An optimized block matching algorithm for motion estimation using logical image

The fundamental of motion estimation is that the objects in a frame of video sequence should only move within the original frame to form corresponding objects on the next frame. The concept of block matching method is to partition current frame into a matrix of macro blocks, that are compared with corresponding block and its adjoining neighbors in the last frame to create a vector that determines the motion of a macro block from one position to another in the preceding frame. This movement calculated for all the macro blocks consisting a frame, calculates the motion approximated in current frame. The search region for a perfect macro block equivalent is limited up to s pixels on all fours sides of the corresponding macro block in previous frame. This `s´ is known as the search region. Bigger movements require a big s, and the bigger the search region, the process of motion estimation becomes costly. Motion estimation is the main process of block matching algorithms. The motion vector for every block is calculated by obtaining perfect equivalent block of pixels within the search area region in the last frame according to matching criterion. The efficiency of Block Matching Algorithms can be improved by increasing the PSNR of compensated image, changing Block size on Block Matching Algorithm and decreasing number of computations. Block Matching Algorithm is a part in Video Compression Process. Improving the working of Block Matching Algorithm reduces the time taken to compress a video. Motion estimation is used widely in various image sequence coding schemes. It plays a important part in the carrying & storage of video signals at reduced bit rates.