[1] Yang Biao, Lin Guoyu, Zhang Weigong,. Integration of Lab model and EHOGfor human appearance matching across disjoint camera views [J]. Journal of Southeast University (English Edition), 2012, 28 (4): 422-427. [doi:10.3969/j.issn.1003-7985.2012.04.009]

Copy

Integration of Lab model and EHOGfor human appearance matching across disjoint camera views()

结合Lab模型与EHOG特征的摄像机离散视域人物外表匹配

Share：

Journal of Southeast University (English Edition)[ISSN:1003-7985/CN:32-1325/N]

Volumn:

28

Issue:

2012 4

Page:

422-427

Research Field:

Computer Science and Engineering

Publishing date:

2012-12-30

Info

Title:

Integration of Lab model and EHOGfor human appearance matching across disjoint camera views

结合Lab模型与EHOG特征的摄像机离散视域人物外表匹配

Author(s):

Yang Biao;  Lin Guoyu;  Zhang Weigong

School of Instrument Science and Engineering, Southeast University, Nanjing 210096, China

杨彪;  林国余;  张为公

东南大学仪器科学与工程学院, 南京 210096

Keywords:

human appearance matching;  Lab model;  extended histogram of oriented gradients(EHOG);  disjoint camera views

人物外表匹配;  Lab模型;  扩展梯度方向直方图;  摄像机离散视域

PACS:

TP391

DOI:

10.3969/j.issn.1003-7985.2012.04.009

Abstract:

The integration of the Lab model with the extended histogram of oriented gradients(EHOG)is proposed to improve the accuracy of human appearance matching across disjoint camera views under perturbations such as illumination changes and different viewing angles. For the Lab model that describes the global information of observations, a sorted nearest neighbor clustering method is proposed for color clustering and then a partitioned color matching method is used to calculate the color similarity between observations. The Bhattacharya distance is employed for the textural similarity calculation of the EHOG which describes the local information. The global information, which is robust to different viewing angles and scale changes, describes the observations well. Meanwhile, the use of local information, which is robust to illumination changes, can strengthen the discriminative ability of the method. The integration of global and local information improves the accuracy and robustness of the proposed matching approach. Experiments are carried out indoors, and the results show a high matching accuracy of the proposed method.

针对摄像头离散区域存在的光照变化、视角变化等干扰, 提出一种结合Lab模型以及扩展梯度方向直方图特征的方法来改善人物外表匹配的准确率.对于描述目标全局信息的Lab模型, 提出一种排序最近邻聚类算法进行颜色聚类, 然后使用分块颜色匹配算法计算观察值之间的颜色相似度.对于描述目标局部信息的扩展梯度方向直方图特征, 使用巴氏距离计算2个观察值之间的相似度.全局信息可以很好地描述目标外形, 并且能够适应摄像头视角的变化以及目标尺度上的改变.局部信息对光照变化具有更强的鲁棒性, 它能够增强模型的辨别能力.全局信息和局部信息的结合保证了所提出算法的精确度和鲁棒性.室内实验结果显示所提出的算法具有较高的正确匹配率.

References:

[1] Huang T, Russell S. Object identification in a Bayesian context [C]//Proceedings of the Fifteenth International Joint Conference on Artificial Intelligence. San Francisco, CA, USA, 1997: 1276-1282.
[2] Huang J G, Kong B. A new method of unstructured road detection based on HSV color space and road features [C]//International Conference on Information Acquisition. Seogwipo-si, South Korea, 2007:596-601.
[3] Andrew G, Richard B. Tracking objects across cameras by incrementally learning inter-camera color calibration and patterns of activity [C]//Proceedings of the 9th European Conference on Computer Vision. Graz, Austria, 2006: 125-136.
[4] Pier L M, Paolo S, Tiziana D O. Object tracking by non-overlapping distributed camera network [C]//11th International Conference on Advanced Concepts for Intelligent Vision Systems. Bordeaux, France, 2009:516-527.
[5] Omar J, Khurram S, Mubarak S. Appearance modeling for tracking in multiple non-overlapping cameras [C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, DC, USA, 2005:26-33.
[6] Lin Mingxiu, Liu Weijia, Xu Xinhe. Algorithm for multimode vehicle tracking based on template match [J]. Journal of System Simulation, 2007, 19(7):1519-1522.(in Chinese)
[7] Liu X B, Lin L, Yan S C. Adaptive object tracking by learning hybrid template online [J]. Circuits and Systems for Video Technology, 2011, 21(11):1588-1599.
[8] Raul M, Carlos R D, Fernando J. Robust 3D people tracking and positioning system in a semi-overlapped multi-camera environment [C]//15th International Conference on Image Processing. San Diego, CA, USA, 2008:2656-2659.
[9] Lam E Y. Combining gray world and retinex theory for automatic white balance in digital photography [C]//Proceedings of the 9th International Symposium on Consumer Electronics. Macau, China, 2005:134-139.
[10] Li E, Cai Lidong. An image compensation method based on the illumination component correction [J]. Journal of Changchun University of Science and Technology: Natural Science Edition, 2010, 33(2):137-139.(in Chinese)
[11] Hu Zhengping, Yang Jianxiu. Object localization algorithm based on mixture model of HOG feature and LSVM [J]. Signal Processing, 2011, 27(8):1206-1212.(in Chinese)
[12] Massimo P, Eric D C. Track matching over disjoint camera views based on an incremental major color spectrum histogram [C]//IEEE Conference on Advanced Video and Signal Based Surveillance. Teatro Sociale, Como, Italy, 2005:147-152.
[13] Yang H D, Lee S W. Multiple pedestrian detection and tracking based on weighted temporal texture features [C]//Proceedings of the 17th International Conference on Pattern Recognition. Washington, DC, USA, 2004:248-251.

Memo

Memo:

Biographies: Yang Biao(1987—), male, graduate; Zhang Weigong(corresponding author), male, doctor, professor, zhangwg@seu.edu.cn.
Foundation items: The National Natural Science Foundation of China(No.60972001), the Science and Technology Plan of Suzhou City(No.SG201076).
Citation: Yang Biao, Lin Guoyu, Zhang Weigong. Integration of Lab model and EHOG for human appearance matching across disjoint camera views[J].Journal of Southeast University(English Edition), 2012, 28(4):422-427.[doi:10.3969/j.issn.1003-7985.2012.04.009]

Common functions