Twin support vector machine is a novel classifier, it construct two nonparallel hyper planes instead of a single hyper plane to obtain four times faster than the usual SVM. With the result of traditional incremental learning method of SVM, we analyze the characteristics of twin support vector machine and the distribution of the training sample set. In this paper, we propose a fast incremental learning algorithm based on twin support vector machine. It can deal with the newly added training samples and utilize the result of the previous training effectively. Experimental results prove that the given algorithm has excellent classification performance on runtime and recognition rate, and therefore confirm the above conclusion further.
{"title":"A Fast Incremental Learning Algorithm Based on Twin Support Vector Machine","authors":"Yunhe Hao, Haofeng Zhang","doi":"10.1109/ISCID.2014.38","DOIUrl":"https://doi.org/10.1109/ISCID.2014.38","url":null,"abstract":"Twin support vector machine is a novel classifier, it construct two nonparallel hyper planes instead of a single hyper plane to obtain four times faster than the usual SVM. With the result of traditional incremental learning method of SVM, we analyze the characteristics of twin support vector machine and the distribution of the training sample set. In this paper, we propose a fast incremental learning algorithm based on twin support vector machine. It can deal with the newly added training samples and utilize the result of the previous training effectively. Experimental results prove that the given algorithm has excellent classification performance on runtime and recognition rate, and therefore confirm the above conclusion further.","PeriodicalId":385391,"journal":{"name":"2014 Seventh International Symposium on Computational Intelligence and Design","volume":"134 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133566343","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}
Assuming that one sentence in review expresses one opinion, LDA based on sentence is performed to analysis the massive online reviews. When computing the topic's word, word relevance measure is designed which penalizes the word frequency by a factor that captures how much the word is shared across topics, words for topics can been selected more accurately. Experiments on massive review crawled from network show that the result of analyzing is better than the standard LDA, there is clearer topic cue, and recognition is improved among the topics.
{"title":"Reviews Analysis Based on Sentence and Word Relevance","authors":"Shibo Zhang, Yun Sha, Xiaojie Wang","doi":"10.1109/ISCID.2014.21","DOIUrl":"https://doi.org/10.1109/ISCID.2014.21","url":null,"abstract":"Assuming that one sentence in review expresses one opinion, LDA based on sentence is performed to analysis the massive online reviews. When computing the topic's word, word relevance measure is designed which penalizes the word frequency by a factor that captures how much the word is shared across topics, words for topics can been selected more accurately. Experiments on massive review crawled from network show that the result of analyzing is better than the standard LDA, there is clearer topic cue, and recognition is improved among the topics.","PeriodicalId":385391,"journal":{"name":"2014 Seventh International Symposium on Computational Intelligence and Design","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127844514","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}
Content-based image classification is such a technique which adapt to mass image data access and classification operation and it is based on the color, texture and shape feature. Image automatic classification using computer is one of the current hot topic. The traditional image classification method based on a single feature is ineffective. In this paper, we use multi-kernel SVM classifiers and the multi-feature fusion of feature weighting for image classification. Feature weighting is to set a certain weight for various features according to a certain standards and it is an effective way to find the most effective features. We use Corel Image Library as the database. The experimental result shows that the accuracy of image classification based on multi-feature fusion with multi-kernel SVM is much higher than a single feature. The method in this paper is an effective approach to improve the accuracy of image classification and expand possibilities for other application.
{"title":"An Image Classification Method Based on Multi-feature Fusion and Multi-kernel SVM","authors":"Zixi Xiang, Xueqiang Lv, Kai Zhang","doi":"10.1109/ISCID.2014.25","DOIUrl":"https://doi.org/10.1109/ISCID.2014.25","url":null,"abstract":"Content-based image classification is such a technique which adapt to mass image data access and classification operation and it is based on the color, texture and shape feature. Image automatic classification using computer is one of the current hot topic. The traditional image classification method based on a single feature is ineffective. In this paper, we use multi-kernel SVM classifiers and the multi-feature fusion of feature weighting for image classification. Feature weighting is to set a certain weight for various features according to a certain standards and it is an effective way to find the most effective features. We use Corel Image Library as the database. The experimental result shows that the accuracy of image classification based on multi-feature fusion with multi-kernel SVM is much higher than a single feature. The method in this paper is an effective approach to improve the accuracy of image classification and expand possibilities for other application.","PeriodicalId":385391,"journal":{"name":"2014 Seventh International Symposium on Computational Intelligence and Design","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115234556","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}
Kang Feng, Zhang Hua-xiong, Hu Jie, Zhang Cheng, Zhou Hui
Extracting patterns in fabric images usually should preserve the color of yarns and pattern structures, and suppress textures caused by adjacent yarns. A novel filter is proposed for smoothing fabric images while suppressing textures and preserving fabric pattern structures, by means of a nonlinear combination of local color values and lightness gradients underlying CIE-Lab color space. The proposed filter smooths color values and preserves pattern structures based on photometric similarity of neighboring pixels of the fabric images, and suppresses textures based on local gradient difference between the textures of yarns and the patterns of fabric images. It takes a non-iterative, local, and simple scheme similar as bilateral filter. Compared with popular smoothing filters, the experiments demonstrate that the proposed filter produces satisfactory smoothing images with the properties of texture-suppressing and pattern-preserving.
{"title":"A Novel Smoothing Filter with Texture Suppression for Fabric Images","authors":"Kang Feng, Zhang Hua-xiong, Hu Jie, Zhang Cheng, Zhou Hui","doi":"10.1109/ISCID.2014.200","DOIUrl":"https://doi.org/10.1109/ISCID.2014.200","url":null,"abstract":"Extracting patterns in fabric images usually should preserve the color of yarns and pattern structures, and suppress textures caused by adjacent yarns. A novel filter is proposed for smoothing fabric images while suppressing textures and preserving fabric pattern structures, by means of a nonlinear combination of local color values and lightness gradients underlying CIE-Lab color space. The proposed filter smooths color values and preserves pattern structures based on photometric similarity of neighboring pixels of the fabric images, and suppresses textures based on local gradient difference between the textures of yarns and the patterns of fabric images. It takes a non-iterative, local, and simple scheme similar as bilateral filter. Compared with popular smoothing filters, the experiments demonstrate that the proposed filter produces satisfactory smoothing images with the properties of texture-suppressing and pattern-preserving.","PeriodicalId":385391,"journal":{"name":"2014 Seventh International Symposium on Computational Intelligence and Design","volume":"124 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115614091","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 fault-tolerant design of analog circuit is a very meaning thing. Negative-correlation redundant fault-tolerant design is a new approach. This paper proposes a method which uses single population to evolve negative-correlation circuit. Each chromosome represents only one circuit and evolves one circuit that meets requirements each evolution time. Then use the evolved circuits to guide the subsequent evolution process. According to the information of correlation in population, we change the evolved circuits dynamically. Meanwhile in order to further improve the efficiency of the algorithm, increasing diversity of population, we add variable structure operation and self-adaptive operation of genetic operator. When all the negative-correlation circuits are evolved, the algorithm terminates and outputs these circuits. Analog low-pass filter experiment shows that this method can evolve negative-correlation circuits which meet our requirements successfully. Compared with the conventional methods, it can shorten evolution time and improve the evolution efficiency.
{"title":"An Efficient Differential Evoluiton Method Used for Evolving Negative-Correlation Circuits","authors":"Zaisheng Huang, Jingsong He","doi":"10.1109/ISCID.2014.172","DOIUrl":"https://doi.org/10.1109/ISCID.2014.172","url":null,"abstract":"The fault-tolerant design of analog circuit is a very meaning thing. Negative-correlation redundant fault-tolerant design is a new approach. This paper proposes a method which uses single population to evolve negative-correlation circuit. Each chromosome represents only one circuit and evolves one circuit that meets requirements each evolution time. Then use the evolved circuits to guide the subsequent evolution process. According to the information of correlation in population, we change the evolved circuits dynamically. Meanwhile in order to further improve the efficiency of the algorithm, increasing diversity of population, we add variable structure operation and self-adaptive operation of genetic operator. When all the negative-correlation circuits are evolved, the algorithm terminates and outputs these circuits. Analog low-pass filter experiment shows that this method can evolve negative-correlation circuits which meet our requirements successfully. Compared with the conventional methods, it can shorten evolution time and improve the evolution efficiency.","PeriodicalId":385391,"journal":{"name":"2014 Seventh International Symposium on Computational Intelligence and Design","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115661087","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}
According to the requirement of grape pruning in winter, an algorithm of detecting the buds of grape vines based on machine vision was proposed, laying the foundation of grape vines automotive pruning. The grapevines' color image was captured indoor. The blue component of the color image was selected for the image preprocessing such as filter, threshold segmentation and noise removal. After that a binary image was gained. With the binary image, Rosenfeld algorithm was used in thinning to extract the skeleton of the grape branches. Because the morphological characteristic of the buds was similar to the corners, Harris algorithm was chosen to detect the point of buds from the skeleton image. The experiment result showed that it's effective to detect the buds with the strategy of this paper. The recognition rate reached 70.2%.
{"title":"Detection Method for the Buds on Winter Vines Based on Computer Vision","authors":"Sheng Xu, Y. Xun, Tingmeng Jia, Qinghua Yang","doi":"10.1109/ISCID.2014.26","DOIUrl":"https://doi.org/10.1109/ISCID.2014.26","url":null,"abstract":"According to the requirement of grape pruning in winter, an algorithm of detecting the buds of grape vines based on machine vision was proposed, laying the foundation of grape vines automotive pruning. The grapevines' color image was captured indoor. The blue component of the color image was selected for the image preprocessing such as filter, threshold segmentation and noise removal. After that a binary image was gained. With the binary image, Rosenfeld algorithm was used in thinning to extract the skeleton of the grape branches. Because the morphological characteristic of the buds was similar to the corners, Harris algorithm was chosen to detect the point of buds from the skeleton image. The experiment result showed that it's effective to detect the buds with the strategy of this paper. The recognition rate reached 70.2%.","PeriodicalId":385391,"journal":{"name":"2014 Seventh International Symposium on Computational Intelligence and Design","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115782566","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}
Terrain is the cause of formation of spatial distribution pattern of land use. Based on the land use data and the digital elevation model (DEM), the study analyzed the relationship between spatial distribution pattern of land use and landform factors - elevation, slope and aspect in Hang Zhou. A concept of elevation-slope united distribution index is proposed. The result shows that the terrain had great impact on spatial distribution pattern of land use. The advantageous land use type on each level of terrains were extracted according to the number and the structure of the land use types. It is founded that some distribution features induces the corresponding land use change pattern.
{"title":"Land Use and Landform Impact Factors Co-occurrence Matrix Interpretation","authors":"Y. Liu, Yan Li","doi":"10.1109/ISCID.2014.190","DOIUrl":"https://doi.org/10.1109/ISCID.2014.190","url":null,"abstract":"Terrain is the cause of formation of spatial distribution pattern of land use. Based on the land use data and the digital elevation model (DEM), the study analyzed the relationship between spatial distribution pattern of land use and landform factors - elevation, slope and aspect in Hang Zhou. A concept of elevation-slope united distribution index is proposed. The result shows that the terrain had great impact on spatial distribution pattern of land use. The advantageous land use type on each level of terrains were extracted according to the number and the structure of the land use types. It is founded that some distribution features induces the corresponding land use change pattern.","PeriodicalId":385391,"journal":{"name":"2014 Seventh International Symposium on Computational Intelligence and Design","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115887396","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}
This paper proposes an approach of detecting information leakage based on subtracting matrix to determine files when and where were leaked from file system. Due to the low efficiency of detecting leak operation and the rapid growth of the size of storage device make it difficult to locate the place where the leakage occurred. We build a time matrix model by file system access timestamps in a suspicious information system. Then three kinds of two-value (0-1) matrices are generated based on the similarity of access timestamps in the time matrix. The behavior of information leakage can be finally determined by comparing the degree of the similarity in these matrices. The experimental results show the method can detect information leakage more quickly and accurately.
{"title":"Detecting Information Leakage Based on Subtracting Matrix","authors":"Zongda Han, Binglong Li","doi":"10.1109/ISCID.2014.249","DOIUrl":"https://doi.org/10.1109/ISCID.2014.249","url":null,"abstract":"This paper proposes an approach of detecting information leakage based on subtracting matrix to determine files when and where were leaked from file system. Due to the low efficiency of detecting leak operation and the rapid growth of the size of storage device make it difficult to locate the place where the leakage occurred. We build a time matrix model by file system access timestamps in a suspicious information system. Then three kinds of two-value (0-1) matrices are generated based on the similarity of access timestamps in the time matrix. The behavior of information leakage can be finally determined by comparing the degree of the similarity in these matrices. The experimental results show the method can detect information leakage more quickly and accurately.","PeriodicalId":385391,"journal":{"name":"2014 Seventh International Symposium on Computational Intelligence and Design","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124136899","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}
Aircraft Landing Scheduling (ALS) plays an important role in ensuring safety and efficiency of flight operation in terminal area in the air traffic management domain. Since ALS problem is multi-constraint and large-scale, existing algorithms can hardly deal with it efficiently. This paper firstly introduces the mathematic model of ALS problem and the 1D Cellular-Automaton model for simulating aircraft landing process. Then, based on above two models and a practice terminal area scenario, an improved 2D-Cellular-Automaton-based algorithm is presented for solving real-time ALS problem. Compared with previous approaches for ALS, the improved algorithm puts emphasis on not only flight speed, but also flight actual landing route, which can more exactly simulate practical aircraft landing process and provide feasible aircraft landing sequence. Empirical studies, using the scenario of Chengdu Shuangliu Airport and corresponding records of flight tracks, show that our proposed algorithm outperforms existing ones.
{"title":"An Improved Cellular-Automaton-Based Algorithm for Real-Time Aircraft Landing Scheduling","authors":"Yueshuai He, Kaiquan Cai, Yongliang Li, Mingming Xiao","doi":"10.1109/ISCID.2014.243","DOIUrl":"https://doi.org/10.1109/ISCID.2014.243","url":null,"abstract":"Aircraft Landing Scheduling (ALS) plays an important role in ensuring safety and efficiency of flight operation in terminal area in the air traffic management domain. Since ALS problem is multi-constraint and large-scale, existing algorithms can hardly deal with it efficiently. This paper firstly introduces the mathematic model of ALS problem and the 1D Cellular-Automaton model for simulating aircraft landing process. Then, based on above two models and a practice terminal area scenario, an improved 2D-Cellular-Automaton-based algorithm is presented for solving real-time ALS problem. Compared with previous approaches for ALS, the improved algorithm puts emphasis on not only flight speed, but also flight actual landing route, which can more exactly simulate practical aircraft landing process and provide feasible aircraft landing sequence. Empirical studies, using the scenario of Chengdu Shuangliu Airport and corresponding records of flight tracks, show that our proposed algorithm outperforms existing ones.","PeriodicalId":385391,"journal":{"name":"2014 Seventh International Symposium on Computational Intelligence and Design","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124543352","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}
Reasonable and right decisions are the keys to the successful financing guarantee project, and the core of decision-making is the correct evaluation. The improved spectral clustering algorithm is used to build the financing guarantee project evaluation model, which can avoid the set of scale factor, and reduce the computational complexity of matrix eigenvalue decomposition. The financing guarantee project evaluation model is established by MATLAB software, and the effectiveness and high efficiency of CMSC can be verified through the trainings and simulation experiments.
{"title":"Project Evaluation of Financial Guarantee Based on Improved Spectral Clustering","authors":"Weiquan Sang, Xiaoping Zhang, Hui Li","doi":"10.1109/ISCID.2014.224","DOIUrl":"https://doi.org/10.1109/ISCID.2014.224","url":null,"abstract":"Reasonable and right decisions are the keys to the successful financing guarantee project, and the core of decision-making is the correct evaluation. The improved spectral clustering algorithm is used to build the financing guarantee project evaluation model, which can avoid the set of scale factor, and reduce the computational complexity of matrix eigenvalue decomposition. The financing guarantee project evaluation model is established by MATLAB software, and the effectiveness and high efficiency of CMSC can be verified through the trainings and simulation experiments.","PeriodicalId":385391,"journal":{"name":"2014 Seventh International Symposium on Computational Intelligence and Design","volume":"188 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117304136","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：