A Novel Fast Block-Matching Motion Estimation Algorithm Based on Artificial Immune System

In this paper, a novel fast block-matching algorithm (BMA) for motion estimation is proposed, named fast block-matching motion estimation algorithm based on artificial immune system (AIS). This algorithm inspired from immunology can effectively overcome the inherent drawback of being liable to get trapped in local optima in traditional fast BMAs such as TSS and DS. In the meantime, we proposed an improved immune selection approach based on antibody suppression to reduce computational complexity. Moreover, the immune memory mechanism is utilized in AIS to improve search performance further. The experimental results demonstrate that the proposed AIS outperforms TSS and DS in searching quality at comparable computational cost, which takes a better tradeoff between exploration and exploitation.