提出了一種基于DPCM的從有損到無(wú)損的漸進(jìn)圖像壓縮和編碼格式。算法分為兩個(gè)
階段。第一個(gè)階段采用基于馬爾可夫模型預(yù)測(cè)的DPCM算法得到低分辨率的近似圖像；第二個(gè)階段對(duì)量化誤差進(jìn)行漸進(jìn)傳輸，使圖像被漸進(jìn)地完善。它通過(guò)對(duì)量化誤差實(shí)施小波變換，根據(jù)量化誤差的局部頻率特性，用簡(jiǎn)單的規(guī)則把小波變換后的系數(shù)進(jìn)行分組，最終使得漸進(jìn)圖像傳輸成為可能。實(shí)驗(yàn)結(jié)果表明，由于使用了馬爾可夫模型和小波變換，可以更快的得到細(xì)節(jié)豐富的圖像。
關(guān)鍵詞:圖像壓縮；漸進(jìn)圖像傳輸；DPCM；自適應(yīng)預(yù)測(cè)；小波變換
Abstract: A DPCM-based lossy-to-lossless progressive image compression and transmission
scheme is proposed. The method consists of two phases: the first phase makes use of DPCM
algorithm transmitting a low-resolution approximation of the image, in which Markov model is
adopted when predicting; The second phase transmits quantitative errors progressively, therefore the low-resolution image can be then progressively improved. The second phase first transforms the quantitative errors by wavelet transform, then employing a low-complexity classification process to separate transformed quantitative errors into multiple teams according to local frequency feature, so that progressive transmission becomes possible. The experimental results show that the presented method could refine the approximation image as quickly as possible because of the use of Markov model and wavelet transform.
Keywords: Image compression; Progressive image transmission; DPCM; Adaptive
prediction; Wavelet transform