A New Hybrid Search Algorithm with Novel Cross-Diagonal-Hexagon Search Video Coding Algorithm for Block Motion Estimation

摘要

Computational complexity and motion estimation time are the main factors which affect the performance of video codec for real time applications. In block matching motion estimation algorithm, search patterns have more effect on searching speed and distortion performance. The major block matching algorithm search types are full search and fast search. The full search will search all points of the reference frame and the fast search will use only less search points. There are many algorithms in the literature and in which the hybrid algorithms outperforms the other existing fast motion estimation algorithms. In this paper, a new hybrid search algorithm with novel cross diagonal-hexagon search pattern is proposed for fast block motion estimation. It uses a novel search patterns in both global and local search is performed. Compared to the existing algorithms, the proposed algorithm uses an average of 8.791 search points, resulting in reduced computational complexity and faster processing with better video coding quality. The proposed algorithm is implemented in Cyclone IV E (EP4CE115F29C7) and a better performance improvement in terms of area and speed is achieved.