A Rapid Body Limb Detection Algorithm for Video Sequence Image

Abstract:

We describe an efficient algorithm to detect the limbs of a human. In order to realize the real-time and robustness, we utilize a detection algorithm by using integral image and haar-like feature. The integral image is the original image for video, the pixel value of each point is the point of the original image in the top left of all the pixel values. After the convolution of the edge detection template and the original video for each frame, convolution images obtained information and suppress background noise, the body physical location can be calculated by accumulating the image of the integral image and by using haar-like feature. The experiment results show the algorithm can detect the location of the limbs at the rate of 30 frames per second.

• Periodical: Advanced Materials Research (Volumes 433-440)
• Pages: 4324-4329
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.433-440.4324
• Online since: January 2012
• Authors: Hua Chen, Chun Hai Hu, Shu Tao Wang
• Keywords: Haar-Like Feature, Integral Image, Limb Detection, Video Sequence Image

References

[1] Huang Shike, Tao Lin, Zhang Tianxu, An improved algorithm of moving object detection based on optical flow, Journal of Hua Zhong University of Science and Technology: Nature Science. vol. 33, pp.39-41, May (2005).

[2] Cheung C S, Kamath C, Robust techniques for background subtraction in urban traffic video, Proc of the SPIE., vol. 5308, pp.881-892, (2004).

[3] Yu Ting, Zhang Cha, Cohen M, Monocular video foreground /background segmentation by tracking spatial-color Gaussian Mixture Models, Proc of the IEEE Workshop on Motion and Vision Computing, Austin, USA, pp.5-12, (2007).

DOI: 10.1109/wmvc.2007.27

[4] Duan Shengchen, Zheng Kailiu, Generalized Haar-like features for fast face detection, Proceedings of the Sixth International Conference on Machine Learning and Cybernetics, Hong Kong, pp.19-22, (2007).

DOI: 10.1109/icmlc.2007.4370496

[5] T. Mita, T. Kaneko, O. Hori, Joint haar-like features for face detection, Proceeding Of 10th IEEE International Conference on Computer Vision, vol. 2, pp.1619-1626, (2005).

DOI: 10.1109/iccv.2005.129

[6] J. Nishimura, T. Kuroda, Low cost speech detection using haar-like filtering for sensornet, Proceedings of ICSP, (2008).

DOI: 10.1109/icosp.2008.4697683

[7] Paul Viola, Michael Jones, Rapid object detection using a boosted cascade of simple features, Conference on Computer Vsion and Pattern Recognition, (2001).

DOI: 10.1109/cvpr.2001.990517

[8] N. Oliver, A. Garg, E. Horvitz, Layered representations for learning and inferring office activity from multiple sensory channels[, CVIU. Vol. 96, pp.163-180, Feb. (2004).

DOI: 10.1016/j.cviu.2004.02.004