作　　者： ; ; ;

机构地区： 复旦大学信息科学与工程学院电子工程系

出　　处： 《复旦学报（自然科学版）》 2013年第3期371-379,共9页

摘　　要： 高光谱图像拥有巨大的数据量,造成其保存和传输非常不便,因此高光谱图像的压缩显得尤为重要.提出一种基于张量Tucker分解和小波包变换的高光谱图像压缩算法.首先,该算法利用Tucker分解的性质,充分提取高光谱图像中各个模式下的信息,并利用其中包含有空间信息的光谱模式对高光谱图像的光谱维进行解相关.然后,运用比经典Mallat小波分解更为有效的小波包变换对光谱去相关后保留下来的主成分进行JPEG2000压缩.实验结果表明,本算法压缩性能远远好于经典的三维小波算法,并且由于张量分解的应用,不论在码率失真表现还是信息保真度上,本算法均比基于一维主成分分析的高光谱压缩算法更具优势. The volumetric data of hyperspectral imagery makes it difficult for its storage and transmission,and thus its compression becomes an urgent issue in the practical applications.An efficient method for hyperspectral image compression based on Tucker decomposition and wavelet packet transform is proposed.The Tucker decomposition can reduce the spatial and spectral redundancies and extract the interactive information from different modes of the hyperspectral imagery simultaneously,while wavelet packet transform can provide more flexible and finer decomposition results on the high-frequency components of images.The spectral factor matrix of Tucker decomposition is firstly used to decorrelate the spectral dimension in the proposed algorithm;then,with JPEG2000,the more efficient wavelet packet transform is applied instead of Mallat decomposition to compress the principal components preserved after spectral decorrelation.Experimental results on real hyperspectral images demonstrate that the proposed method not only outperforms the conventional 3D wavelet-based algorithms,but also has considerable advantages over the one-dimensional principal components analysis based hyperspectral images compression algorithms on both rate-distortion performance and information preservation.

关 键 词： 高光谱图像 图像压缩 张量分析 小波包变换 分解

领　　域： [自动化与计算机技术] [自动化与计算机技术]