Pub Date : 2013-09-01DOI: 10.1109/IRANIANMVIP.2013.6779959
A. Dehghani, Alistair Sutherland
A local-spatial interest point matching algorithm for articulated human upper body tracking application is proposed in this paper. The first stage finds confidently matched pairs of interest points from the reference and target interest point lists through a local-feature-descriptors-based matching method. Applying two cross-checking and displacement-checking steps reduces the number of mismatched pairs and results confidently matched pairs. Using these confidently matched pairs, the second stage recovers more matched interest point pairs from the remaining unmatched through the graph matching by a cyclic string matching algorithm. The proposed approach benefits from the speed of local matching algorithms as well as the accuracy and robustness of spatial matching methods. In addition, it compensates for the reference list leakage problem. Experimental results show that the combined two-stage interest matching method efficiently improves the matching process for articulated human upper body tracking.
{"title":"A combined two-stage local-spatial interest point matching algorithm","authors":"A. Dehghani, Alistair Sutherland","doi":"10.1109/IRANIANMVIP.2013.6779959","DOIUrl":"https://doi.org/10.1109/IRANIANMVIP.2013.6779959","url":null,"abstract":"A local-spatial interest point matching algorithm for articulated human upper body tracking application is proposed in this paper. The first stage finds confidently matched pairs of interest points from the reference and target interest point lists through a local-feature-descriptors-based matching method. Applying two cross-checking and displacement-checking steps reduces the number of mismatched pairs and results confidently matched pairs. Using these confidently matched pairs, the second stage recovers more matched interest point pairs from the remaining unmatched through the graph matching by a cyclic string matching algorithm. The proposed approach benefits from the speed of local matching algorithms as well as the accuracy and robustness of spatial matching methods. In addition, it compensates for the reference list leakage problem. Experimental results show that the combined two-stage interest matching method efficiently improves the matching process for articulated human upper body tracking.","PeriodicalId":297204,"journal":{"name":"2013 8th Iranian Conference on Machine Vision and Image Processing (MVIP)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115124168","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 : 2013-09-01DOI: 10.1109/IRANIANMVIP.2013.6779946
R. Saniei, K. Faez
Application of the lossless compression method to hide texts is considered as a novel trend in research projects. Evaluation of the proposed methods in the field of steganography reflects a variety of approaches to create covert communication via text files. The extensiveness of steganographic issues and the presence of a huge variety of approaches make it difficult to precisely compare and evaluate these methods. Therefore, in this article a new steganography method that uses a statistical compression technique called `arithmetic coding', will be presented. In addition, the comparison of this method capacity with other methods will be explained. The arithmetic coding technique that has very high compression rates, shall guarantee even a higher growth capacity and higher security compared to its similar techniques. Meanwhile, the secret messages were not revealed through rewriting or syntax/semantic checking and compared with similar methods, increased the capacity by up to 68.9%, and compared with other methods; this method improved the capacity of fifteen times.
{"title":"The capacity of arithmetic compression based text steganography method","authors":"R. Saniei, K. Faez","doi":"10.1109/IRANIANMVIP.2013.6779946","DOIUrl":"https://doi.org/10.1109/IRANIANMVIP.2013.6779946","url":null,"abstract":"Application of the lossless compression method to hide texts is considered as a novel trend in research projects. Evaluation of the proposed methods in the field of steganography reflects a variety of approaches to create covert communication via text files. The extensiveness of steganographic issues and the presence of a huge variety of approaches make it difficult to precisely compare and evaluate these methods. Therefore, in this article a new steganography method that uses a statistical compression technique called `arithmetic coding', will be presented. In addition, the comparison of this method capacity with other methods will be explained. The arithmetic coding technique that has very high compression rates, shall guarantee even a higher growth capacity and higher security compared to its similar techniques. Meanwhile, the secret messages were not revealed through rewriting or syntax/semantic checking and compared with similar methods, increased the capacity by up to 68.9%, and compared with other methods; this method improved the capacity of fifteen times.","PeriodicalId":297204,"journal":{"name":"2013 8th Iranian Conference on Machine Vision and Image Processing (MVIP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115500073","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 : 2013-09-01DOI: 10.1109/IRANIANMVIP.2013.6779976
A. Dehghani, A. Pourmohammad
Image and video processing techniques are one of the commonly used methods for traffics monitoring. This paper investigates the image processing techniques based vehicles speed measurement issue using only a fixed single camera. Therefore, a geometrical calculation based method is proposed. Based on this method, first a moving vehicle is detected in a video background and then the vehicle speed is estimated based on some geometrical calculations. A comparison is made between this method and two other same case vehicles speed measurement methods for evaluation. The simulations results on 160×112 pixels real recorded video images shows the average vehicles speed error less than %10.
{"title":"Single camera vehicles speed measurement","authors":"A. Dehghani, A. Pourmohammad","doi":"10.1109/IRANIANMVIP.2013.6779976","DOIUrl":"https://doi.org/10.1109/IRANIANMVIP.2013.6779976","url":null,"abstract":"Image and video processing techniques are one of the commonly used methods for traffics monitoring. This paper investigates the image processing techniques based vehicles speed measurement issue using only a fixed single camera. Therefore, a geometrical calculation based method is proposed. Based on this method, first a moving vehicle is detected in a video background and then the vehicle speed is estimated based on some geometrical calculations. A comparison is made between this method and two other same case vehicles speed measurement methods for evaluation. The simulations results on 160×112 pixels real recorded video images shows the average vehicles speed error less than %10.","PeriodicalId":297204,"journal":{"name":"2013 8th Iranian Conference on Machine Vision and Image Processing (MVIP)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125200408","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 : 2013-09-01DOI: 10.1109/IRANIANMVIP.2013.6779965
Elham Mohammadi, E. Fatemizadeh, H. Sheikhzadeh, Sahar Khoubani
Breast cancer is considered as the most important cause of death among women. Architectural distortions are very important signs of breast cancer and early detection of them is a rewarding work. In this paper we propose a method to recognize architectural distortion from normal parenchyma. In our proposed method, appropriate features are extracted by the analysis of oriented textures with the application of orientation component of recent the state-of-the-art local texture descriptor called Monogenic Binary Coding (MBC). In addition, we transform Region of Interests (ROIs) to polar coordinates in order to highlight some specific patterns in mammograms. Various classifiers are used over a set of mammograms from Digital Database for Screening Mammography (DDSM). The results show that proposed method is very encouraging. The best performance achieved is 91.25% in terms of the average accuracy using the Nearest Neighbor classifier.
{"title":"A textural approach for recognizing architectural distortion in mammograms","authors":"Elham Mohammadi, E. Fatemizadeh, H. Sheikhzadeh, Sahar Khoubani","doi":"10.1109/IRANIANMVIP.2013.6779965","DOIUrl":"https://doi.org/10.1109/IRANIANMVIP.2013.6779965","url":null,"abstract":"Breast cancer is considered as the most important cause of death among women. Architectural distortions are very important signs of breast cancer and early detection of them is a rewarding work. In this paper we propose a method to recognize architectural distortion from normal parenchyma. In our proposed method, appropriate features are extracted by the analysis of oriented textures with the application of orientation component of recent the state-of-the-art local texture descriptor called Monogenic Binary Coding (MBC). In addition, we transform Region of Interests (ROIs) to polar coordinates in order to highlight some specific patterns in mammograms. Various classifiers are used over a set of mammograms from Digital Database for Screening Mammography (DDSM). The results show that proposed method is very encouraging. The best performance achieved is 91.25% in terms of the average accuracy using the Nearest Neighbor classifier.","PeriodicalId":297204,"journal":{"name":"2013 8th Iranian Conference on Machine Vision and Image Processing (MVIP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130551089","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 : 2013-09-01DOI: 10.1109/IRANIANMVIP.2013.6779947
S. M. Marvasti-Zadeh, Hossein Ghanei-Yakhdan, S. Kasaei
Error concealment is a useful method for improving the damaged video quality in the decoder side. In this paper, a dynamic method with low computational complexity is presented to improve the visual quality of videos when up to 50% of the frames are damaged. In the proposed method, temporal replacement and the improved outer boundary matching algorithm are used for dynamical error concealment in inter-frames of videos. With the use of motion vectors (MVs) which are close to the damaged macroblock (MB) the method can determine whether the motion in specific areas is either regular, irregular, or zero. Then, based on this knowledge, different methods are performed. It adaptively selects a set of candidate MVs and external boundaries for comparison purposes. Furthermore, to increase the accuracy, depending on the correctness of adjacent MVs, a specific weight is given to the boundaries of adjacent MBs. Experimental results show that the proposed method enhances both objective and subjective quality of damaged frames without any considerable increase in complexity. The average PSNR in some frames of test sequences has increased about 1.01 dB more than the outer boundary matching algorithm.
{"title":"Dynamic temporal error concealment for video data in error-prone environments","authors":"S. M. Marvasti-Zadeh, Hossein Ghanei-Yakhdan, S. Kasaei","doi":"10.1109/IRANIANMVIP.2013.6779947","DOIUrl":"https://doi.org/10.1109/IRANIANMVIP.2013.6779947","url":null,"abstract":"Error concealment is a useful method for improving the damaged video quality in the decoder side. In this paper, a dynamic method with low computational complexity is presented to improve the visual quality of videos when up to 50% of the frames are damaged. In the proposed method, temporal replacement and the improved outer boundary matching algorithm are used for dynamical error concealment in inter-frames of videos. With the use of motion vectors (MVs) which are close to the damaged macroblock (MB) the method can determine whether the motion in specific areas is either regular, irregular, or zero. Then, based on this knowledge, different methods are performed. It adaptively selects a set of candidate MVs and external boundaries for comparison purposes. Furthermore, to increase the accuracy, depending on the correctness of adjacent MVs, a specific weight is given to the boundaries of adjacent MBs. Experimental results show that the proposed method enhances both objective and subjective quality of damaged frames without any considerable increase in complexity. The average PSNR in some frames of test sequences has increased about 1.01 dB more than the outer boundary matching algorithm.","PeriodicalId":297204,"journal":{"name":"2013 8th Iranian Conference on Machine Vision and Image Processing (MVIP)","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125717100","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 : 2013-09-01DOI: 10.1109/IRANIANMVIP.2013.6779950
S. Mehralian, M. Palhang
In this paper we proposed a new method for pedestrian detection in images and videos. Our method uses a sliding window to search through images. In order to extract the features, each window is divided into overlapping cells and features are extracted from them. The feature that we extracted to describe each window is based on analysis of gradient distribution of each cell. After gradient distribution of a cell computed, the PCA is applied on it and using a mathematical expression that gauges the attitude of edges we got the feature of the cell. Putting the features of the cells next to each other forms the feature vector of the window. Then, the extracted features are classified using Support Vector Machine (SVM). Finally, the learned SVM model tested on the INRIA pedestrian dataset. The proposed method was compared with Histograms of Oriented Gradient (HOG) approach and the results show that our method has comparable detection accuracy as well as having more robustness when facing with noise.
{"title":"Pedestrian detection using principal components analysis of gradient distribution","authors":"S. Mehralian, M. Palhang","doi":"10.1109/IRANIANMVIP.2013.6779950","DOIUrl":"https://doi.org/10.1109/IRANIANMVIP.2013.6779950","url":null,"abstract":"In this paper we proposed a new method for pedestrian detection in images and videos. Our method uses a sliding window to search through images. In order to extract the features, each window is divided into overlapping cells and features are extracted from them. The feature that we extracted to describe each window is based on analysis of gradient distribution of each cell. After gradient distribution of a cell computed, the PCA is applied on it and using a mathematical expression that gauges the attitude of edges we got the feature of the cell. Putting the features of the cells next to each other forms the feature vector of the window. Then, the extracted features are classified using Support Vector Machine (SVM). Finally, the learned SVM model tested on the INRIA pedestrian dataset. The proposed method was compared with Histograms of Oriented Gradient (HOG) approach and the results show that our method has comparable detection accuracy as well as having more robustness when facing with noise.","PeriodicalId":297204,"journal":{"name":"2013 8th Iranian Conference on Machine Vision and Image Processing (MVIP)","volume":"149 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121401501","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 : 2013-09-01DOI: 10.1109/IRANIANMVIP.2013.6779986
P. Rasti, H. Demirel, G. Anbarjafari
In this paper, we propose a new super resolution technique based on the interpolation followed by registering them using iterative back projection (IBP). Firstly the low resolution image is interpolated and then decimate it to four lower low resolution images. The four low resolution images are interpolated and registered by using IBP in order to generate a sharper high resolution image. The proposed method has been tested on Lena, Elaine, Pepper, and Baboon. The quantitative peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) results as well as the visual results show the superiority of the proposed technique over the conventional and alternative image super resolution techniques. For Lena's image, the PSNR is 6.21 dB higher than the bicubic interpolation.
{"title":"Iterative back projection based image resolution enhancement","authors":"P. Rasti, H. Demirel, G. Anbarjafari","doi":"10.1109/IRANIANMVIP.2013.6779986","DOIUrl":"https://doi.org/10.1109/IRANIANMVIP.2013.6779986","url":null,"abstract":"In this paper, we propose a new super resolution technique based on the interpolation followed by registering them using iterative back projection (IBP). Firstly the low resolution image is interpolated and then decimate it to four lower low resolution images. The four low resolution images are interpolated and registered by using IBP in order to generate a sharper high resolution image. The proposed method has been tested on Lena, Elaine, Pepper, and Baboon. The quantitative peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) results as well as the visual results show the superiority of the proposed technique over the conventional and alternative image super resolution techniques. For Lena's image, the PSNR is 6.21 dB higher than the bicubic interpolation.","PeriodicalId":297204,"journal":{"name":"2013 8th Iranian Conference on Machine Vision and Image Processing (MVIP)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129310987","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 : 2013-09-01DOI: 10.1109/IRANIANMVIP.2013.6780002
M. Mohammadi, E. Fatemizadeh
The Local Binary Patterns (LBP) feature extraction method is a theoretically and computationally simple and efficient methodology for texture analysis. The LBP operator is used in many applications such as facial expression recognition and face recognition. The original LBP is based on hard thresholding the neighborhood of each pixel, which makes texture representation sensitive to noise. In addition, LBP cannot distinguish between a strong and a weak pattern. In order to enhance the LBP approach, Fuzzy Local Binary Patterns (FLBP) is proposed. In FLBP, any neighborhood does not represented only by one code, but, it is represented by all existing codes with different degrees. In FLBP, any fuzzy Intersection and Union operators may be used. In this study, the following operators are applied and their results are compared together: Dot-Sum, Min-Max and normalized Min-Max. Based on the extensive experiments, the fuzzy Min-Max operators are more useful and can improve the accuracy in the application of Facial Expression Recognition (FER) about 4% (i.e., form 82.98% to 86.88%).
{"title":"Fuzzy local binary patterns: A comparison between Min-Max and Dot-Sum operators in the application of facial expression recognition","authors":"M. Mohammadi, E. Fatemizadeh","doi":"10.1109/IRANIANMVIP.2013.6780002","DOIUrl":"https://doi.org/10.1109/IRANIANMVIP.2013.6780002","url":null,"abstract":"The Local Binary Patterns (LBP) feature extraction method is a theoretically and computationally simple and efficient methodology for texture analysis. The LBP operator is used in many applications such as facial expression recognition and face recognition. The original LBP is based on hard thresholding the neighborhood of each pixel, which makes texture representation sensitive to noise. In addition, LBP cannot distinguish between a strong and a weak pattern. In order to enhance the LBP approach, Fuzzy Local Binary Patterns (FLBP) is proposed. In FLBP, any neighborhood does not represented only by one code, but, it is represented by all existing codes with different degrees. In FLBP, any fuzzy Intersection and Union operators may be used. In this study, the following operators are applied and their results are compared together: Dot-Sum, Min-Max and normalized Min-Max. Based on the extensive experiments, the fuzzy Min-Max operators are more useful and can improve the accuracy in the application of Facial Expression Recognition (FER) about 4% (i.e., form 82.98% to 86.88%).","PeriodicalId":297204,"journal":{"name":"2013 8th Iranian Conference on Machine Vision and Image Processing (MVIP)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129253472","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 : 2013-09-01DOI: 10.1109/IRANIANMVIP.2013.6780025
A. Nazemi, M. Shafiee, Z. Azimifar
Automatic vehicle Make and Model Recognition (MMR) system offers a competent way to vehicle classification and recognition systems. This paper proposes a real time while robust vehicle make and model recognition system extracting the vehicle sub-image from the background and studies some sparse feature coding methods such as Orthogonal Matching Pursuit (OMP), some variation of Sparse Coding (SC) methods and compares them to choose the best one. Our method employs the sparse feature coding methods on dense Scale-Invariant Feature Transform (SIFT) features and Support Vector Machine (SVM) for classification. The proposed system is examined by an Iranian on road vehicles dataset, which its samples are in different point of views, various weather conditions and illuminations.
{"title":"On road vehicle make and model recognition via sparse feature coding","authors":"A. Nazemi, M. Shafiee, Z. Azimifar","doi":"10.1109/IRANIANMVIP.2013.6780025","DOIUrl":"https://doi.org/10.1109/IRANIANMVIP.2013.6780025","url":null,"abstract":"Automatic vehicle Make and Model Recognition (MMR) system offers a competent way to vehicle classification and recognition systems. This paper proposes a real time while robust vehicle make and model recognition system extracting the vehicle sub-image from the background and studies some sparse feature coding methods such as Orthogonal Matching Pursuit (OMP), some variation of Sparse Coding (SC) methods and compares them to choose the best one. Our method employs the sparse feature coding methods on dense Scale-Invariant Feature Transform (SIFT) features and Support Vector Machine (SVM) for classification. The proposed system is examined by an Iranian on road vehicles dataset, which its samples are in different point of views, various weather conditions and illuminations.","PeriodicalId":297204,"journal":{"name":"2013 8th Iranian Conference on Machine Vision and Image Processing (MVIP)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115297495","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 : 2013-09-01DOI: 10.1109/IRANIANMVIP.2013.6780009
S. Rahimi, A. Aghagolzadeh, Hadi Seyedarabi
Human tracking is an interesting topic in computer vision domain. In this paper, a human detection and tracking algorithm based on new features combination in one camera system is proposed. In detection part, first, mixture of Gaussian background subtraction method is used to find moving regions, then histogram of oriented gradient (HOG) feature of these regions are extracted. At the end, SVM classifier is used to distinguish human from non-human according to their HOG features. In tracking part, first, color, cellular local binary pattern (Cell-LBP) and HOG features of humans are extracted, then their next positions are estimated using particle filter framework. Color, Cell-LBP and HOG features are used to model humans. Color is an effective feature in dealing with object deformation and partial occlusion but has some restriction in cases where background or objects have same color. Cell-LBP is an improved texture descriptor that is robust against partial occlusion, this feature compensates color's restriction. HOG is a shape descriptor that can separate humans from background and is robust against illumination changes. Combination of these three features improves tracking result despite challenges like partial occlusion, object's deformation and illumination changes. Experimental results show advantage of the proposed algorithm.
{"title":"Human detection and tracking using new features combination in particle filter framework","authors":"S. Rahimi, A. Aghagolzadeh, Hadi Seyedarabi","doi":"10.1109/IRANIANMVIP.2013.6780009","DOIUrl":"https://doi.org/10.1109/IRANIANMVIP.2013.6780009","url":null,"abstract":"Human tracking is an interesting topic in computer vision domain. In this paper, a human detection and tracking algorithm based on new features combination in one camera system is proposed. In detection part, first, mixture of Gaussian background subtraction method is used to find moving regions, then histogram of oriented gradient (HOG) feature of these regions are extracted. At the end, SVM classifier is used to distinguish human from non-human according to their HOG features. In tracking part, first, color, cellular local binary pattern (Cell-LBP) and HOG features of humans are extracted, then their next positions are estimated using particle filter framework. Color, Cell-LBP and HOG features are used to model humans. Color is an effective feature in dealing with object deformation and partial occlusion but has some restriction in cases where background or objects have same color. Cell-LBP is an improved texture descriptor that is robust against partial occlusion, this feature compensates color's restriction. HOG is a shape descriptor that can separate humans from background and is robust against illumination changes. Combination of these three features improves tracking result despite challenges like partial occlusion, object's deformation and illumination changes. Experimental results show advantage of the proposed algorithm.","PeriodicalId":297204,"journal":{"name":"2013 8th Iranian Conference on Machine Vision and Image Processing (MVIP)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116212947","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：