基于整数小波变换与双阈值交替的遥感图像压缩

提出一种基于双阈值交替的遥感图像编码（DTAC）算法。根据整数小波变换（IWT）分解后不同阈值上的重要系数分布状况，对重要系数相对集中的区间用整数平方代替2的整数幂（IPT）作为量化阈值，在其他阈值区间中仍使用IPT，然后对每个阈值平面上的重要系数采用统一的编码策略。实验表明，新算法具有较低编码复杂度，不仅支持单一码流下从有损到无损的遥感图像编码，而且峰值信噪比（PSNR）高于采用IWT的SPECK编码算法。

Remote Sensing Image Compression Based on Integer Wavelet Transform and Double Threshold Alternating

A new compression algorithm(DTAC) based on double threshold alternating is presented.According to the significant coefficient distribution on different thresholds based on integer wavelet transform(IWT),the new algorithm uses the integer square as the quantization thresholds instead of the integer powers of two(IPT) on the field where the significant coefficients concentrate relatively.However,the IPT are reserved as the quantization thresholds of other threshold planes.On every threshold plane,the single encoding strategy is applied for the significant coefficients.Experiment results show that the new algorithm can lower the coding complexity,support the remote sensing image compression using a single bitstream from lossy to lossless,and provide PSNR performance better than SPECK algorithm using the same IWT.