Pub Date : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207440
Juan Huang, Hong Chen, Yanfang Tao
The goal of semi-supervised learning algorithm is to effectively incorporate labeled and unlabeled data in a general-purpose learner with small misclassification error. Although there are various algorithms to implement semi-supervised learning task, the crucial issue of dependence of generalization error on the number of labeled and unlabeled data is still poorly understood. In this paper, we consider the Laplacian Support Vector Machines (LapSVMs) and establish its error analysis.
{"title":"Analysis of Laplacian Support Vector Machines","authors":"Juan Huang, Hong Chen, Yanfang Tao","doi":"10.1109/ICWAPR.2009.5207440","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207440","url":null,"abstract":"The goal of semi-supervised learning algorithm is to effectively incorporate labeled and unlabeled data in a general-purpose learner with small misclassification error. Although there are various algorithms to implement semi-supervised learning task, the crucial issue of dependence of generalization error on the number of labeled and unlabeled data is still poorly understood. In this paper, we consider the Laplacian Support Vector Machines (LapSVMs) and establish its error analysis.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"31 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113973871","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 : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207478
Zhi-Heng Wang, Hong-Min Liu, Xiao Xue
In this paper, the problem of removing highlight and shading in color image is addressed, and a novel method called highlight and shading invariant color transform (HSICT) is proposed for this purpose. HSICT can be accomplished in a process of three steps: (1) Illumination color estimation is achieved by using two different color distributions; (2) A linear transform is applied to eliminate the influence of highlight; (3) The effect of shading is removed by normalization. Experiments illustrate that HSICT can not only effectively remove highlight and shading in color image, but also can be easily combined with other algorithms in many fields, such as segmentation and edge detection.
{"title":"HSICT: A method for romoving highlight and shading in color image","authors":"Zhi-Heng Wang, Hong-Min Liu, Xiao Xue","doi":"10.1109/ICWAPR.2009.5207478","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207478","url":null,"abstract":"In this paper, the problem of removing highlight and shading in color image is addressed, and a novel method called highlight and shading invariant color transform (HSICT) is proposed for this purpose. HSICT can be accomplished in a process of three steps: (1) Illumination color estimation is achieved by using two different color distributions; (2) A linear transform is applied to eliminate the influence of highlight; (3) The effect of shading is removed by normalization. Experiments illustrate that HSICT can not only effectively remove highlight and shading in color image, but also can be easily combined with other algorithms in many fields, such as segmentation and edge detection.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115137967","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 : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207411
Lin Jiang, Bin Fang, Taiping Zhang, Yuanyan Tang, Donghui Li
A novel method to extract illumination invariant features using the adaptive filter is proposed for face recognition under varying lighting conditions, the proposed method estimates illumination by minimizing the difference between the normalized illumination and estimated original illumination in the logarithm domain. To evaluate effectiveness of our method, three illumination methods (MSR, SQI and LTV) were implemented using Yale B database. It shows that the performance of the proposed method is better than other methods.
{"title":"Face recognition under varying illumination using adaptive filtering","authors":"Lin Jiang, Bin Fang, Taiping Zhang, Yuanyan Tang, Donghui Li","doi":"10.1109/ICWAPR.2009.5207411","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207411","url":null,"abstract":"A novel method to extract illumination invariant features using the adaptive filter is proposed for face recognition under varying lighting conditions, the proposed method estimates illumination by minimizing the difference between the normalized illumination and estimated original illumination in the logarithm domain. To evaluate effectiveness of our method, three illumination methods (MSR, SQI and LTV) were implemented using Yale B database. It shows that the performance of the proposed method is better than other methods.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"107 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125017026","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 : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207427
Jinfang Han, Fachao Li
In this paper, the exponential stability problem of a class of stochastic interval delayed cellular neural networks is studied. Firstly, a kind of equivalent description of this stochastic interval delayed cellular neural networks is presented. Then by using the Itô formula, Razumikhin theorems, Lyapunov function and norm inequalities, several simple sufficient conditions are obtained which guarantee the exponential stability of the stochastic interval cellular neural networks. and some recent results reported in the literature are generalized.
{"title":"Exponential stability of stochastic interval cellular neural networks with delays","authors":"Jinfang Han, Fachao Li","doi":"10.1109/ICWAPR.2009.5207427","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207427","url":null,"abstract":"In this paper, the exponential stability problem of a class of stochastic interval delayed cellular neural networks is studied. Firstly, a kind of equivalent description of this stochastic interval delayed cellular neural networks is presented. Then by using the Itô formula, Razumikhin theorems, Lyapunov function and norm inequalities, several simple sufficient conditions are obtained which guarantee the exponential stability of the stochastic interval cellular neural networks. and some recent results reported in the literature are generalized.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"525 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123903722","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 : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207429
Ying-hui Kong, Haijun Xiao
Intrusion detection is a important network security research direction. SVM(Support Vector Machine) is considered as a good substitute for traditional learning classification approach, and has a good generalization performance especially in small samples in non-linear case. LLE(Local Linear Embedding) is a good nonlinear dimensionality reduction method, which is good for the data that lies on the nonlinear manifold. This paper proposes an approach using SVM and LLE in intrusion detection system. In the Matlab simulation experiment, we can achieve higher classification accuracy rate, lower false positive rare and false negative rate using the method, compared to PCA(Principal Component Analysis) and ICA(Independent Component Analysis) approach.
{"title":"A new approach for intrusion detection based on Local Linear Embedding algorithm","authors":"Ying-hui Kong, Haijun Xiao","doi":"10.1109/ICWAPR.2009.5207429","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207429","url":null,"abstract":"Intrusion detection is a important network security research direction. SVM(Support Vector Machine) is considered as a good substitute for traditional learning classification approach, and has a good generalization performance especially in small samples in non-linear case. LLE(Local Linear Embedding) is a good nonlinear dimensionality reduction method, which is good for the data that lies on the nonlinear manifold. This paper proposes an approach using SVM and LLE in intrusion detection system. In the Matlab simulation experiment, we can achieve higher classification accuracy rate, lower false positive rare and false negative rate using the method, compared to PCA(Principal Component Analysis) and ICA(Independent Component Analysis) approach.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"3 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126117068","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 : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207458
Hui Yu, Jigui Jian
This paper is devoted to the study of multi-agent consensus with a time-varying reference state in directed networks with both switching topology and constant time delay. Stability analysis is performed based on a proposed Lyapunov-Krasovskii function. Sufficient conditions based on linear matrix inequalities (LMIs) are given to guarantee multi-agent consensus on a time-vary reference state under arbitrary switching of the network topology even if the network communication is affected by time delay. Finally, simulation examples are given to validate the theoretical results.
{"title":"Multi-agent consensus with a time-varying reference state in directed network with switching topology and time-delay","authors":"Hui Yu, Jigui Jian","doi":"10.1109/ICWAPR.2009.5207458","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207458","url":null,"abstract":"This paper is devoted to the study of multi-agent consensus with a time-varying reference state in directed networks with both switching topology and constant time delay. Stability analysis is performed based on a proposed Lyapunov-Krasovskii function. Sufficient conditions based on linear matrix inequalities (LMIs) are given to guarantee multi-agent consensus on a time-vary reference state under arbitrary switching of the network topology even if the network communication is affected by time delay. Finally, simulation examples are given to validate the theoretical results.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128008677","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 : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207421
Hui Zeng, Zhichun Mu, Li Yuan
In this paper, we propose a novel method for ear recognition using the contourlet transform. As first, we decompose the image using the contourlet transform. Then the features of the lowpass subband and the bandpass directional subbands are extracted respectively. Here we use the normalized gray-level co-occurrence matrix and the generalized Gaussian density to extract ear features. Finally, the two kinds of features are connected and the SVM method is used for classification. Extensive experiments have performed to valid its efficiency and robustness. Moreover, we can conclude that for ear feature extraction, the contourlet transform is more suitable for wavelet transform.
{"title":"Contourlet transform based EAR recognition","authors":"Hui Zeng, Zhichun Mu, Li Yuan","doi":"10.1109/ICWAPR.2009.5207421","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207421","url":null,"abstract":"In this paper, we propose a novel method for ear recognition using the contourlet transform. As first, we decompose the image using the contourlet transform. Then the features of the lowpass subband and the bandpass directional subbands are extracted respectively. Here we use the normalized gray-level co-occurrence matrix and the generalized Gaussian density to extract ear features. Finally, the two kinds of features are connected and the SVM method is used for classification. Extensive experiments have performed to valid its efficiency and robustness. Moreover, we can conclude that for ear feature extraction, the contourlet transform is more suitable for wavelet transform.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125088745","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 : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207469
Caixia Deng, Ting-ting Bai, Ying Geng
A method for image edge detection based on image fusion is presented in this paper. Since traditional wavelet transforms are unable to control the noise well and the edge is not consistent with direction properties, some improvements are made and a new kind of wavelet transform is proposed. It detects the edge of original image by means of new wavelet transform and Canny operator respectively, then produces a new image by fusing and analyzing the two results based on the experimental results. It is shown that the proposed method provides clearer and smoother edges than that using Sobel or wavelet transformation algorithms alone. This algorithm is simple, useful and easy to implement.
{"title":"Image edge detection based on wavelet transform and Canny operator","authors":"Caixia Deng, Ting-ting Bai, Ying Geng","doi":"10.1109/ICWAPR.2009.5207469","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207469","url":null,"abstract":"A method for image edge detection based on image fusion is presented in this paper. Since traditional wavelet transforms are unable to control the noise well and the edge is not consistent with direction properties, some improvements are made and a new kind of wavelet transform is proposed. It detects the edge of original image by means of new wavelet transform and Canny operator respectively, then produces a new image by fusing and analyzing the two results based on the experimental results. It is shown that the proposed method provides clearer and smoother edges than that using Sobel or wavelet transformation algorithms alone. This algorithm is simple, useful and easy to implement.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127159202","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 : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207461
Chang Liu, Song-nian Yu, Qiang Guo
Considering that data searched from the search engine is not comprehensive, and the inconsistencies between desired results and received results are inevitable, a more effective search tool called Distributed Document Clustering for Search Engine (DDCSE) is proposed in this paper. In the DDCSE, the utilizing of distributed clustering and several search engines is used to categorize the results, in order to feedback a set of better refined results. Experiments show that a significant improvement is achieved via the distribution document clustering, so as to refine the results and reduce the time used to filter out irrelevant data for the search engines.
{"title":"Distributed Document Clustering for Search Engine","authors":"Chang Liu, Song-nian Yu, Qiang Guo","doi":"10.1109/ICWAPR.2009.5207461","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207461","url":null,"abstract":"Considering that data searched from the search engine is not comprehensive, and the inconsistencies between desired results and received results are inevitable, a more effective search tool called Distributed Document Clustering for Search Engine (DDCSE) is proposed in this paper. In the DDCSE, the utilizing of distributed clustering and several search engines is used to categorize the results, in order to feedback a set of better refined results. Experiments show that a significant improvement is achieved via the distribution document clustering, so as to refine the results and reduce the time used to filter out irrelevant data for the search engines.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"108 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127946484","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 : 2009-07-12DOI: 10.1109/ICWAPR.2009.5207497
Xiang-Chun Xiao, Xiao-Ming Zeng
In this paper, we define a Α-linear bounded operator about two Bessel sequences in Hilbert C*-Module and, by means of which we make some characterzations of the properties of modular frames in Hilbert C*-module.
{"title":"Some properties of mudular frames in Hilbert C*-Modules","authors":"Xiang-Chun Xiao, Xiao-Ming Zeng","doi":"10.1109/ICWAPR.2009.5207497","DOIUrl":"https://doi.org/10.1109/ICWAPR.2009.5207497","url":null,"abstract":"In this paper, we define a Α-linear bounded operator about two Bessel sequences in Hilbert C*-Module and, by means of which we make some characterzations of the properties of modular frames in Hilbert C*-module.","PeriodicalId":424264,"journal":{"name":"2009 International Conference on Wavelet Analysis and Pattern Recognition","volume":"07 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127411749","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：