科研之友
微信
新浪微博
Facebook
分享链接
摘要
Vector Quantization is lossy data compression technique and has various applications. Key to Vector Quantization is good codebook. Once the codebook size is fixed then for any codebook generation algorithm the MSE reaches a value beyond which it cannot be reduced unless the codebook size is increased. In this paper we are proposing two-level codebook generation algorithm which reduces mean squared error (MSE) for the same codebook size. For demonstration we have used codebooks obtained from Keker＊s Proportionate error (KPE) algorithm. The proposed method is can be applied to any codebook generation algorithm.
