论文部分内容阅读
本文利用人眼的视觉特性提出了改进DPCM(差分脉码调制)图象编码的两种方法:一种是采用膈行隔象素取样和优化修正插值的自适应DPCM,可使图象的比特率由8b/pel压缩到1.0～1.5b/pel,量化信噪比可达39dB以上;另一种是采用方块编码的自适应DPCM,可在同样图象质量的条件下,把此特率进一步压缩到0.8～1.0b/pel。这样,基本上解决了DPCM编码中压缩比提高与图象质量下降的矛盾。
In this paper, two methods to improve the DPCM (Differential Pulse Code Modulation) image coding are proposed by using the visual characteristics of the human eye. One is to use the DPCM with pixel-by-pixel sampling and optimized interpolation to optimize the interpolation, The rate of compression from 8b / pel to 1.0 ~ 1.5b / pel, quantization SNR up to 39dB above; the other is the use of block-coded adaptive DPCM, in the same image quality conditions, the rate further Compressed to 0.8 ~ 1.0b / pel. In this way, the contradiction between the improvement of compression ratio and the degradation of image quality in DPCM coding is basically solved.