场景自适应跟踪特征选择研究

Q4 Physics and Astronomy

光学技术 Pub Date : 2014-01-01 DOI:10.3788/GXJS20144006.0551

付毅 Fu Yi, 吴泽民 Wu Zemin, 田畅 Tian Chang, 曾明勇 Zeng Mingyong, 揭斐然 Jie Feiran

{"title":"场景自适应跟踪特征选择研究","authors":"付毅 Fu Yi, 吴泽民 Wu Zemin, 田畅 Tian Chang, 曾明勇 Zeng Mingyong, 揭斐然 Jie Feiran","doi":"10.3788/GXJS20144006.0551","DOIUrl":null,"url":null,"abstract":"Video target tracking is a key issue in the field of computer vision.Based on the scene classification an online method for video target feature selection is proposed.During off-line,the modified spatial pyramid matching algorithm(NEW-SPM)is used to complete the sample-videos classification at first.And then the rough localization of target and background region is realized with TLD tracking algorithm,and the characteristics description the target and background is completed with different descriptors.After this,the feature distance between target and background is calculated through the form of log variance ratio.And finally,the fusion mean and entropy statistical analysis is carried out on the characteristics of distance for the saliency sorting of different descriptors.On this basis,the scenarios related characteristics saliency lists are constructed.Combining with the testing scenes judgment,the lists are applied to the on-line particle filter tracking system to select the best descriptor in the current category,and it is tested on the public library of tracking,the validity of the lists and the necessity of selection method are proven.","PeriodicalId":35591,"journal":{"name":"光学技术","volume":"40 1","pages":"551-559"},"PeriodicalIF":0.0000,"publicationDate":"2014-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Research on scene adaptive tracking feature selection\",\"authors\":\"付毅 Fu Yi, 吴泽民 Wu Zemin, 田畅 Tian Chang, 曾明勇 Zeng Mingyong, 揭斐然 Jie Feiran\",\"doi\":\"10.3788/GXJS20144006.0551\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Video target tracking is a key issue in the field of computer vision.Based on the scene classification an online method for video target feature selection is proposed.During off-line,the modified spatial pyramid matching algorithm(NEW-SPM)is used to complete the sample-videos classification at first.And then the rough localization of target and background region is realized with TLD tracking algorithm,and the characteristics description the target and background is completed with different descriptors.After this,the feature distance between target and background is calculated through the form of log variance ratio.And finally,the fusion mean and entropy statistical analysis is carried out on the characteristics of distance for the saliency sorting of different descriptors.On this basis,the scenarios related characteristics saliency lists are constructed.Combining with the testing scenes judgment,the lists are applied to the on-line particle filter tracking system to select the best descriptor in the current category,and it is tested on the public library of tracking,the validity of the lists and the necessity of selection method are proven.\",\"PeriodicalId\":35591,\"journal\":{\"name\":\"光学技术\",\"volume\":\"40 1\",\"pages\":\"551-559\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2014-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"光学技术\",\"FirstCategoryId\":\"1089\",\"ListUrlMain\":\"https://doi.org/10.3788/GXJS20144006.0551\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"Physics and Astronomy\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"光学技术","FirstCategoryId":"1089","ListUrlMain":"https://doi.org/10.3788/GXJS20144006.0551","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"Physics and Astronomy","Score":null,"Total":0}

引用次数: 0

摘要

视频目标跟踪是计算机视觉领域的一个关键问题。提出了一种基于场景分类的视频目标特征在线选择方法。离线时，首先使用改进的空间金字塔匹配算法(NEW-SPM)完成样本视频分类。然后利用TLD跟踪算法实现目标和背景区域的粗略定位，并用不同的描述符完成目标和背景的特征描述。之后，通过对数方差比的形式计算目标与背景之间的特征距离。最后，对距离特征进行融合均值和熵统计分析，对不同描述符进行显著性排序。在此基础上，构建了场景相关特征显著性列表。结合测试场景判断，将列表应用于在线粒子滤波跟踪系统中，在当前类别中选择最佳描述子，并在公共跟踪库上进行了测试，验证了列表的有效性和选择方法的必要性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Research on scene adaptive tracking feature selection

Video target tracking is a key issue in the field of computer vision.Based on the scene classification an online method for video target feature selection is proposed.During off-line,the modified spatial pyramid matching algorithm(NEW-SPM)is used to complete the sample-videos classification at first.And then the rough localization of target and background region is realized with TLD tracking algorithm,and the characteristics description the target and background is completed with different descriptors.After this,the feature distance between target and background is calculated through the form of log variance ratio.And finally,the fusion mean and entropy statistical analysis is carried out on the characteristics of distance for the saliency sorting of different descriptors.On this basis,the scenarios related characteristics saliency lists are constructed.Combining with the testing scenes judgment,the lists are applied to the on-line particle filter tracking system to select the best descriptor in the current category,and it is tested on the public library of tracking,the validity of the lists and the necessity of selection method are proven.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

光学技术 Physics and Astronomy-Atomic and Molecular Physics, and Optics

CiteScore

0.60

自引率

0.00%

发文量

6699

期刊介绍： The predecessor of Optical Technology was Optical Technology, which was founded in 1975. At that time, the Fifth Ministry of Machine Building entrusted the School of Optoelectronics of Beijing Institute of Technology to publish the journal, and it was officially approved by the State Administration of Press, Publication, Radio, Film and Television for external distribution. From 1975 to 1979, the magazine was named Optical Technology, a quarterly with 4 issues per year; from 1980 to the present, the magazine is named Optical Technology, a bimonthly with 6 issues per year, published on the 20th of odd months. The publication policy is: to serve the national economic construction, implement the development of the national economy, serve production and scientific research, and implement the publication policy of "letting a hundred flowers bloom and a hundred schools of thought contend".