COMPARISON AND IMPLEMENTATION OF FAST BLOCK MATCHING MOTION ESTIMATION ALGORITHMS FOR VIDEO COMPRESSSION

In Digital video communication it is not practical, to store the full digital video without processing, because of the problems encountered in storage and transmission, so the processing technique called video compression is essential. In video compression, one of the computationally expensive and resource hungry key element is the Motion Estimation. The Motion estimation is a process which determines the motion between two or more frames of video. In this paper, Four block matching motion estimation algorithms, namely Exhaustive Search (ES), Three Step Search (TSS), New Three Step Search (NTSS), and Diamond Search (DS) algorithms are compared and implemented for different distances between the frames of the video by exploiting the temporal correlation between successive frames of mristack and foreman slow motion videos and proved that Diamond Search (DS) algorithm is the best matching motion estimation algorithm that achieve best tradeoff between search speed (number of computations) and reconstructed picture quality with extensive simulation results and comparative analysis.