论文部分内容阅读
为了进一步降低矢量量化的运算量,提出了一种新的快速搜索算法。在最近邻域搜索算法的基础上,提出了一个基于矢量分割的一般性码字排除准则。该准则综合利用子矢量的均值和方差参数,构造了一个判决不等式来排除不可能的码字。算法中子矢量的个数设定为2。实验结果表明,该算法的运算时间是改进的等均值等方差最近邻域搜索(IEENNS)算法的80%左右。该算法的性能要优于以往的几种基于不等式判决的快速搜索算法,可以应用在语音和图像编码算法中。
In order to further reduce the computational complexity of vector quantization, a new fast search algorithm is proposed. Based on the nearest neighbor search algorithm, a general code word exclusion criterion based on vector segmentation is proposed. The criterion comprehensively uses the mean and variance parameters of sub-vectors to construct a decision inequality to exclude the impossible code words. The number of neutron vectors in the algorithm is set to two. The experimental results show that the algorithm’s computation time is about 80% of the improved equal mean nearest neighbor search (IEENNS) algorithm. The performance of the algorithm is better than the previous several fast search algorithms based on inequality decision, which can be used in speech and image coding algorithms.