Real time vision based object tracking using CAMSHIFT algorithm with enhanced color image segmentation

In this paper we implement a vision based moving Object Tracking system with Wireless Surveillance Camera which uses a color image segmentation and color histogram with background subtraction for tracking any objects in non-ideal environment. The implementation of the moving video objects based on the Continuously Adaptive Mean Shift (CAMSHIFT) algorithm is presented by optimizing the kernel variants by adjusting the HSV value for various environmental conditions. The object occlusions are also removed by calculating the minimal distance between the two objects using Bhattacharya coefficients and it is robust to changes in shape with complete occlusion. Based on the selection of the users Region of Interest (ROI) the HSV value of the object being tracked by means of CAMSHIFT algorithm which uses an Adaptive block based approach for continuous object tracking. A software approach for real time implementation of moving object tracking is done through MATLAB.