一种基于小波的Contourlet变换的图像压缩算法

Image Compression Algorithm Using Wavelet-Based Contourlet Transform

针对JPEG2000图像压缩不能有效地保护图像边缘的局限，提出了一种结合基于小波的Contourlet变换与最优截断嵌入码块编码（EBCOT）的静止图像压缩算法（CEBCOT）．在Contourlet变换中采用小波变换取代拉普拉斯塔式变换获得了非冗余基于小波的Contourlet变换，它通过方向滤波器组把小波变换的高频子带进一步分解为多个方向子带，从而更稀疏地表示了图像的边缘和纹理．CEBCOT采用了改进的EBCOT编码，它依照方向子带大小进行编码块分割，提高了编码性能．实验结果表明，相对于JPEG2000，所提出的算法获得的压缩图像边缘更加清晰，峰值信噪比提高了0．1～0．8dB．

To overcome the limitation of JPEG2000 in image edge preservation, a novel still image compression algorithm called CEBCOT was presented, in which the algorithm was constructed by wavelet-based contourlet transform and embedded block coding with optimized truncation （EBCOT）. In the contourlet transform, the Laplacian pyramid transform was replaced by the wavelet transform to obtain non-redundant wavelet-based contourlet transform. The high frequency subband decomposed by the wavelet transform was further decomposed into multiple directional subbands by directional filterbank so as to explain the edge and texture of the image more sparsely. A modified EBCOT coder with code-block partition according to the size of directional subband was adopted to increase the coding performance. Experiments demonstrate that compared to JPEG2000, the edges of the compressed images obtained by our proposed algorithm are clearer, the peak signal-noise ratio is increased by 0. 1 dB to 0.8 dB.