混合PCA/ICA与JPEG2000结合的高光谱图像压缩

Hyperspectral image compression using mixed PCA/ICA in conjunction with JPEG2000

主成分分析(PCA)常常结合JPEG2000压缩标准用来对高光谱图像进行压缩。然而,由PCA得到的主成分仅利用了二阶统计信息。对于高光谱图像应用来说,只采用二阶统计信息是远远不够的,如异常像素的处理常常需要用到更高阶的统计信息。研究了一种混合PCA/ICA与JPEG2000相结合的高光谱图像压缩算法。首先,对原始高光谱图像进行PCA变换,提取出前m个主成分对应的特征向量矩阵WPCA;然后,对其余的特征向量进行ICA变换,得到n个特征向量矩阵WICA;最后,将得到的混合投影矩阵、原始高光谱图像及其均值向量共同嵌入JPEG2000比特流,从而完成对高光谱图像的压缩。在不同码率的情况下,通过空间相关系数(ρ)、信噪比(SNR)、光谱角填图(SAM)等技术指标对混合PCA/ICA+JPEG2000算法的压缩性能进行评估。实验结果表明,混合PCA/ICA+JPEG2000算法不但能有效去除高光谱图像的谱间相关性,而且能够有效提高光谱保真度,保护异常像素信息。

The principal component analysis （PCA） method combined with JPEG2000 is widely used in hyperspectral image compression. However, the covariance matrix of the PCA only represents the second order statistics. In many applications of hyperspectral image analysis, only preserving the infor- mation of the second order statistics is not sufficient. Taking the anomalous pixels for example, more subtle information needs to be captured by using higher-order statistics. To solve this problem, we pro- pose a hyperspectral image compression algorithm using mixed PCA/ICA in conjunction with JPEG2000 standard. Firstly, the PCA is performed for the original hyperspectral image to find the eigenvector ma- trix WpCA corresponding to the first rn largest eigenvalues. Then, the ICA is employed for the remaining eigenvectors to find n eigenvector matrix W_ICA . Finally, the mixed projection matrix, original hyperspec- tral image and it＇s mean vectors are embedded into the JPEG2000 bitstream for compression. At different bit rates, the performance of the mixed PCA/ICA＋JPEG2000 is evaluated by spatial correlation coeffi- cient （ρ）, signal noise ratio （SNR） and spectral angel map （SAM）. Experimental results reveal that the proposed algorithm is not only better in spectral correlation reduction, but also can improve spectral fi-delity and protect anomaly pixel information.