Identification of Human Motion Using Stationary Camera

Abstract

Video surveillance is currently one of the most active research topics in the computer vision community. During motion, the surveillance system can detect moving objects and identify them as humans, animals, vehicles. This strong interest is driven by a wide spectrum of promising applications in surveillance system such as Military security, Public and commercial security, etc. Recognition of Human Motion (RHM) is proposed system which includes detection, feature extraction and recognition of people from image sequences involving humans. In proposed system frame differencing method is used for moving object detection and Neural Network is used for recognition of human motion. System uses half bottom part of the frames for generating more accurate result which is identify the moving object is human. Experimental results show that human motion can be correctly classified.