使用频谱词包的接力跟踪目标交接方法

doi:10.3969/j.issn.1001-2400.2016.04.033

西安电子科技大学学报 ›› 2016, Vol. 43 ›› Issue (4): 191-196.doi: 10.3969/j.issn.1001-2400.2016.04.033

• 研究论文 • 上一篇

使用频谱词包的接力跟踪目标交接方法

张小骏;刘志镜;薛鸿民

(西安电子科技大学计算机学院，陕西西安 710071)

收稿日期:2016-01-19 出版日期:2016-08-20 发布日期:2016-10-12
作者简介:张小骏(1964-)，男，西安电子科技大学博士研究生，E-mail: 1479781033@qq.com．
基金资助:
国家自然科学基金资助项目(61173091)

Method for relay tracked target handover using the spectrum bag of words

ZHANG Xiaojun;LIU Zhijing;XUE Hongmin

(School of Computer Science and Technology, Xidian Univ., Xi'an 710071, China)

Received:2016-01-19 Online:2016-08-20 Published:2016-10-12

摘要/Abstract

摘要：

针对接力跟踪中目标跨摄像机交接困难问题，提出了使用频谱词包的目标交接方法．首先对第1摄像机的两幅或多幅目标图像分块提取傅里叶特征；再对特征集进行K均值聚类，选择其中紧密类的质心特征作为正常单词，另引入一个异常单词，与正常单词一起共同构成词包；最后用词包分别描述两摄像机的目标，以欧氏距离度量相似性．实验表明，该方法的目标识别性能和鲁棒性比目前几种常见的基于纹理的方法有一定的提高．

关键词: 多摄像机, 接力跟踪, 目标交接, 傅里叶频谱, 频谱词包

Abstract:

Aiming at the difficulty of the relay-tracked target handover across cameras, a method using the spectrum Bag Of Words(BOW) is proposed. Firstly, 2 or more target images from the first camera are divided into blocks. The Fourier features of the blocks are extracted and then clustered by Kmeans. The averaged vectors of the compact clusters are used as the normal words of the BOW, while an abnormal vector is used as the abnormal word. The targets from the first and second cameras are described in BOW words.The Euclidean distance between the BOW descriptions is used as the target similarity metrics. Experiments show that the proposed method has some definite improvement in target recognition and robustness compared with several common methods based on the texture.

Key words: multiple cameras, relay tracking, target handover, Fourier spectrum, spectrum bag of words

张小骏;刘志镜;薛鸿民. 使用频谱词包的接力跟踪目标交接方法[J]. 西安电子科技大学学报, 2016, 43(4): 191-196.

ZHANG Xiaojun;LIU Zhijing;XUE Hongmin. Method for relay tracked target handover using the spectrum bag of words[J]. Journal of Xidian University, 2016, 43(4): 191-196.

参考文献

［1］ FEI Y, MAKRIS D, VELASTIN S A, et al. Calibration and Object Correspondence in Camera Networks with Widely Separated Overlapping Views［J］. IET Computer Vision, 2015, 9(3): 354-367.
［2］ WANG X G. Intelligent Multi-camera Video Surveillance: a View［J］. Pattern Recognition Letters, 2013, 34: 3-19.
［3］ L Z, XING X, WANG K, et al. Class Energy Image Analysis for Video Sensor-based Gait Recognition: a Review［J］. Sensors, 2015, 15(1): 932-964.
［4］闫允一, 姜帅, 郭宝龙. 结合稳定兴趣点和Gabor小波的图像检索［J］. 西安电子科技大学学报, 2014, 41(5): 118-123.
YAN Yunyi, JIANG Shuai, GUO Baolong. Image Retrieval Using Stable Interest Points and Gabor Wavelet［J］. Journal of Xidian University, 2014, 41(5): 118-123.
［5］ ARIVAZHAGAN S, NIRMALA S. Rotation and Scale Invariant Texture Classification Using Gabor and Curvelet Transforms［J］. International Journal of Tomography and Simulation, 2015, 28(2): 94-105.
［6］刘丽, 匡纲要. 图像纹理特征提取方法综述［J］. 中国图象图形学报, 2009, 14(4): 622-635.
LIU Li, KUANG Gangyao. Overview of Image Textural Feature Extraction Methods［J］. Journal of Image and Graphics, 2009, 14(4): 622-635.
［7］ MENNESSONA J, JEANB C S, MASCARILLA L. Color Fourier-Mellin Descriptors for Image Recognition［J］. Pattern Recognition Letters, 2014, 40: 27-35.
［8］ ALTANTSETSEG E, KATSUTSUGU M, KOUICHI K. Pairwise matching of 3D Fragments Using Fast Fourier Transform［J］.The Visual Computer, 2014, 30(6/8): 929-938.
［9］赵理君, 唐娉, 霍连志, 等. 图像场景分类中视觉词包模型方法综述［J］. 中国图象图形学报, 2014, 19(3): 333-343.
ZHAO Lijun, TANG Ping, HUO Lianzhi, et al. Review of the Bag of Visual Words Models in Image Scene Classification［J］. Journal of Image and Graphics, 2014,19(3): 333-343.
［10］ KHAN R, BARAT C, MUSELET D, et al. Spatial Histograms of Soft Pairwise Similar Patches to Improve the Bag of Words Model［J］. Computer Vision and Image Understanding, 2015, 132: 102-112.
［11］ FARAJI M, SHANBEHZADEH J. Bag of Visual Words,its Detectors and Descriptors: a Survey in Detail［J］. Advances in Computer Science an International Journal, 2015, 4(2): 8-12.
［12］ IONESCU R T, POPESCU M, KERNEL P Q. A Rank Correlation Kernel for Visual Word Histograms［J］. Pattern Recognition Letters, 2015, 55: 51-57.
［13］闫晶莹, 王成儒. 一种新的纹理特征提取算法［J］. 西安邮电学院学报, 2011, 16(1): 49-53.
YAN Jingying, WANG Chengru. A New Method for Texture Feature Extraction［J］. Journal o Xi'an University of Posts and Telecommunications, 2011, 16(1): 49-53.
［14］刘丽, 谢毓湘, 魏迎梅, 等. 局部二进制模式方法综述［J］. 中国图象图形学报, 2014, 19(12): 1696-1720.
LIU Li, XIE Yuxiang, WEI Yingmei, et al. Survey of Local Binary Pattern Method［J］. Journal of Image and Graphics, 2014, 19(12): 1696-1720.

使用频谱词包的接力跟踪目标交接方法

Method for relay tracked target handover using the spectrum bag of words

PDF (PC)

赞

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

Metrics

本文评价

推荐阅读 10