Least Square Mean Optimization-Based Real Object Detection and Tracking

In this paper, automatic real-time object detection and tracking is implemented via means of Kalman filter in which the system output is actually tracking the input canceling out any variation due to input and output noises. This paper can be used to develop a surveillance system of static camera and robotic automation visual systems. Whenever a new object comes in camera frame, system uses the concepts of frame subtraction then threshold image by Otsu’s method, and later Kalman filtering is being processed to estimate the next following coordinates of its movement. The work presented here is extended to work at video processing stage. And finally, least square mean optimization technique is used to evaluate the set of system parameters for perfect tracking of forthcoming new objects, and once that parameter is evaluated it be can used to execute tracking process perfectly.