Abstract Human detection is fundamental for human localization, recognition, and tracking. And it is still a difficult problem because of environment complexity and vision system movement. A new human detection algorithm based on mobile robot vision system is proposed to solve the problem effectively. The approach consists of the following two steps: (1)Waveletbased multiscale edge detection combined with edgelinked operator method is introduced to extract the edges of images. In this image a new morphology method is employed to get the object contourclosed for improvement of the correct recognition rate. And invariant Hu moments are calculated as pattern features vectors. (2)The adaptive Gaussian kernel soft margin support vector machine (CSVM) classifier is designed to distinguish human images from nonhuman ones. Experimental comparisons have been conducted,including adaptive Gaussian CSVM classifiers based on different features and the classifiers with different classification methods. The results validate effectiveness and robustness of the algorithm.
HU ChunHua,MA XuDong,DAI XianZhong. Human Detection Based on Support Vector Machine of Adaptive Gaussian Kernel for Indoor Application[J]. , 2007, 20(4): 492-498.
[1] Cielniak G, Duckett T. Person Identification by Mobile Robots in Indoor Environments // Proc of the IEEE International Workshop on Robotic Sensing. Orebro, Sweden, 2003: 15 [2] Han Hongzhe, Wang Zhiliang, Liu Jiwei, et al. Gait Recognition Based on Linear Discrimination Analysis and Support Vector Machine. Pattern Recognition and Artificial Intelligence, 2005, 18(2): 160164 (in Chinese) (韩鸿哲,王志良,刘冀伟,等. 基于线性判别分析和支持向量机的步态识别. 模式识别与人工智能, 2005, 18(2): 160164) [3] Pan Feng, Wang Xuanyin. Support Vector MachineBased Human Detection under Complex Background. Journal of Image and Graphics, 2005, 18(2): 181186 (in Chinese) (潘 锋,王宣银.基于支持向量机的复杂背景下的人体检测. 中国图象图形学报, 2005, 18(2): 181186) [4] Viola P, Jones M J, Snow D. Detecting Pedestrians Using Patterns of Motion and Appearance // Proc of the IEEE International Conference on Computer Vision. Beijing, China, 2003: 734741 [5] Elzein H, Lakshmanan S, Watta P. A Motion and ShapeBased Pedestrian Detection Algorithm // Proc of the IEEE Intelligent Vehicles Symposium. Columbus, USA, 2003: 500504 [6] Xu Fengliang, Liu Xia, Fujimura K. Pedestrian Detection and Tracking with Night Vision. IEEE Trans on Intelligent Transportation Systems, 2005, 6(1): 6371 [7] Oren M, Papageorgion C, Sinha P, et al. Pedestrian Detection Using Wavelet Templates // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Santa Barbara, USA, 1997: 193199 [8] Hotta K. Object Detection Method Based on Local Kernels and Automatic Kernel Selection by KullbackLeibler Divergence // Proc of the 6th IEEE Workshop on Application of Computer Vision. Orlando, USA, 2002: 105111 [9] Gonzalez R C, Woods R E. Digital Image Processing. 3rd Edition. Milano, Italy: AddisonWesley, 1992 [10] Mallat S, Zhong Sifen. Characterization of Signals from Multiscale Edges. IEEE Trans on Pattern Analysis and Machine Intelligence, 1992, 14(7): 710732 [11] Cheng Lizhi, Wang Hongxia, Luo Yong. Theory and Application of Wavelet. Beijing, China: Chinese Science Press, 2004 (in Chinese) (成礼智,王红霞,罗 永.小波的理论与应用.北京:科学出版社, 2004) [12] Cristianini N, ShaweTaylor J. An Introduction to Support Vector Machines and Other KernelBased Learning Methods. Cambridge, UK: Cambridge University Press, 2000 [13] Cristianini N, Campbell C, ShaweTylor J. Dynamically Adapting Kernels in Support Vector Machines // Kearns M S, Solla S A, Cohn D D, eds. Advances in Neural Information Processing Systems. Cambridge, USA: MIT Press, 1999, 11: 204210
2007, Vol. 20 
