一种基于混合变换的分层图像表示与编码

提出一种基于混合变换的分层图像编码算法，它将原始图像划分成包含不同结构特征的图像平面，如平滑边缘层和纹理层等等，然后采用不同的压缩算法对不同的图像层进行编码，从而达到对整幅图像最优编码的目的。本文实际上给出了一种新的编码框架，在这个框架内综合了不同的信号变换方式，将它们有机结合起来能够提供一种有效的图像表示方式。实验结果表明，与采用单一小波图像编码算法相比，本文算法能够获得更好的图像恢复质量。

Multi-layered image representation and coding based on mixed transforms

This paper proposes a new multi-layered image coding algorithm based on mixed transforms, which divides the original image into multiple image layers, such as smooth and edge layer and texture layer etc, then adopts different compression algorithm to code each layer. As a result, we can achieve the aim of coding whole image optimally. In fact, here we put forward a new image coding frame, in which different signal transforms are combined together to propose an efficient image representation. The experiments shows better visual quality of reconstructed image, compared with traditional single wavelet image coding algorithm.