in one or more frames. The curve-skeleton of an object is a line-like representation that preserves topology and geometrical information. Finding the curve-skeleton of a volume corresponding to the person is a good starting point for approximating the underlying skeletal structure. In this paper a GPU implementation of a fully parallel thinning algorithm based on the critical kernels framework is presented. The algorithm is compared to another state-of-the-art thinning method, and while it is demonstrated that both achieve real-time frame rates, the proposed algorithm yields superior accuracy and robustness when used in a pose estimation context. The GPU implementation is > 8× faster than a sequential version, and the positions of the four extremities are estimated with rms error ~6 cm and ~98 % of frames correctly labelled.
{"title":"Real-time 3D skeletonisation in computer vision-based human pose estimation using GPGPU","authors":"R. Bakken, Lars Moland Eliassen","doi":"10.1109/IPTA.2012.6469538","DOIUrl":"https://doi.org/10.1109/IPTA.2012.6469538","url":null,"abstract":"Human pose estimation is the process of approximating the configuration of the body's underlying skeletal articulation in one or more frames. The curve-skeleton of an object is a line-like representation that preserves topology and geometrical information. Finding the curve-skeleton of a volume corresponding to the person is a good starting point for approximating the underlying skeletal structure. In this paper a GPU implementation of a fully parallel thinning algorithm based on the critical kernels framework is presented. The algorithm is compared to another state-of-the-art thinning method, and while it is demonstrated that both achieve real-time frame rates, the proposed algorithm yields superior accuracy and robustness when used in a pose estimation context. The GPU implementation is > 8× faster than a sequential version, and the positions of the four extremities are estimated with rms error ~6 cm and ~98 % of frames correctly labelled.","PeriodicalId":267290,"journal":{"name":"2012 3rd International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124892412","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 : 2012-10-01DOI: 10.1109/IPTA.2012.6469561
Lavdie Rada, Ke Chen, B. Ghanbari
Total variational segmentation models provide effective tools for identifying all features and their boundaries in two and three dimensional images and have been proven to be useful and successful. Speeding up a simulation is one of the remaining challenges. In this paper we propose a restarted homotopy analysis method to improve the computational efficiency in three-dimensional image segmentation. The algorithm replaces the nonlinear variational problem by a sequence of linear approximations by working with linear equations instead of nonlinear ones which lead to efficient energy minimization while maintaining the segmentation quality. Experimental results will show that the computational efficiency is significantly improved.
{"title":"A restarted iterative homotopy analysis method for three-dimensional image segmentation","authors":"Lavdie Rada, Ke Chen, B. Ghanbari","doi":"10.1109/IPTA.2012.6469561","DOIUrl":"https://doi.org/10.1109/IPTA.2012.6469561","url":null,"abstract":"Total variational segmentation models provide effective tools for identifying all features and their boundaries in two and three dimensional images and have been proven to be useful and successful. Speeding up a simulation is one of the remaining challenges. In this paper we propose a restarted homotopy analysis method to improve the computational efficiency in three-dimensional image segmentation. The algorithm replaces the nonlinear variational problem by a sequence of linear approximations by working with linear equations instead of nonlinear ones which lead to efficient energy minimization while maintaining the segmentation quality. Experimental results will show that the computational efficiency is significantly improved.","PeriodicalId":267290,"journal":{"name":"2012 3rd International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128436709","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 : 2012-10-01DOI: 10.1109/IPTA.2012.6469567
Marco Aldinucci, C. Spampinato, M. Drocco, M. Torquati, S. Palazzo
In this paper a two-phase filter for removing “salt and pepper” noise is proposed. In the first phase, an adaptive median filter is used to identify the set of the noisy pixels; in the second phase, these pixels are restored according to a regularization method, which contains a data-fidelity term reflecting the impulse noise characteristics. The algorithm, which exhibits good performance both in denoising and in restoration, can be easily and effectively parallelized to exploit the full power of multi-core CPUs and GPGPUs; the proposed implementation based on the FastFlow library achieves both close-to-ideal speedup and very good wall-clock execution figures.
{"title":"A parallel edge preserving algorithm for salt and pepper image denoising","authors":"Marco Aldinucci, C. Spampinato, M. Drocco, M. Torquati, S. Palazzo","doi":"10.1109/IPTA.2012.6469567","DOIUrl":"https://doi.org/10.1109/IPTA.2012.6469567","url":null,"abstract":"In this paper a two-phase filter for removing “salt and pepper” noise is proposed. In the first phase, an adaptive median filter is used to identify the set of the noisy pixels; in the second phase, these pixels are restored according to a regularization method, which contains a data-fidelity term reflecting the impulse noise characteristics. The algorithm, which exhibits good performance both in denoising and in restoration, can be easily and effectively parallelized to exploit the full power of multi-core CPUs and GPGPUs; the proposed implementation based on the FastFlow library achieves both close-to-ideal speedup and very good wall-clock execution figures.","PeriodicalId":267290,"journal":{"name":"2012 3rd International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116339416","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 : 2012-10-01DOI: 10.1109/IPTA.2012.6469543
Merza Klaghstan, R. Hänsch, David Coquil, O. Hellwich
Image categorization refers to the process of assigning images to a number of predefined categories. The difficulty of problem solving is proportional to the number of categories the system addresses. This paper proposes an image categorization system, and studies the impact of dividing the categorization problem into smaller problems in a hierarchical structure. We compare examples solved with and without the proposed approach, to conclude its pros and cons.
{"title":"Impact of hierarchical structures in image categorization systems","authors":"Merza Klaghstan, R. Hänsch, David Coquil, O. Hellwich","doi":"10.1109/IPTA.2012.6469543","DOIUrl":"https://doi.org/10.1109/IPTA.2012.6469543","url":null,"abstract":"Image categorization refers to the process of assigning images to a number of predefined categories. The difficulty of problem solving is proportional to the number of categories the system addresses. This paper proposes an image categorization system, and studies the impact of dividing the categorization problem into smaller problems in a hierarchical structure. We compare examples solved with and without the proposed approach, to conclude its pros and cons.","PeriodicalId":267290,"journal":{"name":"2012 3rd International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"118 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128178336","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 : 2012-10-01DOI: 10.1109/IPTA.2012.6469570
M. A. Larhmam, S. Mahmoudi, M. Benjelloun
Vertebra detection presents the first step of any automatic spinal column diagnosis. This task becomes more difficult in the case of the cervical X-ray images characterized by their low contrasts and noise due to skull bones. In this paper, we describe an efficient modified template matching method for detecting cervical vertebrae using Generalized Hough Transform (GHT). The proposed method consists of three main steps toward vertebrae detection: 1) Offline training to obtain a robust average model of cervical vertebra. 2) Detecting the potential vertebra centers. 3) Adaptive Post-processing filter. X-ray Image data of 40 healthy cases were used to validate our approach by using a total of 200 cervical vertebrae. We obtained an accuracy of 89%.
{"title":"Semi-automatic detection of cervical vertebrae in X-ray images using generalized hough transform","authors":"M. A. Larhmam, S. Mahmoudi, M. Benjelloun","doi":"10.1109/IPTA.2012.6469570","DOIUrl":"https://doi.org/10.1109/IPTA.2012.6469570","url":null,"abstract":"Vertebra detection presents the first step of any automatic spinal column diagnosis. This task becomes more difficult in the case of the cervical X-ray images characterized by their low contrasts and noise due to skull bones. In this paper, we describe an efficient modified template matching method for detecting cervical vertebrae using Generalized Hough Transform (GHT). The proposed method consists of three main steps toward vertebrae detection: 1) Offline training to obtain a robust average model of cervical vertebra. 2) Detecting the potential vertebra centers. 3) Adaptive Post-processing filter. X-ray Image data of 40 healthy cases were used to validate our approach by using a total of 200 cervical vertebrae. We obtained an accuracy of 89%.","PeriodicalId":267290,"journal":{"name":"2012 3rd International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130307893","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 : 2012-10-01DOI: 10.1109/IPTA.2012.6469506
Meriem Hacini, K. Djemal, F. Hachouf
In this paper, a novel image denoising algorithm based on a multiplicative regularization technique is proposed. The regularization employs a weighted total variation (TV) that is included as a multiplicative constraint. In this way, the appropriate regularization parameter will be controlled by the optimization process itself. The new proposed method not only overcomes the disadvantage of generating artificial edges but also has the advantages of denoising and edges preservation of TV model. Experimental results show that the new method is effective in removing speckle noise and image details are kept well.
{"title":"Speckled images restoration filter based on weighted multiplicative regularization approach","authors":"Meriem Hacini, K. Djemal, F. Hachouf","doi":"10.1109/IPTA.2012.6469506","DOIUrl":"https://doi.org/10.1109/IPTA.2012.6469506","url":null,"abstract":"In this paper, a novel image denoising algorithm based on a multiplicative regularization technique is proposed. The regularization employs a weighted total variation (TV) that is included as a multiplicative constraint. In this way, the appropriate regularization parameter will be controlled by the optimization process itself. The new proposed method not only overcomes the disadvantage of generating artificial edges but also has the advantages of denoising and edges preservation of TV model. Experimental results show that the new method is effective in removing speckle noise and image details are kept well.","PeriodicalId":267290,"journal":{"name":"2012 3rd International Conference on Image Processing Theory, Tools and Applications (IPTA)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121265650","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：