Human detection using HOG features of head and shoulder based on depth map

被引：10

作者：

Tian, Qing ^{[1
]}

Zhou, Bo ^{[1
]}

Zhao, Wen-Hua ^{[1
]}

Wei, Yun ^{[2
,3
]}

Fei, Wei-Wei ^{[4
]}

机构：

[1] College of Information Engineering, North China University of Technology, Beijing

[2] Intelligent Transportation System Research Center, Southeast University, Nanjing

[3] Beijing Urban Engineering Design and Research Institute, Beijing

[4] Systems Engineering Research Institute, CSSC, Beijing

来源：

Journal of Software | 2013年 / 8卷 / 09期

关键词：

Depth Map; Hog Features Of Head And Shoulder; Human Detection; SVM;

D O I：

10.4304/jsw.8.9.2223-2230

中图分类号：

学科分类号：

摘要：

Conventional moving objects detection and tracking using visible light image was often affected by the change of moving objects, change of illumination conditions, interference of complex backgrounds, shaking of camera, shadow of moving objects and moving objects of self-occlusion or mutual-occlusion phenomenon. We propose a human detection method using HOG features of head and shoulder based on depth map and detecting moving objects in particular scene in this paper. In-depth study on Kinect to get depth map with foreground objects. Through the comprehensive analysis based on distance information of the moving objects segmentation extraction removal diagram of background information, by analyzing and comprehensively applying segmentation a method based on distance information to extract pedestrian's Histograms of Oriented Gradients (HOG) features of head and shoulder[1], then make a comparison to the SVM classifier. SVM classifier isolate regions of interest (features of head and shoulder) and judge to achieve real-time detection of objects (pedestrian). The human detection method by using features of head and shoulder based on depth map is a good solution to the problem of low efficiency and identification in traditional human detection system. The detection accuracy of our algorithm is approximate at 97.4% and the average time processing per frame is about 51.76 ms. © 2013 ACADEMY PUBLISHER.

引用

页码：2223 / 2230

页数：7

共 24 条

[1]

Fei jie W., Real-time Detection and Tracking of Human Based on Head and Shoulder Feature, (2010)

[2]

Zhang Q., Reconstruction of intermediate view based on depth map enhancement, Journal of Multimedia, 7, 6, pp. 415-419, (2012)

[3]

Dalal N., Triggs B., Histograms of Oriented Gradients for Human Detection, Computer Vision and Pattern Recognition, 1, pp. 886-893, (2005)

[4]

Yu Q., Shan H., Yu W., Pedestrian detection and counting in video surveillance, Microcomputer information (measurement and control automation), 10, pp. 187-188, (2010)

[5]

Bei H., Guijin W., Xinggang L., Quick matting for vedio based on depth image of the kinect, Journal of Tsinhua University (Science and Technology), 52, pp. 29-32, (2012)

[6]

Yi-bo A., Hui G., Mao-jun Z., Cascade features based method for pedestrian detection in street scene, Joumal of Computer Applications, 31, pp. 129-132, (2011)

[7]

Fanelli G., Weisi T., Gall J., Real time head pose estimation from consumer depth cameras, Lecture Notes in Computer Science, 6835, pp. 101-110, (2011)

[8]

Yong-yu Q., Qing L., Jian-ming G., Sheng-hui Z., HOG and Color Based Pedestrian Detection, Journal of Wuhan University of Technology, 33, pp. 134-138, (2011)

[9]

Feng P., Xuan-yin W., Quan-qiang W., Human detection based on head and shoulder feature intelligent surveillance system, Journal of Zhejiang University (Engineering Science), 38, pp. 10-14, (2004)

[10]

Xia L., Chen C.-C., Aggarwal J.K., Human Detection Using Depth Information by Kinect, IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, (2011)

← 1 2 3 →