Adaptive Visual Tracking for Human Motion Detection

Abstract

Abstract—Object detection is one of the fundamental steps for automated video analysis in many vision applications. Object detection in video is usually performed by background subtraction techniques. In the existing method they proposed object detection by pixel variation of the image from one frame to another and the background subtracted by the training process in the recorded videos. In the proposed method the object is detected in the live video that is used for the security purpose. This method can be applicable in bank, jewellery shops, military etc., in an efficient way. Camera is fixed at the required spot and if there is any human object is recognized, it is handled and makes the framework to acknowledge and delivers the cautioning sound in the meantime gatecrasher's picture get exchanged to the comparing specialist through mail correspondence. The mailing system is composed with a counter, which sends the images of the coverage area at regular intervals. Advantages over the existing system are cost and power consumption is reduced as it does not require any sensors. Based on the Camera’s range the monitoring area may be increased. In live video 18 frame is processed at a unit time and it takes again 18 frames to process output. In existing system, they took 5secs to process 1 frame. Proposed method going to achieve 10 frames/sec. For this process MATLAB 7.12(R2011A) software is used.

Keywords – Frame matching algorithm, Object Detection

Cite this Article: N. Manikandaprabu, S. Vijayachitra. Adaptive Visual Tracking for Human Motion Detection. International Journal of Image Processing and Pattern Recognition. 2019; 5 (1): 1-5p.