一种非线性图象矢量量化的重建方法——梯度中值算子法

A Non Linear Reconstruction Algorithm for Vector Quantization in Image Coding

针对图象矢量量化编码复杂度高，并存在方块效应等缺点，本文基于亚－Ｎｙｑｕｉｓｔ采样方式构成训练码矢，并提出了一种新的重建图象方法——梯度中值算子法，它不仅降低了运算的复杂度，而且在插值的同时改善了图象的质量。实验结果表明，在压缩比为０．

For reducing the high complexity of vector quantization and the block effects in image coding, a non linear reconstruction algorithm is proposed in this paper. Based on Half Nyquist sampling pattern, it is not only reducting the operating comlexity, but also improving the quality of the reconstructed image. Results show that the performance quality ( R SN ) of the reconstructed image is more than 33 dB and the subjective quality is much better than others under the compression ratio 0.312 5 bit/pixel.