Object Tracking Model Based on Fusing Multiple Weighted Distribution Field Features
LUO Huilan1, SHAN Shunyong1, KONG Fansheng2
1.School of Information Engineering, Jiangxi University of Science and Technology, Ganzhou 341000 2.College of Computer Science and Technology, Zhejiang University, Hangzhou 310027
Abstract Object tracking is difficult to be implemented by using single feature. Histogram is simple and convenient to describe the image features. However, the spatial information of the features can not be expressed by the statistical histograms, and the distribution field descriptors can reflect the spatial information of the features. Based on their advantages, an object tracking algorithm is proposed by fusing gray value features, texture features and edge features. Three kinds of features are combined through distribution field descriptors to form joint representations. And the distribution layers of dense distribution field features are multiplied by the corresponding weights to construct an efficient target model. An adaptive object model updating scheme is used to update the target model and adapt to varietiesof the background and the illumination. The experimental results on commonly used testing video sequences show that the proposed algorithm generates better performance in complicated situations, such as pose change, rotation, occlusion and illumination changes and it has stronger robustness.
LUO Huilan,SHAN Shunyong,KONG Fansheng. Object Tracking Model Based on Fusing Multiple Weighted Distribution Field Features[J]. , 2016, 29(2): 131-142.
[1] YILMAZ A, JAVED O, SHAH M. Object Tracking: A Survey[EB/OL].[2015-01-02]. http://crcv.ucf.edu/papers/Object%20Tracking.pdf. [2] CHEN W G. Moving Shadow Detection in Video Surveillance Based on Multi-feature Analysis // Proc of the 2nd International Conference on Multimedia and Signal Processing. Shanghai, China, 2012: 224-231. [3] CHAU D P, THONNAT M, BRMOND F, et al. Online Parameter Tuning for Object Tracking Algorithms. Image and Vision Computing, 2014, 32(4): 287-302. [4] 邬大鹏,程卫平,于盛林.基于帧间差分和运动估计的Camshift 目标跟踪算法.光电工程, 2010, 37(1): 55-60. (WU D P, CHENG W P, YU S L. Camshift Object Tracking Algorithm Based on Inter-Frame Difference and Motion Prediction. Opto-Electronic Engineering, 2010, 37(1): 55-60.) [5] 吴 垠,李良福,肖樟树,等.基于尺度不变特征的光流法目标跟踪技术研究.计算机工程与应用, 2013, 49(15): 157-161. (WU Y, LI L F, XIAO Z S, et al. Optical Flow Motion Tracking Algorithm Based on SIFT Feature. Computer Engineering and Applications, 2013, 49(15): 157-161.) [6] 罗 艳,项 俊,严明君,等.基于多示例学习和随机蕨丛检测的在线目标跟踪.电子与信息学报, 2014, 36(7): 1605-1611. (LUO Y, XIANG J, YAN M J, et al. Online Target Tracking Based on Multiple Instance Learning and Random Ferns Detection. Journal of Electronics & Information Technology, 2014, 36(7): 1605-1611.) [7] ZHANG K H, SONG H H. Real-Time Visual Tracking via Online Weighted Multiple Instance Learning. Pattern Recognition, 2013, 46(1): 397-411. [8] ZHANG K H, ZHANG L, YANG M H. Real-Time Compressive Tracking // Proc of the 12th European Conference on Computer Vision. Florence, Italy, 2012: 864-877. [9] 刘献如,蔡自兴,唐 琎.基于SAD与UKF-Mean Shift的主动目标跟踪.模式识别与人工智能, 2010, 23(5): 646-652. (LIU X R, CAI Z X, TANG J. Automatic Object Tracking Method Based on SAD and UKF-Mean Shift. Pattern Recognition and Artificial Intelligence, 2010, 23(5): 646-652.) [10] BAO C L, WU Y, LING H B, et al. Real Time Robust l1 Tracker Using Accelerated Proximal Gradient Approach // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Providence, USA, 2012: 1830-1837. [11] WANG Q, CHEN F, XU W L, et al. Object Tracking via Partial Least Squares Analysis. IEEE Trans on Image Processing, 2012, 21(10): 4454-4465. [12] NING J, ZHANG L, ZHANG D, et al. Robust Mean-Shift Tra-cking with Corrected Background-Weighted Histogram. IET Computer Vision, 2012, 6(1): 62-69. [13] YANG M, PEI M T, WU Y W, et al. Learning Online Structural Appearance Model for Robust Object Tracking. Science China Information Sciences, 2015, 58(3): 1-14. [14] LEICHTER I, LINDENBAUM M, RIVLIN E. Tracking by Affine Kernel Transformations Using Color and Boundary Cues. IEEE Trans on Pattern Analysis and Machine Intelligence, 2008, 31(1): 164-171. [15] COMANICIU D, RAMESH V, MEER P. Kernel-Based Object Tra- cking. IEEE Trans on Pattern Analysis and Machine Intelligence, 2003, 25(5): 564-577. [16] WANG H, LIU C, XU L, et al. Multiple Feature Fusion for Tracking of Moving Objects in Video Surveillance // Proc of the International Conference on Computational Intelligence and Security. Suzhou, China, 2008, I: 554-559. [17] SEVILLA-LARA L, LEARNED-MILLER E. Distribution Fields for Tracking // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Providence, USA, 2012: 1910-1917. [18] 王 宪,张方生,宋书林,等.基于分布域描述算子的视频目标跟踪.光电工程, 2013, 40(5): 21-27. (Wang X, Zhang F S, Song S L, et al. Visual Object Tracking Algorithm Based on Distribution Fields Descriptor. Opto-Electronic Engineering, 2013, 40(5): 21-27.) [19] HUANG Y S, OU Z H, YU H H, et al. Non-uniform SURF Feature Point Detection and Matching [EB/OL]. [2015-01-10]. http://www.scientific.net/amm.284-287.3184.pdf. [20] 王保云,范保杰.基于颜色纹理联合特征直方图的自适应 Meanshift 跟踪算法.南京邮电大学学报(自然科学版), 2013, 33(3): 18-25. (WANG B Y, FAN B J. Adoptive Meanshift Tracking Algorithm Based on the Combined Feature Histogram of Color and Texture. Journal of Nanjing University of Posts and Telecommunications (Natural Science), 2013, 33(3): 18-25.) [21] KIM G, KIM T, LIM D H, et al. Sketch Filter for Feature Extra-ction and Rendering Applications. Electronics Letters, 2013, 49(13): 805-806. [22] HSIAO Y Z, PEI S C. Edge Detection, Color Quantization, Se-gmentation, Texture Removal, and Noise Reduction of Color Image Using Quaternion Iterative Filtering[EB/OL].[2015-01-03].http://electronicimaging.spiedigitallibrary.org/article.aspx?articleid=1887507. [23] ZHANG K H, ZHANG L, YANG M H. Real-Time Object Tracking via Online Discriminative Feature Selection. IEEE Trans on Image Processing, 2013, 22(12): 4664-4677.
2016, Vol. 29 
