A novel approach on object detection and tracking using adaptive background subtraction method

Image processing is an ever increasing research scope area where real time surveillance systems will increases the opportunity to the researchers for developing new modules for all the problems. Particularly in complex video processing operations security, intelligence processing is much needed in the society to satisfy the individuals. Basic object detection and tracking has different techniques and many automated systems are available now days to analyze the particular portion or object from the video. Estimation of moving object from the video sequence provides robustness for same colors for object and the background. In view of reducing the robustness and improving the performance of object detecting and tracking system the proposed model used Markov model based background subtraction. It uses neighborhood method to improve the background performance and Markov random field is used to estimate the energy function to optimize the real time experimental results. Generating saliency map combines the texture and cues to explore the linearly generated objects and tracked using component labeling.