基于免疫克隆选择的块匹配运动估计

运动估计是视频压缩编码中的关键技术.从运动矢量的特点出发,采用搜索点预测、Gray码编码以及有效的迭代终止准则等策略,提出了基于免疫克隆选择的块匹配运动估计.该方法将块匹配运动估计问题的性质与免疫克隆选择算法所具有的全局搜索特性、解的多样性和不易早熟的特点相融合,在能够获得接近全搜索方法所得到的平均峰值信噪比的前提下,使得平均搜索点数大为降低.仿真实验结果表明,在大多数序列上,该算法都比已有的快速搜索算法具有更高的性能和更少的平均搜索点数.同时,该算法适用面广,对大运动和小运动序列都能得到较好的效果.

Block Motion Estimation Based on Immune Clonal Selection

Motion estimation is a key technique in video compress and coding. Based on the analysis of the character of motion vector, a novel block motion estimation based on immune clonal selection (BMEICS) is proposed in this paper with some strategies like prediction of initial search point, Gray encoding and effective stop criteria. BMEICS synthesizes the character of block motion estimation and that of global search, diversity, and no prone to premature in immune clonal selection. It speeds up the process of motion estimation while maintaining the average with little loss. Experimental results show that BMEICS obtains almost the same as the full search algorithm with fewer search points, and outperforms the existing fast block-matching algorithms for most sequences in terms of speed and quality. Furthermore, BMEICS is applicable to all types of video sequences in spite of the degree of motion.