Computer analysis of images and patterns : proceedings of the ... International Conference on Automatic Image Processing. International Conference on Automatic Image Processing最新文献
Pub Date : 1997-10-26DOI: 10.1109/ICIP.1997.638578
K. Ma, L. Huang
First, we propose a DCT-based subband absolute moment block truncation coding (SAMBTC) method and compare its coding performance with that of QMF-based SAMBTC. Second, the characteristics of the human visual system based on the Weber's law model is incorporated into these schemes, independently. The objective is to further reduce the bit rate without incurring more noticeable degradation. By comparison with the JPEG standard, our image coders suffer much less blocking artifacts at low bit rates and obtain superior subjective image quality.
{"title":"A DCT embedded subband AMBTC image coder","authors":"K. Ma, L. Huang","doi":"10.1109/ICIP.1997.638578","DOIUrl":"https://doi.org/10.1109/ICIP.1997.638578","url":null,"abstract":"First, we propose a DCT-based subband absolute moment block truncation coding (SAMBTC) method and compare its coding performance with that of QMF-based SAMBTC. Second, the characteristics of the human visual system based on the Weber's law model is incorporated into these schemes, independently. The objective is to further reduce the bit rate without incurring more noticeable degradation. By comparison with the JPEG standard, our image coders suffer much less blocking artifacts at low bit rates and obtain superior subjective image quality.","PeriodicalId":92344,"journal":{"name":"Computer analysis of images and patterns : proceedings of the ... International Conference on Automatic Image Processing. International Conference on Automatic Image Processing","volume":"12 1","pages":"645-648 vol.2"},"PeriodicalIF":0.0,"publicationDate":"1997-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87292138","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 : 1997-10-26DOI: 10.1109/ICIP.1997.638848
K. Sawada, Takehiro Yoshida
Resolution scalability refers to a picture coding property where pictures at lower different resolutions can be reconstructed by decoding only the subsets of a single coded bit-stream, while the full resolution picture is reconstructed by decoding the total bit-stream. This paper presents a resolution scalable video coding scheme for interlaced video sequences. The proposed scalable coding scheme employs adaptive in-field/in-frame subband for spatial scalability, where in-field subband and in-frame subband are adaptively used for moving portions and stationary portions respectively. It also employs adaptive field/frame subsampling for temporal scalability. Experimental results have demonstrated that the proposed adaptive subband gives better picture quality especially for reconstructed lower resolution pictures compared to non adaptive subband.
{"title":"Scalable coding based on adaptive subband for interlaced video sequences","authors":"K. Sawada, Takehiro Yoshida","doi":"10.1109/ICIP.1997.638848","DOIUrl":"https://doi.org/10.1109/ICIP.1997.638848","url":null,"abstract":"Resolution scalability refers to a picture coding property where pictures at lower different resolutions can be reconstructed by decoding only the subsets of a single coded bit-stream, while the full resolution picture is reconstructed by decoding the total bit-stream. This paper presents a resolution scalable video coding scheme for interlaced video sequences. The proposed scalable coding scheme employs adaptive in-field/in-frame subband for spatial scalability, where in-field subband and in-frame subband are adaptively used for moving portions and stationary portions respectively. It also employs adaptive field/frame subsampling for temporal scalability. Experimental results have demonstrated that the proposed adaptive subband gives better picture quality especially for reconstructed lower resolution pictures compared to non adaptive subband.","PeriodicalId":92344,"journal":{"name":"Computer analysis of images and patterns : proceedings of the ... International Conference on Automatic Image Processing. International Conference on Automatic Image Processing","volume":"26 1","pages":"621-624 vol.2"},"PeriodicalIF":0.0,"publicationDate":"1997-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87312328","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 : 1997-10-26DOI: 10.1109/ICIP.1997.638582
G. Poulalion, O. Flous, S. Morvan, M. Najim
Tomographic imaging deals with reconstructing an image from its projections. In the electromagnetic field, cross-range projections can be obtained from a spectral analysis of the radar echoes of a rotating target. However, the lack of resolution of classical methods like the Fourier transform leads to very poor quality images. We thus couple a proposed high resolution spectral analysis method with the backprojection algorithm. We illustrate the efficiency of the whole method on a measured radar echo of a target.
{"title":"High resolution radar tomography","authors":"G. Poulalion, O. Flous, S. Morvan, M. Najim","doi":"10.1109/ICIP.1997.638582","DOIUrl":"https://doi.org/10.1109/ICIP.1997.638582","url":null,"abstract":"Tomographic imaging deals with reconstructing an image from its projections. In the electromagnetic field, cross-range projections can be obtained from a spectral analysis of the radar echoes of a rotating target. However, the lack of resolution of classical methods like the Fourier transform leads to very poor quality images. We thus couple a proposed high resolution spectral analysis method with the backprojection algorithm. We illustrate the efficiency of the whole method on a measured radar echo of a target.","PeriodicalId":92344,"journal":{"name":"Computer analysis of images and patterns : proceedings of the ... International Conference on Automatic Image Processing. International Conference on Automatic Image Processing","volume":"241 1","pages":"661-664 vol.2"},"PeriodicalIF":0.0,"publicationDate":"1997-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85174067","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 : 1997-10-26DOI: 10.1109/ICIP.1997.632109
T. Satonaka, T. Baba, T. Otsuki, T. Chikamura, T. Meng
We propose a neural network based on image synthesis, histogram adaptive quantization and the discrete cosine transformation (DCT) for object recognition with luminance, rotation and location invariance. An efficient representation of the invariant features is constructed using a three-dimensional memory structure. The performance of luminance and rotation invariance is illustrated by reduced error rates in face recognition. The error rate of using a two-dimensional DCT is improved from 13.6% to 2.4% with the aid of the proposed image synthesis procedure. The 2.4% error rate is better than all previously reported results using Karhunen-Loeve (1990) transform convolution networks and eigenface models. In using the DCT, our approach also enjoys the additional advantage of greatly reduced computational complexity.
{"title":"Object recognition with luminance, rotation and location invariance","authors":"T. Satonaka, T. Baba, T. Otsuki, T. Chikamura, T. Meng","doi":"10.1109/ICIP.1997.632109","DOIUrl":"https://doi.org/10.1109/ICIP.1997.632109","url":null,"abstract":"We propose a neural network based on image synthesis, histogram adaptive quantization and the discrete cosine transformation (DCT) for object recognition with luminance, rotation and location invariance. An efficient representation of the invariant features is constructed using a three-dimensional memory structure. The performance of luminance and rotation invariance is illustrated by reduced error rates in face recognition. The error rate of using a two-dimensional DCT is improved from 13.6% to 2.4% with the aid of the proposed image synthesis procedure. The 2.4% error rate is better than all previously reported results using Karhunen-Loeve (1990) transform convolution networks and eigenface models. In using the DCT, our approach also enjoys the additional advantage of greatly reduced computational complexity.","PeriodicalId":92344,"journal":{"name":"Computer analysis of images and patterns : proceedings of the ... International Conference on Automatic Image Processing. International Conference on Automatic Image Processing","volume":"125 1","pages":"336-339 vol.3"},"PeriodicalIF":0.0,"publicationDate":"1997-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79514077","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 : 1997-10-26DOI: 10.1109/ICIP.1997.632028
Peter Morguet, M. Lang
A universal approach to the classification of video image sequences by hidden Markov models (HMMs) is presented. The extraction of low level features allows the HMM to build an internal image representation using standard training algorithms. As a result, the states of the HMMs contain probability density functions, so called image density functions, which reflect the structure of the underlying images preserving their geometry. The successful application of the approach to both the recognition of dynamic head and hand gestures demonstrates the universal validity and sensitivity of our method. Even sequences containing only small detail changes are reliably recognized.
{"title":"A universal HMM-based approach to image sequence classification","authors":"Peter Morguet, M. Lang","doi":"10.1109/ICIP.1997.632028","DOIUrl":"https://doi.org/10.1109/ICIP.1997.632028","url":null,"abstract":"A universal approach to the classification of video image sequences by hidden Markov models (HMMs) is presented. The extraction of low level features allows the HMM to build an internal image representation using standard training algorithms. As a result, the states of the HMMs contain probability density functions, so called image density functions, which reflect the structure of the underlying images preserving their geometry. The successful application of the approach to both the recognition of dynamic head and hand gestures demonstrates the universal validity and sensitivity of our method. Even sequences containing only small detail changes are reliably recognized.","PeriodicalId":92344,"journal":{"name":"Computer analysis of images and patterns : proceedings of the ... International Conference on Automatic Image Processing. International Conference on Automatic Image Processing","volume":"13 1","pages":"146-149 vol.3"},"PeriodicalIF":0.0,"publicationDate":"1997-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85341378","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 : 1997-10-26DOI: 10.1109/ICIP.1997.638585
O. Déforges, N. Normand
This paper presents a new algorithm for an efficient implementation of morphological operations for gray images. It defines a recursive morphological decomposition method of convex structuring elements by only causal two pixel structuring elements. Whatever the element size, erosion or/and dilation can then be performed during a unique raster-like image scan, involving a fixed reduced analysis neighborhood. The resulting process offers a low computational complexity, combined with an easiness for describing the element form. The algorithm is exemplified with granulometry. Quantum dots are segmented using a multiscale morphologic decomposition. Our new algorithm is particularly well suited for this type of morphological treatments, as they use structuring elements with both a large size and a form fitting the object to extract, that is to say depending on the application.
{"title":"Recursive morphological operators for gray image processing. Application in granulometry analysis","authors":"O. Déforges, N. Normand","doi":"10.1109/ICIP.1997.638585","DOIUrl":"https://doi.org/10.1109/ICIP.1997.638585","url":null,"abstract":"This paper presents a new algorithm for an efficient implementation of morphological operations for gray images. It defines a recursive morphological decomposition method of convex structuring elements by only causal two pixel structuring elements. Whatever the element size, erosion or/and dilation can then be performed during a unique raster-like image scan, involving a fixed reduced analysis neighborhood. The resulting process offers a low computational complexity, combined with an easiness for describing the element form. The algorithm is exemplified with granulometry. Quantum dots are segmented using a multiscale morphologic decomposition. Our new algorithm is particularly well suited for this type of morphological treatments, as they use structuring elements with both a large size and a form fitting the object to extract, that is to say depending on the application.","PeriodicalId":92344,"journal":{"name":"Computer analysis of images and patterns : proceedings of the ... International Conference on Automatic Image Processing. International Conference on Automatic Image Processing","volume":"9 1","pages":"672-675 vol.2"},"PeriodicalIF":0.0,"publicationDate":"1997-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85314012","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 : 1997-10-26DOI: 10.1109/ICIP.1997.647969
S. Craver, N. Memon, B. Yeo, M. Yeung
We shall show that non-invertibility is a necessary but not sufficient condition in resolving ownership disputes. We then define quasi-invertible watermarking schemes, and, present analysis that links invertibility and quasi-invertibility to some classes of watermarking techniques with different properties (which may or may not require original versions in watermark decoding), as well as to the different classes of attacks we have developed.
{"title":"On the invertibility of invisible watermarking techniques","authors":"S. Craver, N. Memon, B. Yeo, M. Yeung","doi":"10.1109/ICIP.1997.647969","DOIUrl":"https://doi.org/10.1109/ICIP.1997.647969","url":null,"abstract":"We shall show that non-invertibility is a necessary but not sufficient condition in resolving ownership disputes. We then define quasi-invertible watermarking schemes, and, present analysis that links invertibility and quasi-invertibility to some classes of watermarking techniques with different properties (which may or may not require original versions in watermark decoding), as well as to the different classes of attacks we have developed.","PeriodicalId":92344,"journal":{"name":"Computer analysis of images and patterns : proceedings of the ... International Conference on Automatic Image Processing. International Conference on Automatic Image Processing","volume":"20 1","pages":"540-543 vol.1"},"PeriodicalIF":0.0,"publicationDate":"1997-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90890330","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 : 1997-10-26DOI: 10.1109/ICIP.1997.638633
S. Kalitzin, B. H. Romeny, M. Viergever
A hierarchical segmentation model is obtained by using linear scale evolution of gray-scale images. At each scale segments are generated as Voronoi diagrams with a distance measure defined on the image landscape. The set of centers of the Voronoi cells is the set of local extrema of the gray-scale image. This set is localized by using the winding number distribution of the gradient vector field. Scale evolution induces hierarchical structure of embedded segments. Objects defined at coarser scales "decompose" into sub-objects at finer scales. The process is naturally described in terms of singularity catastrophes within the smooth scale evolution. Alternatively, we present a purely topological segmentation procedure, based on singular isophotes. The last are generated by the set of saddle points in the image which are detected also with the topological winding-number method.
{"title":"On topological deep-structure segmentation","authors":"S. Kalitzin, B. H. Romeny, M. Viergever","doi":"10.1109/ICIP.1997.638633","DOIUrl":"https://doi.org/10.1109/ICIP.1997.638633","url":null,"abstract":"A hierarchical segmentation model is obtained by using linear scale evolution of gray-scale images. At each scale segments are generated as Voronoi diagrams with a distance measure defined on the image landscape. The set of centers of the Voronoi cells is the set of local extrema of the gray-scale image. This set is localized by using the winding number distribution of the gradient vector field. Scale evolution induces hierarchical structure of embedded segments. Objects defined at coarser scales \"decompose\" into sub-objects at finer scales. The process is naturally described in terms of singularity catastrophes within the smooth scale evolution. Alternatively, we present a purely topological segmentation procedure, based on singular isophotes. The last are generated by the set of saddle points in the image which are detected also with the topological winding-number method.","PeriodicalId":92344,"journal":{"name":"Computer analysis of images and patterns : proceedings of the ... International Conference on Automatic Image Processing. International Conference on Automatic Image Processing","volume":"33 1","pages":"863-866 vol.2"},"PeriodicalIF":0.0,"publicationDate":"1997-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91164559","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 : 1997-10-26DOI: 10.1109/ICIP.1997.632095
R. Krtolica
Recognition algorithms encompass segmentation, feature extraction and classification, but these components might be difficult to isolate because of strong interactions between them, and the lack of crisp criteria telling where one stops and where the other begins. Extreme variability of text images, and of hand written texts in particular, makes it difficult to tune any of those three parts of a recognition algorithm to real data. Automatic parameter tuning (training or learning) requires parametrization of at least a part of the algorithm. As it is more convenient to parametrize classification than the rest of the recognition algorithm, machine learned recognition usually means that the recognition classifier has been trained or tuned automatically. We show that our box connectivity approach to feature extraction, and localized interpretation within the classifier, provide solutions to the outlined problems, and allow efficient implementation of direct learning.
{"title":"Learning character recognition by localized interpretation of character-images","authors":"R. Krtolica","doi":"10.1109/ICIP.1997.632095","DOIUrl":"https://doi.org/10.1109/ICIP.1997.632095","url":null,"abstract":"Recognition algorithms encompass segmentation, feature extraction and classification, but these components might be difficult to isolate because of strong interactions between them, and the lack of crisp criteria telling where one stops and where the other begins. Extreme variability of text images, and of hand written texts in particular, makes it difficult to tune any of those three parts of a recognition algorithm to real data. Automatic parameter tuning (training or learning) requires parametrization of at least a part of the algorithm. As it is more convenient to parametrize classification than the rest of the recognition algorithm, machine learned recognition usually means that the recognition classifier has been trained or tuned automatically. We show that our box connectivity approach to feature extraction, and localized interpretation within the classifier, provide solutions to the outlined problems, and allow efficient implementation of direct learning.","PeriodicalId":92344,"journal":{"name":"Computer analysis of images and patterns : proceedings of the ... International Conference on Automatic Image Processing. International Conference on Automatic Image Processing","volume":"1 1","pages":"292-295 vol.3"},"PeriodicalIF":0.0,"publicationDate":"1997-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90395884","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 : 1997-10-26DOI: 10.1109/ICIP.1997.647799
M. Robini, T. Rastello, D. Vray, I. Magnin
The synthetic aperture image formation process can be formulated as a space-variant 2D convolution. The recovery of the original reflection density is an ill-posed inverse problem which is both underdetermined and ill-conditioned. Its stabilization is achieved via concave stabilizers that are well adapted to the preservation of discontinuities. This leads to the minimization of a non-convex functional, a task which is successfully carried out using a Metropolis-type annealing algorithm. For improved performance, we investigate some inexpensive acceleration techniques which do not alter the theoretical convergence results; their efficiency is demonstrated through restorations from simulated data.
{"title":"Space-variant deconvolution for synthetic aperture imaging using simulated annealing","authors":"M. Robini, T. Rastello, D. Vray, I. Magnin","doi":"10.1109/ICIP.1997.647799","DOIUrl":"https://doi.org/10.1109/ICIP.1997.647799","url":null,"abstract":"The synthetic aperture image formation process can be formulated as a space-variant 2D convolution. The recovery of the original reflection density is an ill-posed inverse problem which is both underdetermined and ill-conditioned. Its stabilization is achieved via concave stabilizers that are well adapted to the preservation of discontinuities. This leads to the minimization of a non-convex functional, a task which is successfully carried out using a Metropolis-type annealing algorithm. For improved performance, we investigate some inexpensive acceleration techniques which do not alter the theoretical convergence results; their efficiency is demonstrated through restorations from simulated data.","PeriodicalId":92344,"journal":{"name":"Computer analysis of images and patterns : proceedings of the ... International Conference on Automatic Image Processing. International Conference on Automatic Image Processing","volume":"1 1","pages":"432-435 vol.1"},"PeriodicalIF":0.0,"publicationDate":"1997-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90527387","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}
Computer analysis of images and patterns : proceedings of the ... International Conference on Automatic Image Processing. International Conference on Automatic Image Processing
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：