Real-time coding scheme for high-resolution remote sensing images

Traditional wavelet-based bit-plane coding schemes are not suitable for hardware pipeline processing. To address this issue, a fast hybrid coding framework is proposed based upon the properties of high-resolution images. Wavelet coefficients are partitioned into three types: the low-pass subband is encoded losslessly; the first level high-pass subbands undergo run-length coding on the basis of wide quantization; the other subbands are processed with adaptive quantization. According to the relationship between input and output data ratio, quantization threshold can be adjusted dynamically, and variable compression ratio can be achieved. Experimental results show that, the proposed algorithm provides similar visual quality and PSNR results in comparison with other typical schemes, however, the execution time is less than one-fifth of such ones.