Date of Publication :18th April 2017
Abstract: Pedestrian detection is the process of detecting a person from image or a live processing video. This information is sent to user in the forms of sound or video alarm. It helps the user in taking necessary decisions. Pedestrian detection process comes under security and safety aspect. This paper introduces a concept of machine learning and its consequences. It combines two algorithms, one is motion based detection and another is feature based detection. Both the algorithms complement with each other. Proposed method increased overall accuracy of detection. Motion based detection is a fast detection method which increased speed of detection. Motion based detection faces illumination problem in real time scenarios. This issue is discussed and solved in this paper. Histogram of oriented gradient (HOG) features are used in feature based detection with support vector machine. Support vector machine is a machine learning technique which is capable to classify given features. Depending upon the shapes, appearances and postures of pedestrians, HOG features will be changed. This make detection process more complex
Reference :
-
[1] Navneet Dalal and Bill Triggs, "Histogram of oriented gradient for Human detection”, IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005, pp 1-7.
[2] Chi Qin Lai and Soo Siang Teoh, "A Review on Pedestrian Detection Techniques Based on Histogram of Oriented Gradient Feature", IEEE Student Conference on Research and Development 2014, pp1-6.
[3] P. Viola, M. J. Jones, and D. Snow. "Detecting pedestrians using patterns of motion and appearance" The 9th ICCV, Nice, France, volume 1,2003, pp 734741.
[4]Liming Wang, Jianbo Shi, Gang Song, I-fan Shen, (2007). Penn-Fudan Database for Pedestrian Detection and Segmentation [Online]. Available: https://www.cis.upenn.edu/~jshi/ped_html/