图象信号的变换域锥形矢量量化器的实现

A pyramid vector quantizer realization for the image signals in transform domain

本文提出了用于图象压缩编码的一种变换域锥形矢量量化器(简称PVQ)的实现方法。鉴于 PVQ 希望输入信源具有 Laplacian 分布,而哈达玛变换(WHT)和离散余弦变换(DCT)中的非直流系数近似服从 Laplacian 分布,采用 WHT 与 DCT 与矢量量化技术相结合的方法,实现了静态图象的压缩编码。该方法的特点是量化器设计结构简单、复杂度低、可大维数编码和有较高的压缩比。在均方误差失真测度准则下,该量化器在恢复图象的质量方面和保护图象边缘方面性能较佳,在图象的高效编码和实现实时处理方面具有可观的研究前景。

A method that the pyramid vector quantizer(PVQ)in transform domain used in theimage compressing cncoder is presented.The advantage of the PVQ is that it is an imageprocessing technology with a simple structure,low complexity,large dimension coding anda large compressing ratio.With mean square error criterion(MSE),the performance of thequantizer is good in restructuring image and protecting image edge.The quantizer is an at-tractive project in high quality image coding and real-time processing.