基于多小波变换的图像编码算法研究  

An Image Coding Method Based on Multiwavelet Transform

作　　者：柳薇[1] 陈冬丽[2] 

机构地区：[1]华南师范大学计算机学院,广东广州510631 [2]中山大学图书馆,广东广州510275

出　　处：《中山大学学报（自然科学版）》2011年第5期50-53,共4页Acta Scientiarum Naturalium Universitatis Sunyatseni

基　　金：广东省科技计划资助项目(2009B010900027)

摘　　要：多小波是小波理论的新分支,将传统的单小波拓展到矢量空间。多小波既包括了传统小波的全部优点,同时又克服了单小波的缺陷,将实际应用中十分重要的正交性、对称性、紧支性、光滑性完美地结合在一起,因此多小波在图像处理方面的应用也显示出巨大的潜力。该文在研究了JPEG2000的核心编码技术——EB-COT算法和多小波变换原理的基础上,提出了结合EBCOT算法的多小波图像编码。算法将各种常用的多小波基,结合EBCOT算法,对灰度图像进行压缩编码。实验结果证明基于EBCOT算法的多小波图像编码方法能够对图像进行很好的压缩。Muhiwavelet extends traditional scalar wavelet to vector space, and includes all the advantages of the traditional wavelet. Muhiwavelet shows the perfect union of symmetry, orthogonally, finitely support and smoothness. As a result, the applications of multiwavelet in image processing have shown great potential. An image coding method of image coding based on EBCOT algorithm and multiwavelet transform was proposed. In this research, a variety of commonly-used multiwavelet bases were used to combine EBCOT algorithm. The experimental result shows that the proposed method is a valuable attempt of image compression.

关 键 词：图像编码 小波变换 多小波变换 EBCOT 

分 类 号：TN911.7[电子电信—通信与信息系统]