通过采用自适应提升小波分析,根据处理信息的局部特征自适应地调整预测和更新算子实现与处理信息的准确匹配,有效地降低信息小波分析的运算量和计算的复杂性,并且能较好地实现同址运算,便于采用DSP等硬件实现;其次,通过定义浮动阈值,在编码中提出了采用自适应深度优先搜索的扫描策略,进一步改善了数字图像的编码效率。实验结果表明,通过这些措施实现的数字图像压缩算法,其重构的图像质量较好,有效地降低了压缩后图像的比特数,图像的压缩效果、编码时间等亦有较大的改善。

This paper introduces adaptive lifting wavelet transform algorithm firstly.Prediction and update operator have been adjusted adaptively according to information partial characteristic and accurate match to the process information.The algorithm can decrease the amount of calculation and computational complexity in the image wavelet analysis,can carry out the same address operation,and is advantageous to recur to the DSP hardware realization.Secondly,this paper proposes to adopt adaptive depth first search strategy through definition floating threshold,and improved image coding efficiency.The experiment results show that the new image compression scheme has been improved and it is better than traditional algorithm in the aspects of image quality,reduced the image bit number,compression effect and compression coding efficiency.