基于比特平面SBI变换的遥感图像无损压缩

Lossless Remote Sensing Image Compression Based on Subblock Interchange in Bit Plane

提出一种基于子块互换(subb lock in terchange,SB I)的遥感图像无损压缩方案.采用小波变换对遥感图像进行分解,将生成的频域数据按不同比特平面分别进行SB I变换并用算术编码进行压缩.该方案改进了基于列的扫描方式,将小波系数按幅值大小进行重新排序,根据SB I变换后数据特点取消了文本压缩中常用的前移编码(M TF),在降低复杂度的前提下提高了压缩率,同时具有信噪比和分辨率可扩展特性.

A lossless remote sensing image compression method based on subblock interchange （SBI） was presented. Remote sensing image was decompressed into frequency domain image by wavelet transform and the bit planes in frequency domain image were permuted by SBI transform seperately. The resulting data was compressed by arithmetic encoder. In the method, the column-based scan mode was improved, wavelet coefficients were resorted in magnitude order, MTF （move to front） code, which was often used in text compression method, was cancelled. The method achieves better compression performance and lower complexity. At the same time the method achieves desired feature of PSNR scalability and resolution scalability.