Pub Date : 2003-09-17DOI: 10.1109/ICIAP.2003.1234036
G. Foresti, C. Micheloni, L. Snidaro, C. Marchiol
In this paper, a real-time face detection system for color image sequences is presented. The system applies three different face detection methods and integrates the obtained results to achieve a greater location accuracy. The first method localizes the human head through outline analysis, focusing the attention of the system on a small image area. The second, a skin color method, is applied to the blobs to find skin regions (e.g., faces, hands, etc.). The third. principal component analysis, is used to reduce the dimensionality of the data set and to detect face patterns. Finally. the obtained face locations are fused to increase the detection reliability and to avoid false detections due to occlusions or unfavorable human poses. The proposed approach is used by a video-based surveillance system for monitoring indoor scenes.
{"title":"Face detection for visual surveillance","authors":"G. Foresti, C. Micheloni, L. Snidaro, C. Marchiol","doi":"10.1109/ICIAP.2003.1234036","DOIUrl":"https://doi.org/10.1109/ICIAP.2003.1234036","url":null,"abstract":"In this paper, a real-time face detection system for color image sequences is presented. The system applies three different face detection methods and integrates the obtained results to achieve a greater location accuracy. The first method localizes the human head through outline analysis, focusing the attention of the system on a small image area. The second, a skin color method, is applied to the blobs to find skin regions (e.g., faces, hands, etc.). The third. principal component analysis, is used to reduce the dimensionality of the data set and to detect face patterns. Finally. the obtained face locations are fused to increase the detection reliability and to avoid false detections due to occlusions or unfavorable human poses. The proposed approach is used by a video-based surveillance system for monitoring indoor scenes.","PeriodicalId":218076,"journal":{"name":"12th International Conference on Image Analysis and Processing, 2003.Proceedings.","volume":"491 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123059709","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 : 2003-09-17DOI: 10.1109/ICIAP.2003.1234089
Ghilès Mostafaoui, C. Achard, M. Milgram
The problem of moving person tracking, without knowledge about the number of persons in the scene, and by taking into account occlusion, under-segmentation and over-segmentation, is challenging. A first motion detection gives us regions with several segmentation problems due to bad acquisition conditions. The tracking step, which has to manage all these problems, is realized with the EM algorithm (expectation maximization). It uses a kinematic model: we suppose a rectilinear and uniform apparent motion, this hypothesis seems very restrictive but remains locally accurate in most applications. Good results are obtained with this approach on several sequences, without any initialization.
{"title":"Trajectories extraction from image sequences based on kinematic","authors":"Ghilès Mostafaoui, C. Achard, M. Milgram","doi":"10.1109/ICIAP.2003.1234089","DOIUrl":"https://doi.org/10.1109/ICIAP.2003.1234089","url":null,"abstract":"The problem of moving person tracking, without knowledge about the number of persons in the scene, and by taking into account occlusion, under-segmentation and over-segmentation, is challenging. A first motion detection gives us regions with several segmentation problems due to bad acquisition conditions. The tracking step, which has to manage all these problems, is realized with the EM algorithm (expectation maximization). It uses a kinematic model: we suppose a rectilinear and uniform apparent motion, this hypothesis seems very restrictive but remains locally accurate in most applications. Good results are obtained with this approach on several sequences, without any initialization.","PeriodicalId":218076,"journal":{"name":"12th International Conference on Image Analysis and Processing, 2003.Proceedings.","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127377264","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 : 2003-09-17DOI: 10.1109/ICIAP.2003.1234123
F. Gasparini, R. Schettini
The paper describes a reliable and rapid method for detecting and removing a color cast (i.e. a superimposed dominant color) in a digital image without any a priori knowledge of its semantic content. A multi-step algorithm classifies the input images as having no cast, evident cast, ambiguous cast, or intrinsic cast (images presenting a cast due to a predominant color that must be preserved). If an evident or ambiguous cast is found, a cast remover step, a modified version of the white balance algorithm, is then applied in the two cases of evident or ambiguous casts. The method we propose has been tuned and tested with positive results on a data set of over 650 images.
{"title":"Color correction for digital photographs","authors":"F. Gasparini, R. Schettini","doi":"10.1109/ICIAP.2003.1234123","DOIUrl":"https://doi.org/10.1109/ICIAP.2003.1234123","url":null,"abstract":"The paper describes a reliable and rapid method for detecting and removing a color cast (i.e. a superimposed dominant color) in a digital image without any a priori knowledge of its semantic content. A multi-step algorithm classifies the input images as having no cast, evident cast, ambiguous cast, or intrinsic cast (images presenting a cast due to a predominant color that must be preserved). If an evident or ambiguous cast is found, a cast remover step, a modified version of the white balance algorithm, is then applied in the two cases of evident or ambiguous casts. The method we propose has been tuned and tested with positive results on a data set of over 650 images.","PeriodicalId":218076,"journal":{"name":"12th International Conference on Image Analysis and Processing, 2003.Proceedings.","volume":"111 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126076058","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 : 2003-09-17DOI: 10.1109/ICIAP.2003.1234070
L. D. Stefano, S. Mattoccia, M. Mola
This work proposes a novel technique aimed at improving the performance of exhaustive template matching based on the normalized cross correlation (NCC). An effective sufficient condition, capable of rapidly pruning those match candidates that could not provide a better cross correlation score with respect to the current best candidate, can be obtained exploiting an upper bound of the NCC function. This upper bound relies on partial evaluation of the crosscorrelation and can be computed efficiently, yielding a significant reduction of operations compared to the NCC function and allows for reducing the overall number of operations required to carry out exhaustive searches. However, the bounded partial correlation (BPC) algorithm turns out to be significantly data dependent. In this paper we propose a novel algorithm that improves the overall performance of BPC thanks to the deployment of a more selective sufficient condition which allows for rendering the algorithm significantly less data dependent. Experimental results with real images and actual CPU time are reported.
{"title":"An efficient algorithm for exhaustive template matching based on normalized cross correlation","authors":"L. D. Stefano, S. Mattoccia, M. Mola","doi":"10.1109/ICIAP.2003.1234070","DOIUrl":"https://doi.org/10.1109/ICIAP.2003.1234070","url":null,"abstract":"This work proposes a novel technique aimed at improving the performance of exhaustive template matching based on the normalized cross correlation (NCC). An effective sufficient condition, capable of rapidly pruning those match candidates that could not provide a better cross correlation score with respect to the current best candidate, can be obtained exploiting an upper bound of the NCC function. This upper bound relies on partial evaluation of the crosscorrelation and can be computed efficiently, yielding a significant reduction of operations compared to the NCC function and allows for reducing the overall number of operations required to carry out exhaustive searches. However, the bounded partial correlation (BPC) algorithm turns out to be significantly data dependent. In this paper we propose a novel algorithm that improves the overall performance of BPC thanks to the deployment of a more selective sufficient condition which allows for rendering the algorithm significantly less data dependent. Experimental results with real images and actual CPU time are reported.","PeriodicalId":218076,"journal":{"name":"12th International Conference on Image Analysis and Processing, 2003.Proceedings.","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126111947","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 : 2003-09-17DOI: 10.1109/ICIAP.2003.1234125
E. Ardizzone, M. Cascia, Fabio Testa
A new algorithm for allocating a given bit rate to different image tiles in the JPEG2000 encoding system is proposed. The algorithm outperforms other approaches commonly used in implementations. The new algorithm is suitable when information content is not equally distributed across the image. It is based on the computation of an index of the information content of each tile. To implement the proposed approach, we modified JasPer, a free software-based JPEG2000 coder implementation (Adams, M.D. and Kossentini, F., Proc. IEEE Int. Conf. on Image Process., vol.2, p.53-6, 2000). The experimentation was carried out on a subset of the JPEG2000 test images. Experimental results are reported, showing the PSNR of the decompressed images to be better than the one produced with the traditional approach, when the information content is not equally distributed across the image, and to be comparable to that of the traditional approach, when the information distribution is quite uniform.
在JPEG2000编码系统中，提出了一种将给定比特率分配给不同图像块的新算法。该算法优于实现中常用的其他方法。新算法适用于信息内容不均匀分布的情况。它基于对每个tile的信息内容的索引的计算。为了实现所提出的方法，我们修改了JasPer，一个基于JPEG2000编码器的免费软件实现(Adams, M.D.和Kossentini, F.， Proc. IEEE Int.)。关于图像处理。，第2卷，第53-6页，2000)。实验是在JPEG2000测试图像的一个子集上进行的。实验结果表明，在信息内容不均匀分布的情况下，解压缩后的图像的PSNR优于传统方法;在信息分布较为均匀的情况下，解压缩后的图像的PSNR与传统方法相当。
{"title":"A new algorithm for bit rate allocation in JPEG2000 tile encoding","authors":"E. Ardizzone, M. Cascia, Fabio Testa","doi":"10.1109/ICIAP.2003.1234125","DOIUrl":"https://doi.org/10.1109/ICIAP.2003.1234125","url":null,"abstract":"A new algorithm for allocating a given bit rate to different image tiles in the JPEG2000 encoding system is proposed. The algorithm outperforms other approaches commonly used in implementations. The new algorithm is suitable when information content is not equally distributed across the image. It is based on the computation of an index of the information content of each tile. To implement the proposed approach, we modified JasPer, a free software-based JPEG2000 coder implementation (Adams, M.D. and Kossentini, F., Proc. IEEE Int. Conf. on Image Process., vol.2, p.53-6, 2000). The experimentation was carried out on a subset of the JPEG2000 test images. Experimental results are reported, showing the PSNR of the decompressed images to be better than the one produced with the traditional approach, when the information content is not equally distributed across the image, and to be comparable to that of the traditional approach, when the information distribution is quite uniform.","PeriodicalId":218076,"journal":{"name":"12th International Conference on Image Analysis and Processing, 2003.Proceedings.","volume":"261 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122714893","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 : 2003-09-17DOI: 10.1109/ICIAP.2003.1234112
F. Chaker, M. Bannour, F. Ghorbel
We propose here a study of a new affine-invariant Fourier descriptors (Ghorbel (1998)) which are computed on the projection of a given curve that is assumed to be evolving on three dimensional space and supposed to be far enough from the camera. This set of descriptors is compared to the well known affine curvature. These invariants satisfy the completeness and stability properties.
{"title":"A complete and stable set of affine-invariant Fourier descriptors","authors":"F. Chaker, M. Bannour, F. Ghorbel","doi":"10.1109/ICIAP.2003.1234112","DOIUrl":"https://doi.org/10.1109/ICIAP.2003.1234112","url":null,"abstract":"We propose here a study of a new affine-invariant Fourier descriptors (Ghorbel (1998)) which are computed on the projection of a given curve that is assumed to be evolving on three dimensional space and supposed to be far enough from the camera. This set of descriptors is compared to the well known affine curvature. These invariants satisfy the completeness and stability properties.","PeriodicalId":218076,"journal":{"name":"12th International Conference on Image Analysis and Processing, 2003.Proceedings.","volume":"88 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122524696","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 : 2003-09-17DOI: 10.1109/ICIAP.2003.1234037
Ho-Man Tang, Michael R. Lyu, Irwin King
We propose a dynamic face recognition committee machine (DFRCM) consisting of five well-known state-of-the-art algorithms in this paper. In previous work, we have developed a static committee machine which outperforms all the individual algorithms in the experiments. However, the weight for each expert in the committee is fixed and cannot be changed once the system is trained. We propose a dynamic architecture on the committee machine which uses the input face image in the gating network to improve the overall performance. In addition, we adopt a feedback mechanism on the committee machine to adjust the weight of an individual algorithm according to the performance of the algorithm. Detailed experimental results of different algorithms and the committee machine are given to demonstrate the effectiveness of the proposed system.
{"title":"Face recognition committee machines: dynamic vs. static structures","authors":"Ho-Man Tang, Michael R. Lyu, Irwin King","doi":"10.1109/ICIAP.2003.1234037","DOIUrl":"https://doi.org/10.1109/ICIAP.2003.1234037","url":null,"abstract":"We propose a dynamic face recognition committee machine (DFRCM) consisting of five well-known state-of-the-art algorithms in this paper. In previous work, we have developed a static committee machine which outperforms all the individual algorithms in the experiments. However, the weight for each expert in the committee is fixed and cannot be changed once the system is trained. We propose a dynamic architecture on the committee machine which uses the input face image in the gating network to improve the overall performance. In addition, we adopt a feedback mechanism on the committee machine to adjust the weight of an individual algorithm according to the performance of the algorithm. Detailed experimental results of different algorithms and the committee machine are given to demonstrate the effectiveness of the proposed system.","PeriodicalId":218076,"journal":{"name":"12th International Conference on Image Analysis and Processing, 2003.Proceedings.","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123027040","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 : 2003-09-17DOI: 10.1109/ICIAP.2003.1234069
A. Mérigot
The paper presents a modified version of the classical split and merge algorithm (Horowitz, S. and Pavlidis, T., 1976). Instead of performing a regular decomposition of the image, it relies on a split at an optimal position that makes a good interregion separation. The implementation of the algorithm uses an initial image preprocessing to speed-up computation. Experimental results show that the number of regions generated by the split phase is largely reduced and that the distortion of the segmented image is smaller, while the execution time is slightly increased.
本文提出了经典分割和合并算法的改进版本(Horowitz, S. and Pavlidis, T.， 1976)。它不是对图像进行常规分解，而是依赖于在最佳位置进行分割，从而实现良好的区域间分离。该算法的实现采用初始图像预处理来加快计算速度。实验结果表明，分割相位产生的区域数量大大减少，分割图像的失真较小，而执行时间略有增加。
{"title":"Revisiting image splitting","authors":"A. Mérigot","doi":"10.1109/ICIAP.2003.1234069","DOIUrl":"https://doi.org/10.1109/ICIAP.2003.1234069","url":null,"abstract":"The paper presents a modified version of the classical split and merge algorithm (Horowitz, S. and Pavlidis, T., 1976). Instead of performing a regular decomposition of the image, it relies on a split at an optimal position that makes a good interregion separation. The implementation of the algorithm uses an initial image preprocessing to speed-up computation. Experimental results show that the number of regions generated by the split phase is largely reduced and that the distortion of the segmented image is smaller, while the execution time is slightly increased.","PeriodicalId":218076,"journal":{"name":"12th International Conference on Image Analysis and Processing, 2003.Proceedings.","volume":"116 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124611131","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 : 2003-09-17DOI: 10.1109/ICIAP.2003.1234105
I. Kunttu, Leena Lepistö, J. Rauhamaa, A. Visa
The description of object shape is an important characteristic of an image. In image processing and pattern recognition, several different shape descriptors are used. In human visual perception, shapes are processed in multiple resolutions. Therefore, multiscale shape representation is essential in shape based image classification and retrieval. In the description of an object shape, the multiresolution representation provides also additional accuracy to the shape classification. We introduce a new descriptor for shape classification. This descriptor is called the multiscale Fourier descriptor, and it combines the benefits of a Fourier descriptor and multiscale shape representation. This descriptor is formed by applying a Fourier transform to the coefficients of the wavelet transform of the object boundary. In this way, the Fourier descriptor can be presented in multiple resolutions. We performed classification experiments using three image databases. The classification results of our method are compared to those of Fourier descriptors.
{"title":"Multiscale Fourier descriptor for shape classification","authors":"I. Kunttu, Leena Lepistö, J. Rauhamaa, A. Visa","doi":"10.1109/ICIAP.2003.1234105","DOIUrl":"https://doi.org/10.1109/ICIAP.2003.1234105","url":null,"abstract":"The description of object shape is an important characteristic of an image. In image processing and pattern recognition, several different shape descriptors are used. In human visual perception, shapes are processed in multiple resolutions. Therefore, multiscale shape representation is essential in shape based image classification and retrieval. In the description of an object shape, the multiresolution representation provides also additional accuracy to the shape classification. We introduce a new descriptor for shape classification. This descriptor is called the multiscale Fourier descriptor, and it combines the benefits of a Fourier descriptor and multiscale shape representation. This descriptor is formed by applying a Fourier transform to the coefficients of the wavelet transform of the object boundary. In this way, the Fourier descriptor can be presented in multiple resolutions. We performed classification experiments using three image databases. The classification results of our method are compared to those of Fourier descriptors.","PeriodicalId":218076,"journal":{"name":"12th International Conference on Image Analysis and Processing, 2003.Proceedings.","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129910543","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 : 2003-09-17DOI: 10.1109/ICIAP.2003.1234096
Richard C. Wilson, E. Hancock
Although graph structures have proved useful in high level vision for object recognition and matching, they can prove computationally cumbersome because of the need to establish reliable correspondences between nodes. Hence, standard pattern recognition techniques cannot be easily applied to graphs since feature vectors are not easily constructed. To overcome this problem, we turn to the spectral matrix. We show how the elements of this matrix can be used to construct symmetric polynomials that are permutation invariant. The coefficients of these polynomials can be used as graph-features which can be encoded in a vectorial manner. Hence, the symmetric polynomials lead to a representation which is invariant under node permutations and so represents the graph structure without the need for labelling or correspondence operations. We demonstrate that these features are complete and continuous for 'simple' graphs (those without repeated eigenvalues in their spectrum). The notions of stability and discrimination are discussed, and we present experimental evaluation of these properties. Finally, we show that these graph characterizations can be used to cluster graphs from real datasets.
{"title":"Pattern spaces from graph polynomials","authors":"Richard C. Wilson, E. Hancock","doi":"10.1109/ICIAP.2003.1234096","DOIUrl":"https://doi.org/10.1109/ICIAP.2003.1234096","url":null,"abstract":"Although graph structures have proved useful in high level vision for object recognition and matching, they can prove computationally cumbersome because of the need to establish reliable correspondences between nodes. Hence, standard pattern recognition techniques cannot be easily applied to graphs since feature vectors are not easily constructed. To overcome this problem, we turn to the spectral matrix. We show how the elements of this matrix can be used to construct symmetric polynomials that are permutation invariant. The coefficients of these polynomials can be used as graph-features which can be encoded in a vectorial manner. Hence, the symmetric polynomials lead to a representation which is invariant under node permutations and so represents the graph structure without the need for labelling or correspondence operations. We demonstrate that these features are complete and continuous for 'simple' graphs (those without repeated eigenvalues in their spectrum). The notions of stability and discrimination are discussed, and we present experimental evaluation of these properties. Finally, we show that these graph characterizations can be used to cluster graphs from real datasets.","PeriodicalId":218076,"journal":{"name":"12th International Conference on Image Analysis and Processing, 2003.Proceedings.","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124138903","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：