A document image retrieval method based on the paragraph feature combining up with the relative difference of local pixel distributions and high-dimensional data index structure is developed in the article .First, the definitions of paragraph feature and relative difference of local pixel distributions are given, then the extraction method is described in detail, so is the retrieval method based on the two features. As a global feature, the paragraph feature combined with the relative difference of local pixel distributions, which is a local feature, could describe the document image sufficiently and give a good distinguish. High-dimensional data index structure used in the method could greatly improve the velocity. Based on them, an efficient retrieval method is derived.
{"title":"A Feature-Based Document Image Retrieval Method","authors":"Tian Zhang","doi":"10.1109/CCPR.2008.76","DOIUrl":"https://doi.org/10.1109/CCPR.2008.76","url":null,"abstract":"A document image retrieval method based on the paragraph feature combining up with the relative difference of local pixel distributions and high-dimensional data index structure is developed in the article .First, the definitions of paragraph feature and relative difference of local pixel distributions are given, then the extraction method is described in detail, so is the retrieval method based on the two features. As a global feature, the paragraph feature combined with the relative difference of local pixel distributions, which is a local feature, could describe the document image sufficiently and give a good distinguish. High-dimensional data index structure used in the method could greatly improve the velocity. Based on them, an efficient retrieval method is derived.","PeriodicalId":292956,"journal":{"name":"2008 Chinese Conference on Pattern Recognition","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127633770","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}
The processing effect and recognition accuracy of RMB is an important part in the paper currency classification system. According to the characteristics of RMB images, the paper uses the theory of digital image processing and pattern recognition to put forward the method of RMB image processing based on the processing and recognition of the part of the RMB serial numbers, the arithmetic of linear perception based on rewards and punishment method and the extraction method of serial numbers character. Through the experiment on the paper currency classification system which uses the CIS sensor as the image acquisition, it testifies that this method of recognition has a high feasibility and recognition accuracy.
{"title":"Application of Image Processing Technology in Paper Currency Classification System","authors":"Wenhong Li, Yonggang Li, Kexue Luo","doi":"10.1109/CCPR.2008.96","DOIUrl":"https://doi.org/10.1109/CCPR.2008.96","url":null,"abstract":"The processing effect and recognition accuracy of RMB is an important part in the paper currency classification system. According to the characteristics of RMB images, the paper uses the theory of digital image processing and pattern recognition to put forward the method of RMB image processing based on the processing and recognition of the part of the RMB serial numbers, the arithmetic of linear perception based on rewards and punishment method and the extraction method of serial numbers character. Through the experiment on the paper currency classification system which uses the CIS sensor as the image acquisition, it testifies that this method of recognition has a high feasibility and recognition accuracy.","PeriodicalId":292956,"journal":{"name":"2008 Chinese Conference on Pattern Recognition","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127840177","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}
Shadow mapping is an efficient algorithm for real-time shadow rendering. This algorithm is image based. When a shadow map with limited resolution maps in to a large virtual scene, it will suffer from the aliasing errors due to it's insufficient sample. In this paper we present a real-time shadow mapping algorithm. We render some sub shadow maps respectively instead of render only one shadow map and then map them into virtual scene respectively. This algorithm can handle aliasing errors by virtually increasing the resolution of the shadow map beyond the GPU hardware limit. Further more, a method called "deferred shadowing" is used to speed up the shadowing algorithm we have presented to meet the real-time rendering requirement in large virtual environment.
{"title":"A Real-Time Anti-Aliasing Shadow Algorithm Based on Shadow Maps","authors":"Xu Hu, Yue Qi, Xukun Shen","doi":"10.1109/CCPR.2008.27","DOIUrl":"https://doi.org/10.1109/CCPR.2008.27","url":null,"abstract":"Shadow mapping is an efficient algorithm for real-time shadow rendering. This algorithm is image based. When a shadow map with limited resolution maps in to a large virtual scene, it will suffer from the aliasing errors due to it's insufficient sample. In this paper we present a real-time shadow mapping algorithm. We render some sub shadow maps respectively instead of render only one shadow map and then map them into virtual scene respectively. This algorithm can handle aliasing errors by virtually increasing the resolution of the shadow map beyond the GPU hardware limit. Further more, a method called \"deferred shadowing\" is used to speed up the shadowing algorithm we have presented to meet the real-time rendering requirement in large virtual environment.","PeriodicalId":292956,"journal":{"name":"2008 Chinese Conference on Pattern Recognition","volume":"117 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123103222","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}
Most of 3D modeling tools' operating interface are complex. In this paper, a fast free sketching 3D modeling method with edge understanding approached. Free sketching is this method's input. A 3D model was constructed in a short time. Some self-intersection graph can be recognized though add some edge index information in this method.
{"title":"Fast 3D Model Construct Based on Free Sketching","authors":"Yakui Li, Hongwei Li, Hanqing Lu","doi":"10.1109/CCPR.2008.29","DOIUrl":"https://doi.org/10.1109/CCPR.2008.29","url":null,"abstract":"Most of 3D modeling tools' operating interface are complex. In this paper, a fast free sketching 3D modeling method with edge understanding approached. Free sketching is this method's input. A 3D model was constructed in a short time. Some self-intersection graph can be recognized though add some edge index information in this method.","PeriodicalId":292956,"journal":{"name":"2008 Chinese Conference on Pattern Recognition","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116647642","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}
Hong Lu, S. Fei, Zhang Tao, Tao Li, Jianyong Zheng
Object states estimation and data association are main facets of multi-object tracking. Under complex situations, one object often grouped with others, or occluded by other objects or background, which can increase the difficulty in data association and behaviour reasoning, and even result in being hard to maintain track. To cope with these problems, a new behaviour reasoning and tracking method is proposed. A scene mode is built firstly simulating public transport environment. Then, Kalman filter is used to estimate the object position and bounding box. Then, the center-association based projection ratio and region-association based occlusion ratio are defined and combined to reason and classify object behaviours. Finally, an adaptive tracking scheme aiming at special behaviours (e.g. merging, and static occlusion) is proposed to validate our reasoning method. Experimental results show that the proposed method is efficient.
{"title":"Behaviour Reasoning and Tracking for Mobile Objects Based on Projection and Occlusion Ratios","authors":"Hong Lu, S. Fei, Zhang Tao, Tao Li, Jianyong Zheng","doi":"10.1109/CCPR.2008.42","DOIUrl":"https://doi.org/10.1109/CCPR.2008.42","url":null,"abstract":"Object states estimation and data association are main facets of multi-object tracking. Under complex situations, one object often grouped with others, or occluded by other objects or background, which can increase the difficulty in data association and behaviour reasoning, and even result in being hard to maintain track. To cope with these problems, a new behaviour reasoning and tracking method is proposed. A scene mode is built firstly simulating public transport environment. Then, Kalman filter is used to estimate the object position and bounding box. Then, the center-association based projection ratio and region-association based occlusion ratio are defined and combined to reason and classify object behaviours. Finally, an adaptive tracking scheme aiming at special behaviours (e.g. merging, and static occlusion) is proposed to validate our reasoning method. Experimental results show that the proposed method is efficient.","PeriodicalId":292956,"journal":{"name":"2008 Chinese Conference on Pattern Recognition","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123769773","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}
Based on the analysis of advantages and disadvantages of dynamic histogram method and intensity -pair based method, we present an adaptive contrast enhancement method based on detection of histogram feature. The results of experiments show that it is superior in image processing compared with brightness preserving method and the above-mentioned two methods. It has the benefits of the mentioned methods and also eliminates their defects. It has excellent flexibility and practicability so that can be used extensively.
{"title":"Adaptive Contrast Enhancement Based on Detection of Histogram Feature","authors":"Sijing Chen, Chenyang Ge","doi":"10.1109/CCPR.2008.33","DOIUrl":"https://doi.org/10.1109/CCPR.2008.33","url":null,"abstract":"Based on the analysis of advantages and disadvantages of dynamic histogram method and intensity -pair based method, we present an adaptive contrast enhancement method based on detection of histogram feature. The results of experiments show that it is superior in image processing compared with brightness preserving method and the above-mentioned two methods. It has the benefits of the mentioned methods and also eliminates their defects. It has excellent flexibility and practicability so that can be used extensively.","PeriodicalId":292956,"journal":{"name":"2008 Chinese Conference on Pattern Recognition","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125137408","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}
In this paper, we propose a feature extraction method called two dimension locally principal component analysis (2DLPCA) for face recognition, which is based directly image matrix rather than 1D image vectors. 2DLPCA seeks to discover the intrinsic image local structure. This local structure may contain useful information for discrimination. Experimental results on ORL face database show the effectiveness of the proposed algorithm.
{"title":"Two Dimension Locally Principal Component Analysis for Face Recognition","authors":"Yu-sheng Lin, Jianguo Wang, Jing-yu Yang","doi":"10.1109/CCPR.2008.52","DOIUrl":"https://doi.org/10.1109/CCPR.2008.52","url":null,"abstract":"In this paper, we propose a feature extraction method called two dimension locally principal component analysis (2DLPCA) for face recognition, which is based directly image matrix rather than 1D image vectors. 2DLPCA seeks to discover the intrinsic image local structure. This local structure may contain useful information for discrimination. Experimental results on ORL face database show the effectiveness of the proposed algorithm.","PeriodicalId":292956,"journal":{"name":"2008 Chinese Conference on Pattern Recognition","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133165723","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}
In the field of face recognition (FR), the techniques that can provide effective feature representation with enhanced discriminability are crucial. Although the separable wavelet transform has played an important role in image processing, it suffers from some shortcomings which badly affect its ability for feature generation. In this paper, we propose a novel FR system combining the dual-tree complex wavelet transform (WT) and a two-step discriminant analysis, which is called discriminant complex- WT-face system. We introduce an effective feature construction method to provide more representative feature. Then the two-step discriminant analysis is used to filter and extract enhanced-discriminability features. Experimental results show that the discriminant complex-WT-face system can significantly outperform some other FR systems.
{"title":"Discriminant Dual-Tree Complex Wavelet Features for Face Recognition","authors":"Chao-Chun Liu, D. Dai","doi":"10.1109/CCPR.2008.53","DOIUrl":"https://doi.org/10.1109/CCPR.2008.53","url":null,"abstract":"In the field of face recognition (FR), the techniques that can provide effective feature representation with enhanced discriminability are crucial. Although the separable wavelet transform has played an important role in image processing, it suffers from some shortcomings which badly affect its ability for feature generation. In this paper, we propose a novel FR system combining the dual-tree complex wavelet transform (WT) and a two-step discriminant analysis, which is called discriminant complex- WT-face system. We introduce an effective feature construction method to provide more representative feature. Then the two-step discriminant analysis is used to filter and extract enhanced-discriminability features. Experimental results show that the discriminant complex-WT-face system can significantly outperform some other FR systems.","PeriodicalId":292956,"journal":{"name":"2008 Chinese Conference on Pattern Recognition","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131786098","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}
As to H.264/AVC watermarking technique, it is a key issue to select some proper image blocks for watermark embedding, that can reduce the increase of watermarked-stream bitrate, strengthen the robustness of watermark and improve the quality of the watermarked video. A new H.264/AVC watermarking algorithm based on the sort of prediction-residual energy is proposed in this paper. Through comprehensively predicting the prediction-residual energy of all image blocks, The selecting priority of each image block for watermarking embedding is set according to the prediction-residual energy, the larger the value, the higher the selecting priority. The experiment results show the good performance on bitrate, transparency and robustness.
{"title":"A H.264/AVC Digital Watermarking Algorithm Based on Prediction-Residual Energy Priority","authors":"Jian-Long Tang, A. Wen","doi":"10.1109/CCPR.2008.72","DOIUrl":"https://doi.org/10.1109/CCPR.2008.72","url":null,"abstract":"As to H.264/AVC watermarking technique, it is a key issue to select some proper image blocks for watermark embedding, that can reduce the increase of watermarked-stream bitrate, strengthen the robustness of watermark and improve the quality of the watermarked video. A new H.264/AVC watermarking algorithm based on the sort of prediction-residual energy is proposed in this paper. Through comprehensively predicting the prediction-residual energy of all image blocks, The selecting priority of each image block for watermarking embedding is set according to the prediction-residual energy, the larger the value, the higher the selecting priority. The experiment results show the good performance on bitrate, transparency and robustness.","PeriodicalId":292956,"journal":{"name":"2008 Chinese Conference on Pattern Recognition","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131792375","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}
The method and performance for the state prediction based on the LM neural network was investigated. The way applied to the state prediction of on-board electromechanical BIT was provided. The slide oil pressure affects and reflects the run state of engine, which is adopted as the typical test data to validate the availability of LM neural network. Result shows that the state prediction and integrative analysis with the dynamic and history information can conquer such shortcomings as the low diagnose ability and high false alarm rate etc in the traditional BIT. The prediction precision is high and convergence rate is quick.
{"title":"Investigation on the State Prediction of the On-board Electromechanical BIT Based on the LM Neural Network","authors":"Chuang Guo, Yin-Hui Li, Jian Wang","doi":"10.1109/CCPR.2008.81","DOIUrl":"https://doi.org/10.1109/CCPR.2008.81","url":null,"abstract":"The method and performance for the state prediction based on the LM neural network was investigated. The way applied to the state prediction of on-board electromechanical BIT was provided. The slide oil pressure affects and reflects the run state of engine, which is adopted as the typical test data to validate the availability of LM neural network. Result shows that the state prediction and integrative analysis with the dynamic and history information can conquer such shortcomings as the low diagnose ability and high false alarm rate etc in the traditional BIT. The prediction precision is high and convergence rate is quick.","PeriodicalId":292956,"journal":{"name":"2008 Chinese Conference on Pattern Recognition","volume":"111 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132110972","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：