【作者】 郭少伟； 李德强；

【Author】 GUO Shao-wei1,2, LI De-qiang1 (1.Shenyang Institute of Automation ,Chinese Academy of Sciences, Shenyang 110016, China; 2.Graduate School, Chinese Academy of Sciences, Beijing 100049, China)

【机构】 中国科学院沈阳自动化研究所；

【摘要】 小波变换以其多分辨率分析和时频局域特性在信号处理中得到了广泛的应用。二维Mallat小波分解只能将图像分解为平滑、水平细节、垂直细节、对角线细节等四个部分。大多数图像特别是自然图像存在丰富的方向信息,然而Mallat分解只能对水平、垂直、对角线三个方向分析图像细节,所以在分解方向上存在不足。Contourlet变换除了具有和小波一样的多尺度、时频局部性,还具有方向性和各向异性。但是这种方法变换后数据量大于原始数据,存在冗余。将Mallat分解和Contourlet中的方向滤波相结合,提出了一种新的图像压缩方法,保持了二维Mallat分解的无冗余性,并且充分利用了Contourlet的方向性。仿真实验表明这种方法优于单独采用Mallat分解和Contourlet变换。更多还原

【Abstract】 Wavelet transform has been widely used in signal processing.2-D Mallat decomposition method can only decompose the image into approximation, horizontal, vertical and diagonal details. Most images have rich directional information. However, Mallat method is able to only analyze the image details on the three directions, and is lack of directionality. Contourlet transform owns directionality and anisotropy besides the multi-resolution and localized time/frequency properties. But Contourlet is a redundant transform. Mallat decomposition and directional filtering of Contourlet are combined together, and a new image compression method is proposed, which keeps the non-redundancy of Mallat and multi-directionality of Contourlet. Simulation results show this new method outperforms Mallat decomposition and Contourlet.更多还原