小波低频子带预测的遥感图像分布式编码

传感技术的不断发展使遥感图像的时间、空间和光谱分辨率不断提高,信息量日益增大,给遥感图像的存储和传输带来了极大挑战,传统图像编码策略难以适应这种基于上行的遥感图像应用需求。本文提出一种基于小波低频子带预测的分布式遥感图像编码方案,利用预测插值的方法得到用于解码Wyner-Ziv的辅助信息,图像的低频信息采用小波零树编码方案,而预测误差图像采用基于均匀量化和位平面的Turbo混合编码方案;在解码端,以预测误差图像的码流作为辅助信息,将其与图像的低频信息码流进行联合解码。实验结果表明,提出方法具有编码效率高、失真率小、压缩比高、抗误码性强和码流具有质量可分级等特性。

Distributed coding of remote sensing image based on wavelet low frequency sub-band prediction

As the continuous development of sensor technology to improve spatial, temporal resolution and spectral resolution of remote sensing images and to increase the amount of information, thus the storage and transmission of remote sensing image has brought significant challenges, so traditional image coding strategy is difficult to adapt to the applications of remote sensing images based on the uplink demand. This paper presents a coding scheme which is based on wavelet low-frequency sub-band prediction of distributed remote sensing images, using interpolation prediction method to get the auxiliary information which is used to decode the Wyner-Ziv. low-frequency information of images use embedded zerotree wavelet (EZW) coding scheme, while the prediction-error image use the Turbo hybrid coding scheme based on uniform quantization and the bit plane. At the decoder, the stream of prediction-error image is considered as an auxiliary information and the stream of low-frequency information are joint decoding. Experimental results show that the proposed method has high coding efficiency, small distortion, high compression ratio, strong ability of error control and scalable bit stream with quality and other characteristics.