基于重叠块小波变换的遥感图像高速压缩算法

High Speed Remote-Sensing Image Compression Using Overlapped Block-Based Wavelet Transform

针对大尺寸图像进行分块小波变换带来的块效应问题,采用重叠分块的方法,对各重叠块分别进行整数小波变换,并在分析单块图像整数小波变换的基础上提出了基于子带位平面的编码算法;解码时,对逆变换后得到的各个块按照原空间位置重叠相加,即可得到重建图像.实验表明,该算法复杂度低,占用内存少,速度快,无块效应,且各块的码率分配接近于最优方案,适合于遥感图像的高质量压缩编码.

In order to avoid block-effects caused by block-based wavelet transform, a compression algorithm based on overlapped block wavelet transform is presented. The original image is divided into overlapped blocks and each block is processed by integer-to-integer wavelet transform. Then the wavelet transform coefficients bits are coded by subband bit-plane coding (SBPC) algorithm. While decoding the reconstruct image can be obtained by reverse transformed blocks overlapping in the original place. With fast coding speed and no block-effects in the reconstruct image, the algorithm needs low memory for the large size image, which is suitable for remote-sensing image compression.