| Nowadays, with the rapid development of Internet and multimedia technology, digital video based on video-coding has become more popular. Motion estimation which plays an important part in video-coding can reduce the redundancies among video frames and make more effective in coding. How to find an effective algorithm of motion estimation is one of the most popular topics in video-coding area. Block-matching algorithm is simply and easily carried out by software and hardware, and is widely used in MPEG-1/2/4 and H.26x.Motion estimation takes too much time in video-coding. In order to reduce the computational complexity and improve the accuracy of motion estimation, we present a new block-matching algorithm called linear search algorithm in this paper. First, we check the SAD of the predicted motion vector. If it is smaller than 512, the predicted motion vector can be considered as the motion vector of current block. Otherwise, a square search template composed of 3*3 checking points is used to find the direction of motion vector. We can promote the precision of the motion vector by checking SAD of other 3 points in this direction, and finally go to the position with the smallest mean block distortion rapidly. Experimental results finally show that the proposed algorithm is superior to other widely used algorithms in both performance and computational complexity. |
PDF Full Text Request