适于干涉多光谱图像压缩的自适应率控制算法

提出了一种基于优化截取内嵌码块编码（EBCOT）的感兴趣区域（ROD编码干涉多光谱图像压缩方法。小波变换后，对1级分解的高频系数感兴趣区域即包含光谱信息区域进行垂直方向的分解，再对感兴趣区域进行比特平面提升。T1编码器对不同比特平面的编码过程（Codingpass）赋予不同的重要性权值，由高到低依次编码，T2编码器根据所得的比特率自适应地反馈控制T1的编码深度，最后进行率失真优化截取。实验结果表明，该方法提高了恢复图像质量，有效地减少了优化截取内嵌码块编码算法的计算量和内存使用量（bpp-1时，测试图像的整体、感兴趣区域和背景区域平均峰值信噪比均提高0．1dB以上，计算量和内存使用量平均减少40％和60％以上），编码方式适合干涉多光谱图像压缩系统硬件实现。

Adaptive Rate Control Algorithm for Interference Multi-Spectral Image Compression

A method for interference multi-spectral image coding is proposed based on region of interest ROI coding of embedded block coding with optimized truncation EBCOT.After wavelet transform,the vertical wavelet decomposition is taken for the ROI which include the multi-spectral information in the high frequency subband,and the coefficients of ROI are lifted by bit-plane.The different significant weights are given to different coding passes in different bit-planes,which are encoded in the significant weights order in T_1 encoder.T_2 encoder gives feedback to control the coding depth according to the accumulated bit rate,and the rate-distortion optimization truncation is completed at last.The experimental results show that the proposed scheme can improve the reconstructed image quality and efficiently reduce the computational complexity and memory usage of EBCOT for test image at bpp=1,the average peak signal-noise ratio of the whole,the ROI and the background increases more than 0.1 dB,and the average computational complexity and memory usage reduces more than 40% and 60%,which is suitable for hardware implementation of interference multi-spectral image compression system.