Pub Date : 2009-12-04DOI: 10.1109/CCPR.2009.5344151
Yinfeng Luo, Shenglin Yu
Particle filter is an effective method for non-linear filter and it has been gained special attention of researchers in various fields. There will be a new mixed particle filter (PUPF) proposed in this paper based on the general particle filter and the unscented particle filter. lt first uses the general particle filter to generate particles for estimating the state at time k and then a new auxiliary model will be introduced. We would use the unscented particle filter to estimate the state at time k the second time. This structure makes use of the latest observation information, it has small error and better stability. The experimental results indicate that the proposed particle filter's performance outperforms the other four particle filters .The result indicates that the PUPF is a useful method for nonlinear filter problems.
{"title":"A New Mixed Particle Filter Based on an Auxiliary Model","authors":"Yinfeng Luo, Shenglin Yu","doi":"10.1109/CCPR.2009.5344151","DOIUrl":"https://doi.org/10.1109/CCPR.2009.5344151","url":null,"abstract":"Particle filter is an effective method for non-linear filter and it has been gained special attention of researchers in various fields. There will be a new mixed particle filter (PUPF) proposed in this paper based on the general particle filter and the unscented particle filter. lt first uses the general particle filter to generate particles for estimating the state at time k and then a new auxiliary model will be introduced. We would use the unscented particle filter to estimate the state at time k the second time. This structure makes use of the latest observation information, it has small error and better stability. The experimental results indicate that the proposed particle filter's performance outperforms the other four particle filters .The result indicates that the PUPF is a useful method for nonlinear filter problems.","PeriodicalId":354468,"journal":{"name":"2009 Chinese Conference on Pattern Recognition","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125171618","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 : 2009-12-04DOI: 10.1109/CCPR.2009.5343984
Sumuya Bao, Chonghui Guo, Shanglei Chai
Recently spectral clustering has become one of the most popular clustering algorithms. Although it has many advantages, it still has a lot of shortcomings which should be resolved, such as there are a wide variety of spectral clustering algorithms that use the eigenvectors in slightly different ways and many of these algorithms have no proof that they will actually compute a reasonable clustering. The spectral clustering method based on normalized cut criterion is a very efficient spectral clustering method. In this paper, we give a note on why we choose the first k eigenvectors in the algorithm (rationality of the clustering) and the conditions for indicator vectors under which the clustering problem could lead to the problem of minimizing the objective function of the spectral clustering method based on normalized cut criterion.
{"title":"A Note on Spectral Clustering Method Based on Normalized Cut Criterion","authors":"Sumuya Bao, Chonghui Guo, Shanglei Chai","doi":"10.1109/CCPR.2009.5343984","DOIUrl":"https://doi.org/10.1109/CCPR.2009.5343984","url":null,"abstract":"Recently spectral clustering has become one of the most popular clustering algorithms. Although it has many advantages, it still has a lot of shortcomings which should be resolved, such as there are a wide variety of spectral clustering algorithms that use the eigenvectors in slightly different ways and many of these algorithms have no proof that they will actually compute a reasonable clustering. The spectral clustering method based on normalized cut criterion is a very efficient spectral clustering method. In this paper, we give a note on why we choose the first k eigenvectors in the algorithm (rationality of the clustering) and the conditions for indicator vectors under which the clustering problem could lead to the problem of minimizing the objective function of the spectral clustering method based on normalized cut criterion.","PeriodicalId":354468,"journal":{"name":"2009 Chinese Conference on Pattern Recognition","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128226549","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 : 2009-12-04DOI: 10.1109/CCPR.2009.5344021
Xu Pan, Hong-Bin Gu, Chanjuan Sun
In this paper, we introduce a classification approach to identify definitions of all terms from a aviation professional corpus. The corpora of aviation domain are firstly segmented by LTP platform from HIT. Then four feature selection methods and two classifiers are applied to extract definitions. First of all, we summarize the correct proportion of feature subset used in classification of term definitions, and secondly argue that the naive bayes classifier combined with CHI or ODDS for feature selection achieve the best score in the F1-measure and F2-measure. In the end, we recognize that the use of SVM classifier with linear kernel could achieve very high precision, but the worst recall.
{"title":"A Classification Approach to Identify Definitions in Aviation Domain","authors":"Xu Pan, Hong-Bin Gu, Chanjuan Sun","doi":"10.1109/CCPR.2009.5344021","DOIUrl":"https://doi.org/10.1109/CCPR.2009.5344021","url":null,"abstract":"In this paper, we introduce a classification approach to identify definitions of all terms from a aviation professional corpus. The corpora of aviation domain are firstly segmented by LTP platform from HIT. Then four feature selection methods and two classifiers are applied to extract definitions. First of all, we summarize the correct proportion of feature subset used in classification of term definitions, and secondly argue that the naive bayes classifier combined with CHI or ODDS for feature selection achieve the best score in the F1-measure and F2-measure. In the end, we recognize that the use of SVM classifier with linear kernel could achieve very high precision, but the worst recall.","PeriodicalId":354468,"journal":{"name":"2009 Chinese Conference on Pattern Recognition","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130736971","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 : 2009-12-04DOI: 10.1109/CCPR.2009.5343977
Toshinori Miyoshi, T. Nagasaki, Hiroshi Shinjo
Normalization is a particular important preprocessing operation, and has a large effect on the performance of character recognition. One of the purposes of normalization is to regulate the size, position, and shape of character images so as to reduce within-class shape variations. Among various methods of normalization, moment-based normalizations are known to greatly improve the performance of character recognition. However, conventional moment-based normalization methods are susceptible to the variations of stroke length and/or thickness. In order to alleviate this problem, we propose moment normalization methods that use the moments of character contours instead of character images themselves to estimate the transformation parameters. Our experiments show that the proposed methods are effective particularly for printed character recognition.
{"title":"Character Normalization Methods Using Moments of Gradient Features and Normalization Cooperated Feature Extraction","authors":"Toshinori Miyoshi, T. Nagasaki, Hiroshi Shinjo","doi":"10.1109/CCPR.2009.5343977","DOIUrl":"https://doi.org/10.1109/CCPR.2009.5343977","url":null,"abstract":"Normalization is a particular important preprocessing operation, and has a large effect on the performance of character recognition. One of the purposes of normalization is to regulate the size, position, and shape of character images so as to reduce within-class shape variations. Among various methods of normalization, moment-based normalizations are known to greatly improve the performance of character recognition. However, conventional moment-based normalization methods are susceptible to the variations of stroke length and/or thickness. In order to alleviate this problem, we propose moment normalization methods that use the moments of character contours instead of character images themselves to estimate the transformation parameters. Our experiments show that the proposed methods are effective particularly for printed character recognition.","PeriodicalId":354468,"journal":{"name":"2009 Chinese Conference on Pattern Recognition","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130393936","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}
Identification of sentiment orientation in Chinese words is essential for getting sentiment comprehension of Chinese text, and building a basic semantic lexicon with Chinese emotional words will provide a core subset for identifying emotional words in a special area. It can not only help to identify and enlarge semantic lexicon in corpus effectively but also improve classification efficiency. On the basis of the similarity of Chinese words, the paper has proposed a method of calculating sentiment weight of Chinese emotional words. In addition, a dictionary with basic Chinese emotional words has been constructed based on the HowNet semantic lexicon. By utilizing the dictionary together with TF-IDF, we have done experiments to identify sentiment orientation in Chinese text and have got satisfying classification result.
{"title":"A Method of Building Chinese Basic Semantic Lexicon Based on Word Similarity","authors":"Yan-hui Zhu, Zhi-qiang Wen, Ping Wang, Zhao-yi Peng","doi":"10.1109/CCPR.2009.5344041","DOIUrl":"https://doi.org/10.1109/CCPR.2009.5344041","url":null,"abstract":"Identification of sentiment orientation in Chinese words is essential for getting sentiment comprehension of Chinese text, and building a basic semantic lexicon with Chinese emotional words will provide a core subset for identifying emotional words in a special area. It can not only help to identify and enlarge semantic lexicon in corpus effectively but also improve classification efficiency. On the basis of the similarity of Chinese words, the paper has proposed a method of calculating sentiment weight of Chinese emotional words. In addition, a dictionary with basic Chinese emotional words has been constructed based on the HowNet semantic lexicon. By utilizing the dictionary together with TF-IDF, we have done experiments to identify sentiment orientation in Chinese text and have got satisfying classification result.","PeriodicalId":354468,"journal":{"name":"2009 Chinese Conference on Pattern Recognition","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124943394","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 : 2009-12-04DOI: 10.1109/CCPR.2009.5344142
Haiyang Hua, Huaici Zhao
Many machine learning algorithms can be applied only to data described by categorical attributes. So discretizatioti of continuous attributes is one of the important steps in preprocessing of extracting knowledge. Traditional discretization algorithms based on clustering need a pre-determined clustering number k, also typically are applied in an unsupervised learning framework. This paper describes such an algorithm, called SX-means (Supervised X-means), which is a new algorithm of supervised discretization of continuous attributes on clustering. The algorithm modifies clusters with knowledge of the class distribution dynamically. And this procedure can not stop until the proper k is found. For the number of clusters k is not pre-determined by the user and class distribution is applied, the random of result is decreased greatly. Experimental evaluation of several discretization algorithms on six artificial data sets show that the proposed algorithm is more efficient and can generate a better discretization schema. Comparing the output of C4.5, resulting tree is smaller, less classification rules, and high accuracy of classification.
{"title":"A Discretization Algorithm of Continuous Attributes Based on Supervised Clustering","authors":"Haiyang Hua, Huaici Zhao","doi":"10.1109/CCPR.2009.5344142","DOIUrl":"https://doi.org/10.1109/CCPR.2009.5344142","url":null,"abstract":"Many machine learning algorithms can be applied only to data described by categorical attributes. So discretizatioti of continuous attributes is one of the important steps in preprocessing of extracting knowledge. Traditional discretization algorithms based on clustering need a pre-determined clustering number k, also typically are applied in an unsupervised learning framework. This paper describes such an algorithm, called SX-means (Supervised X-means), which is a new algorithm of supervised discretization of continuous attributes on clustering. The algorithm modifies clusters with knowledge of the class distribution dynamically. And this procedure can not stop until the proper k is found. For the number of clusters k is not pre-determined by the user and class distribution is applied, the random of result is decreased greatly. Experimental evaluation of several discretization algorithms on six artificial data sets show that the proposed algorithm is more efficient and can generate a better discretization schema. Comparing the output of C4.5, resulting tree is smaller, less classification rules, and high accuracy of classification.","PeriodicalId":354468,"journal":{"name":"2009 Chinese Conference on Pattern Recognition","volume":"92 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114363223","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 : 2009-12-04DOI: 10.1109/CCPR.2009.5343952
Tomo Miyazaki, S. Omachi, H. Aso
Extracting structure information from decorative character images is a challenging problem in the field of character recognition. The structure information of a decorative character image can be represented by a graph. However, the topologies of graphs are different even if they are the ones of the same character, because of various decorations. In this paper, we propose a method to extract a representative graph of decorative character images. The proposed method extracts graphs from decorative character images, obtains common nodes in a character and iteratively integrates graphs into one common supergraph using common nodes. To show the validly of the proposed method, experiments are carried out using decorative character images.
{"title":"Extraction of Representative Structure of Decorative Character Images","authors":"Tomo Miyazaki, S. Omachi, H. Aso","doi":"10.1109/CCPR.2009.5343952","DOIUrl":"https://doi.org/10.1109/CCPR.2009.5343952","url":null,"abstract":"Extracting structure information from decorative character images is a challenging problem in the field of character recognition. The structure information of a decorative character image can be represented by a graph. However, the topologies of graphs are different even if they are the ones of the same character, because of various decorations. In this paper, we propose a method to extract a representative graph of decorative character images. The proposed method extracts graphs from decorative character images, obtains common nodes in a character and iteratively integrates graphs into one common supergraph using common nodes. To show the validly of the proposed method, experiments are carried out using decorative character images.","PeriodicalId":354468,"journal":{"name":"2009 Chinese Conference on Pattern Recognition","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126531522","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 : 2009-12-04DOI: 10.1109/CCPR.2009.5344118
Yunyun Cao, F. Da, Y. Sui
In order to solve the well-known streaking effects of dynamic programming, an improved algorithm based on stereo matching technology is proposed to generate 3D building model. This algorithm obtains the feature points of the building by Harris corner detector and Canny edge detector to segment the scan lines of dynamic programming. Moreover, a linearly interpolated dissimilarity measure is introduced into the cost computation which further improves the matching speed. The experimental results show that the proposed algorithm can produce smooth and dense 3D points cloud model of building.
{"title":"A Stereo Matching Based 3D Building Reconstruction Algorithm","authors":"Yunyun Cao, F. Da, Y. Sui","doi":"10.1109/CCPR.2009.5344118","DOIUrl":"https://doi.org/10.1109/CCPR.2009.5344118","url":null,"abstract":"In order to solve the well-known streaking effects of dynamic programming, an improved algorithm based on stereo matching technology is proposed to generate 3D building model. This algorithm obtains the feature points of the building by Harris corner detector and Canny edge detector to segment the scan lines of dynamic programming. Moreover, a linearly interpolated dissimilarity measure is introduced into the cost computation which further improves the matching speed. The experimental results show that the proposed algorithm can produce smooth and dense 3D points cloud model of building.","PeriodicalId":354468,"journal":{"name":"2009 Chinese Conference on Pattern Recognition","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127027096","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 : 2009-12-04DOI: 10.1109/CCPR.2009.5344023
Lili Ma, Dongfeng Cai, Lanhai Zhou, Na Ye
This paper proposes a method which is aimed to translate English patent terms into Chinese based on head-driven method. Firstly, word alignment information and English NP parse tree are formed. The corresponding relation between word alignment information and syntactic structure which is built using restrict of head. The NP translation pattern database is formed as the gist of term reordering. Then the intermediate result is translated using statistical method. The best result is chose according to mutual information between each modifier and head. Experimental results show the significant improvements over the current phrase-base SMT system.
{"title":"Term Translation Based on Head-Driven Method","authors":"Lili Ma, Dongfeng Cai, Lanhai Zhou, Na Ye","doi":"10.1109/CCPR.2009.5344023","DOIUrl":"https://doi.org/10.1109/CCPR.2009.5344023","url":null,"abstract":"This paper proposes a method which is aimed to translate English patent terms into Chinese based on head-driven method. Firstly, word alignment information and English NP parse tree are formed. The corresponding relation between word alignment information and syntactic structure which is built using restrict of head. The NP translation pattern database is formed as the gist of term reordering. Then the intermediate result is translated using statistical method. The best result is chose according to mutual information between each modifier and head. Experimental results show the significant improvements over the current phrase-base SMT system.","PeriodicalId":354468,"journal":{"name":"2009 Chinese Conference on Pattern Recognition","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127749077","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 : 2009-12-04DOI: 10.1109/CCPR.2009.5344062
Tong Luo, Yuquan Chen, Jianfeng Li, Jianhua Li
Typical low level segmentation method like level set method can be explained in maximum a posteriori estimation (MAP) for pixel label. In this paper, CRFs model is introduced in label estimation combined with level set to produce fast low level process and accurate high level inference. The energy term in level set evolution is also extended to contain object spatial factors, gradient is provided as the spatial updating basis, besides the temporal characteristic in curve evolution. Unlike simple CRFs model, a feedback machinery is imported in parameters learning, the reasons lie in the fact that CRFs could has small sample size and its modeling approach is mainly rely on model structure, but image patch is a typical local feature which is not directly applied into. With image patch used in the feedback, the accuracy of learning can be improved. At last, energy function is extended to allow complicated multiple regions competition, the local features is merged in the process.
{"title":"Dynamic Estimation of Curve Evolution in Image Segmentation with CRFs Label Inferring","authors":"Tong Luo, Yuquan Chen, Jianfeng Li, Jianhua Li","doi":"10.1109/CCPR.2009.5344062","DOIUrl":"https://doi.org/10.1109/CCPR.2009.5344062","url":null,"abstract":"Typical low level segmentation method like level set method can be explained in maximum a posteriori estimation (MAP) for pixel label. In this paper, CRFs model is introduced in label estimation combined with level set to produce fast low level process and accurate high level inference. The energy term in level set evolution is also extended to contain object spatial factors, gradient is provided as the spatial updating basis, besides the temporal characteristic in curve evolution. Unlike simple CRFs model, a feedback machinery is imported in parameters learning, the reasons lie in the fact that CRFs could has small sample size and its modeling approach is mainly rely on model structure, but image patch is a typical local feature which is not directly applied into. With image patch used in the feedback, the accuracy of learning can be improved. At last, energy function is extended to allow complicated multiple regions competition, the local features is merged in the process.","PeriodicalId":354468,"journal":{"name":"2009 Chinese Conference on Pattern Recognition","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116957659","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：