首页 | 本学科首页   官方微博 | 高级检索  
     

仿射变换在压缩感知跟踪中的应用
引用本文:黄宏图,葛渊,张继,查宇飞,毕笃彦,侯志强. 仿射变换在压缩感知跟踪中的应用[J]. 西安电子科技大学学报(自然科学版), 2015, 42(2): 162-166,205. DOI: 10.3969/j.issn.1001-2400.2015.02.027
作者姓名:黄宏图  葛渊  张继  查宇飞  毕笃彦  侯志强
作者单位:空军工程大学航空航天工程学院;西北有色金属研究院;空军工程大学防空反导学院;空军工程大学信息与导航学院
基金项目:国家自然科学基金资助项目(61175029,61379104,61372167);国家自然科学基金青年科学基金资助项目(61203268,61202339)
摘    要:针对跟踪中目标尺度变化和旋转问题,将仿射变换和应用到压缩感知跟踪中.首先,以上一帧的跟踪结果为均值,以一定的标准差按照高斯分布,随机生成不同尺度和旋转角度的候选框;然后,通过仿射变换将其转换至直角坐标系中,通过多尺度滤波得到目标在不同尺度下的高维特征向量,采用压缩矩阵将高维特征向量降维至低维空间;最后,将低维特征向量通过贝叶斯分类器选取具有最大响应的候选位置作为目标的跟踪位置.在此基础上分别提取正负样本来更新分类器参数,从而实现持续稳定的跟踪.实验结果表明,该算法能够较好地解决压缩感知跟踪中的目标旋转和尺度变化问题.

关 键 词:视频跟踪  仿射变换  尺度变化  目标旋转  压缩感知
收稿时间:2013-12-25

Application of affine transform in compressed sensing tracking
HUANG Hongtu;GE Yuan;ZHANG Ji;ZHA Yufei;BI Duyan;HOU Zhiqiang. Application of affine transform in compressed sensing tracking[J]. Journal of Xidian University, 2015, 42(2): 162-166,205. DOI: 10.3969/j.issn.1001-2400.2015.02.027
Authors:HUANG Hongtu  GE Yuan  ZHANG Ji  ZHA Yufei  BI Duyan  HOU Zhiqiang
Affiliation:(1. Aeronautics and Astronautics Engineering College, Air Force Engineering Univ., Xi'an 710038, China;2. Northwest Institute of Nonferrous Metal Research, Xi'an 710016, China;3. Air Defense and Antimissile Institute, Air Force Engineering Univ., Xi'an 710051, China;4. Information and Navigation Institute, Air Force Engineering Univ., Xi'an 7100772, China)
Abstract:The affine transform is employed to solve the scale change and rotation in in compressed sensing tracking. Firstly according to the Gauss distribution of a certain variance, a number of candidates are produced based on the former tracking result. Then the patch is transformed into the orthogonal coordinate via affine transformation. And the high-dimension feature vector is acquired through the multiple-scale filter. The high-dimension feature vector is compressed through the compressive matrix. Finally, the low-dimension feature vector is passed through the bayes classifier and the candidate with the highest response is recognized as the tracking result, on the basis of which the positive and negative samples are extracted to update the parameters of the naive bayes classifier. Experimental results show that the proposed algorithm can well cope with the scale variation and rotation in compressed visual tracking.
Keywords:visual tracking   affine transform   scale variation   object rotation   compressed sensing  
本文献已被 CNKI 等数据库收录!
点击此处可从《西安电子科技大学学报(自然科学版)》浏览原始摘要信息
点击此处可从《西安电子科技大学学报(自然科学版)》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号