Moving vehicle segmentation in a dynamic background using self-adaptive kalman background method

Abstract

This paper introduce the adaptive kalman filter to modeling dynamic background for background subtraction. Background subtraction is a method to identify object and famous used in moving object segmentation. In this paper we also investigate a comparison study on Gaussian subtraction method, frame differencing method and approximate median method. The detection of object will be shown in the result.