An efficient object tracking technique using block-matching algorithm

Abstract

A real-time tracking method based on a motion estimation algorithm is presented. The difference technique and block-matching algorithm (BMA) are employed to achieve this purpose. The resultant motion vectors obtained by BMA are scanned to count the repetitions for each motion vector. Hence, the maximum repetition motion vector is taken to represent the motion of the moving object. The method was successfully tested for different objects using real images.