Pub Date : 2011-07-14DOI: 10.1109/CSAE.2011.5952550
Hongqi Han, X. An, Donghua Zhu, Xuefeng Wang
Visualization technique is a powerful method used by science and technology intelligence analysis experts to identify technical competitor groups. Common visualization methods tend to create graphs meeting the aesthetic criteria instead of finding better clusters, and their analysis results may provide misleading information. A process model of technical group identification method was presented using LinLog graph clustering algorithm to find better competitor groups. In the model, technical similarity value of each pair of competitors is measured based on their R&D output in sub-fields, and two competitors have a link when they have high similarity value; LinLog algorithm, which is aimed at producing better clusters, was employed to layout graph with competitors as nodes, their links as edges and technology similarity values as weights of edges. Experiment results show the efficiency of presented method.
{"title":"Visual group identification method of technical competitors using LinLog graph clustering algorithm","authors":"Hongqi Han, X. An, Donghua Zhu, Xuefeng Wang","doi":"10.1109/CSAE.2011.5952550","DOIUrl":"https://doi.org/10.1109/CSAE.2011.5952550","url":null,"abstract":"Visualization technique is a powerful method used by science and technology intelligence analysis experts to identify technical competitor groups. Common visualization methods tend to create graphs meeting the aesthetic criteria instead of finding better clusters, and their analysis results may provide misleading information. A process model of technical group identification method was presented using LinLog graph clustering algorithm to find better competitor groups. In the model, technical similarity value of each pair of competitors is measured based on their R&D output in sub-fields, and two competitors have a link when they have high similarity value; LinLog algorithm, which is aimed at producing better clusters, was employed to layout graph with competitors as nodes, their links as edges and technology similarity values as weights of edges. Experiment results show the efficiency of presented method.","PeriodicalId":138215,"journal":{"name":"2011 IEEE International Conference on Computer Science and Automation Engineering","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116850912","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 : 2011-07-14DOI: 10.1109/CSAE.2011.5953307
Yang Zhi, Guo-en Xia, W. Jin
Class imbalance is one of the main obstacles in data mining. AUC is one of the main criterions to judge the performance of classifiers, which have been applied in class imbalanced datasets. So, optimizing AUC method has been realized by using gradient method to optimize it directly. But optimizing AUC method limits the shortcoming of gradient method, which is generally converged in local minima. So, this paper introduced the genetic algorithm into optimizing AUC method, and compared it with the previous one. The results of the experiment proving the method in this paper is more suitable for imbalanced datasets than the previous one.
{"title":"Optimizing area under the Roc curve using genetic algorithm","authors":"Yang Zhi, Guo-en Xia, W. Jin","doi":"10.1109/CSAE.2011.5953307","DOIUrl":"https://doi.org/10.1109/CSAE.2011.5953307","url":null,"abstract":"Class imbalance is one of the main obstacles in data mining. AUC is one of the main criterions to judge the performance of classifiers, which have been applied in class imbalanced datasets. So, optimizing AUC method has been realized by using gradient method to optimize it directly. But optimizing AUC method limits the shortcoming of gradient method, which is generally converged in local minima. So, this paper introduced the genetic algorithm into optimizing AUC method, and compared it with the previous one. The results of the experiment proving the method in this paper is more suitable for imbalanced datasets than the previous one.","PeriodicalId":138215,"journal":{"name":"2011 IEEE International Conference on Computer Science and Automation Engineering","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129056236","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 : 2011-07-14DOI: 10.1109/CSAE.2011.5953291
Peizhe Zhang, Canping Li
Color image segmentation of fishes with complex back ground in water is considered. Unsupervised color equalization in HSV space is used to enhance and correct nonuniform color cast of image underwater. Region-based segmentation algorithm is presented for the aim of accurately and fast identification of fishes. The segmentation operation is applied in the selected region, which can also avoid a large number of computations without considering information of non interested regions. For comparison, we test segmentation algorithm combined with automatic color equalization on different region, including segmentation of single fish in a small area and more than one fishes in an interested region at the same time. Experiments show the result is encouraging and efficient.
{"title":"Region-based color image segmentation of fishes with complex background in water","authors":"Peizhe Zhang, Canping Li","doi":"10.1109/CSAE.2011.5953291","DOIUrl":"https://doi.org/10.1109/CSAE.2011.5953291","url":null,"abstract":"Color image segmentation of fishes with complex back ground in water is considered. Unsupervised color equalization in HSV space is used to enhance and correct nonuniform color cast of image underwater. Region-based segmentation algorithm is presented for the aim of accurately and fast identification of fishes. The segmentation operation is applied in the selected region, which can also avoid a large number of computations without considering information of non interested regions. For comparison, we test segmentation algorithm combined with automatic color equalization on different region, including segmentation of single fish in a small area and more than one fishes in an interested region at the same time. Experiments show the result is encouraging and efficient.","PeriodicalId":138215,"journal":{"name":"2011 IEEE International Conference on Computer Science and Automation Engineering","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130208158","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 : 2011-07-14DOI: 10.1109/CSAE.2011.5952411
Wenjie Du, Yuxin Deng
Bisimilarity is one of the most important relations for comparing the behaviour of formal systems in concurrency theory. Decision algorithms for bisimilarity in finite state systems are usually classified into two kinds: global algorithms are generally efficient but require to generate the whole state spaces in advance, local algorithms combine the verification of a system's behaviour with the generation of the system's state space, which is often more effective to determine that one system fails to be related to another. In this paper we propose a quasi-local algorithm with worst case time complexity O(m1m2), where m1 and m2 are the numbers of transitions in two labelled transition systems. With mild modifications, the algorithm can be easily adapted to decide similarity with the same time complexity. For deterministic systems, the algorithm can be simplified and runs in time O(min(m1,m2)).
{"title":"A quasi-local algorithm for checking bisimilarity","authors":"Wenjie Du, Yuxin Deng","doi":"10.1109/CSAE.2011.5952411","DOIUrl":"https://doi.org/10.1109/CSAE.2011.5952411","url":null,"abstract":"Bisimilarity is one of the most important relations for comparing the behaviour of formal systems in concurrency theory. Decision algorithms for bisimilarity in finite state systems are usually classified into two kinds: global algorithms are generally efficient but require to generate the whole state spaces in advance, local algorithms combine the verification of a system's behaviour with the generation of the system's state space, which is often more effective to determine that one system fails to be related to another. In this paper we propose a quasi-local algorithm with worst case time complexity O(m1m2), where m1 and m2 are the numbers of transitions in two labelled transition systems. With mild modifications, the algorithm can be easily adapted to decide similarity with the same time complexity. For deterministic systems, the algorithm can be simplified and runs in time O(min(m1,m2)).","PeriodicalId":138215,"journal":{"name":"2011 IEEE International Conference on Computer Science and Automation Engineering","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133367234","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 : 2011-07-14DOI: 10.1109/CSAE.2011.5952535
Jun Li, Yanhui Li, Shuangping Chen
Profile Hidden Markov Models are used as a popular tool in bioinformatics research and a regular task is to compare a set of protein sequences with a database of models according to sequences' score on these models. However, it suffers from long runtimes on PC platforms, and the runtimes are increasing rapidly due to the rapid growth in size of both sequences and models. In this paper, we present a Viterbi algorithm running on graphic processing units to score sequences, a method padding HMMs and a memory hierarchy are also introduced, these strategies can promote running efficiency in parallel and reduce impact of the bottleneck from buses. Experimental results show the runtimes are saved by the method dramatically.
{"title":"The fast Viterbi algorithm caching Profile Hidden Markov Models on graphic processing units","authors":"Jun Li, Yanhui Li, Shuangping Chen","doi":"10.1109/CSAE.2011.5952535","DOIUrl":"https://doi.org/10.1109/CSAE.2011.5952535","url":null,"abstract":"Profile Hidden Markov Models are used as a popular tool in bioinformatics research and a regular task is to compare a set of protein sequences with a database of models according to sequences' score on these models. However, it suffers from long runtimes on PC platforms, and the runtimes are increasing rapidly due to the rapid growth in size of both sequences and models. In this paper, we present a Viterbi algorithm running on graphic processing units to score sequences, a method padding HMMs and a memory hierarchy are also introduced, these strategies can promote running efficiency in parallel and reduce impact of the bottleneck from buses. Experimental results show the runtimes are saved by the method dramatically.","PeriodicalId":138215,"journal":{"name":"2011 IEEE International Conference on Computer Science and Automation Engineering","volume":"98 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127093900","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 : 2011-07-14DOI: 10.1109/CSAE.2011.5953282
Bao Xiaomin, Ni Xiaoqing, Wang Yaming, Zhu Hanyu
3D textile dynamic simulation aims to realize the simulation on fabric in the real environment, which has great significance in the field of virtual reality and computer animation. The modeling methods of the 3D textile are the key technologies and are the first step in the 3D simulation. Based on the model, collision detection of the fabric is needed in order to realize the dynamic, real-time simulation. Meanwhile, to make the simulated texture more true, it is demanded to consider the illumination model, and texture mapping of the fabric. The paper has separately introduced three kinds of modeling methods, two collision detection methods, two typical illumination model, and three texture mapping methods. Finally, the trends of future research are discussed.
{"title":"Overview of 3D textile dynamic simulation research","authors":"Bao Xiaomin, Ni Xiaoqing, Wang Yaming, Zhu Hanyu","doi":"10.1109/CSAE.2011.5953282","DOIUrl":"https://doi.org/10.1109/CSAE.2011.5953282","url":null,"abstract":"3D textile dynamic simulation aims to realize the simulation on fabric in the real environment, which has great significance in the field of virtual reality and computer animation. The modeling methods of the 3D textile are the key technologies and are the first step in the 3D simulation. Based on the model, collision detection of the fabric is needed in order to realize the dynamic, real-time simulation. Meanwhile, to make the simulated texture more true, it is demanded to consider the illumination model, and texture mapping of the fabric. The paper has separately introduced three kinds of modeling methods, two collision detection methods, two typical illumination model, and three texture mapping methods. Finally, the trends of future research are discussed.","PeriodicalId":138215,"journal":{"name":"2011 IEEE International Conference on Computer Science and Automation Engineering","volume":"221 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123291571","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 : 2011-07-14DOI: 10.1109/CSAE.2011.5952605
Wei-jun Zou, Ming-feng Ying, Bo Yu-ming
In this paper, we propose a particle filter algorithm with adaptive layered-optimization and multi-feature, which is used for motion-based tracking of natural object. A novel reliability measure based on the particle's distribution in the state space is designed to evaluate the tracking quality. According to the tracking quality, the particle set is divided into two parts: one is optimized to be concentrative for the tracking precision and the other keeps being original for the tracking robustness. The number of particles in each part is decided adaptively by the function which uses reliability score as parameter. This algorithm is demonstrated using the color and orientation features weighted by reliability score. Experiments over a set of real-world video sequences are done and the result shows that this algorithm achieves better performance when occlusion and object-motion in variable direction happen; the consuming time meets the requirement of real-time.
{"title":"Visual Tracking with adaptive layered-optimizing particles in Multifeature Particle Filtering Framework","authors":"Wei-jun Zou, Ming-feng Ying, Bo Yu-ming","doi":"10.1109/CSAE.2011.5952605","DOIUrl":"https://doi.org/10.1109/CSAE.2011.5952605","url":null,"abstract":"In this paper, we propose a particle filter algorithm with adaptive layered-optimization and multi-feature, which is used for motion-based tracking of natural object. A novel reliability measure based on the particle's distribution in the state space is designed to evaluate the tracking quality. According to the tracking quality, the particle set is divided into two parts: one is optimized to be concentrative for the tracking precision and the other keeps being original for the tracking robustness. The number of particles in each part is decided adaptively by the function which uses reliability score as parameter. This algorithm is demonstrated using the color and orientation features weighted by reliability score. Experiments over a set of real-world video sequences are done and the result shows that this algorithm achieves better performance when occlusion and object-motion in variable direction happen; the consuming time meets the requirement of real-time.","PeriodicalId":138215,"journal":{"name":"2011 IEEE International Conference on Computer Science and Automation Engineering","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126548093","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 : 2011-07-14DOI: 10.1109/CSAE.2011.5952698
Wenhua Ma
Anisotropic diffusion is classified by the eigenvalue of the Hessian matrix associated with the diffusivity function into two categories: one incapable of edge-sharpening and the other capable of selective edge sharpening. A third class is proposed: the eigenvalue starts with a small value and decreases monotonically with image gradient magnitude so that the stronger the edge is, the more it is sharpened. Two such examples are given and one is found to consistently produce the best PSNR at all simulated noise levels.
{"title":"Monotonically decreasing eigenvalue for edge-sharpening diffusion","authors":"Wenhua Ma","doi":"10.1109/CSAE.2011.5952698","DOIUrl":"https://doi.org/10.1109/CSAE.2011.5952698","url":null,"abstract":"Anisotropic diffusion is classified by the eigenvalue of the Hessian matrix associated with the diffusivity function into two categories: one incapable of edge-sharpening and the other capable of selective edge sharpening. A third class is proposed: the eigenvalue starts with a small value and decreases monotonically with image gradient magnitude so that the stronger the edge is, the more it is sharpened. Two such examples are given and one is found to consistently produce the best PSNR at all simulated noise levels.","PeriodicalId":138215,"journal":{"name":"2011 IEEE International Conference on Computer Science and Automation Engineering","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125553491","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 : 2011-06-10DOI: 10.1109/CSAE.2011.5952938
Ke Chen, Zhiyong Xiong, X. Xian, Fusheng Yu
An image retrieval approach combined with relevance feedback is proposed. A set of blobs that are generated from image features using clustering can be used to describe an image. Given a training set of images with annotations, we apply probabilistic models to predict the probability of a blob in image according to the query words. For improving the initial query results, we apply a relevance feedback mechanism to bridge the semantic gap, leading to the improved image retrieval accuracy. A support vector machine classifier can be learned from training data of relevance images and irrelevance images labeled by users. Experimental results show that the proposed approach obtains higher retrieval accuracy than a commonly used approach.
{"title":"An image retrieval approach with relevance feedback","authors":"Ke Chen, Zhiyong Xiong, X. Xian, Fusheng Yu","doi":"10.1109/CSAE.2011.5952938","DOIUrl":"https://doi.org/10.1109/CSAE.2011.5952938","url":null,"abstract":"An image retrieval approach combined with relevance feedback is proposed. A set of blobs that are generated from image features using clustering can be used to describe an image. Given a training set of images with annotations, we apply probabilistic models to predict the probability of a blob in image according to the query words. For improving the initial query results, we apply a relevance feedback mechanism to bridge the semantic gap, leading to the improved image retrieval accuracy. A support vector machine classifier can be learned from training data of relevance images and irrelevance images labeled by users. Experimental results show that the proposed approach obtains higher retrieval accuracy than a commonly used approach.","PeriodicalId":138215,"journal":{"name":"2011 IEEE International Conference on Computer Science and Automation Engineering","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114995986","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 : 2011-06-10DOI: 10.1109/CSAE.2011.5953294
Xiaogang Wang, S. Hua
An extended state feedback controller for current control of paralleled three-phase three-leg active power filter is presented. The primary goal is to eliminate the impacts of time delay caused by sampling and calculation. The controller is developed in synchronous rotating frame, so the decoupled discrete-time model is deduced at first. Then, a proportional feedback and an integral feedback constitute an extended state feedback controller, the impacts of time delay is analyzed. Moreover, in order to remove these effects, a state observer is introduced. Finally, the simulation results show that the APF based on the proposed controller has good steady and transient performance.
{"title":"Extended state feedback control for active power filter","authors":"Xiaogang Wang, S. Hua","doi":"10.1109/CSAE.2011.5953294","DOIUrl":"https://doi.org/10.1109/CSAE.2011.5953294","url":null,"abstract":"An extended state feedback controller for current control of paralleled three-phase three-leg active power filter is presented. The primary goal is to eliminate the impacts of time delay caused by sampling and calculation. The controller is developed in synchronous rotating frame, so the decoupled discrete-time model is deduced at first. Then, a proportional feedback and an integral feedback constitute an extended state feedback controller, the impacts of time delay is analyzed. Moreover, in order to remove these effects, a state observer is introduced. Finally, the simulation results show that the APF based on the proposed controller has good steady and transient performance.","PeriodicalId":138215,"journal":{"name":"2011 IEEE International Conference on Computer Science and Automation Engineering","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115134558","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：