基于FastICA的高光谱图像压缩技术

提出了一种基于快速独立分量分析(FastICA)的高光谱图像压缩算法。首先引入虚拟维数算法估计图像中的目标端元个数,进而提取出感兴趣的目标端元矢量,并初始化快速独立分量分析的混合矩阵;利用最小噪声分量变换对原始数据进行降维,从降维后的主分量中提取独立分量;对独立分量进行恒虚警率检测与形态学滤波,实现目标分割。对高光谱图像进行谱间Karhunen-Loeve变换,利用比例位移法对感兴趣目标的小波系数进行提升,最后对各主分量进行最优码率的SPIHT压缩。实验结果表明,该方法在获得较高压缩性能的同时能够有效地保留感兴趣的目标。

Compression Technique for Hyperspectral Imagery Based on FastICA

Efficient compression for hyperspectral imagery has been the research focus for the developmentof remote sensing technique. The small targets information protection during the compression process without anypreknowledge should be necessarily considered. This paper presents a new lossy compression method forhyperspectral imagery based on fast independent component analysis (FastICA). Virtual dimensionality isintroduced to determine the number of target endmembers. The mixing matrix of FastICA is initialized by targetendmembers. Minimum noise fraction is employed for dimensionality reduction of original data volumes, andFastICA is performed on the selected principal components to generate independent components. Then, constantfalse alarm rate detection is performed on each IC, which is followed by morphologic filtering. Karhunen-Loevetransform is used to decorrelate the spectral redundancy, general scaling-based method is selected to upshift thewavelet coefficients of interested targets. Finally, each principle component is allocated optimal rate andcompressed by SPIHT algorithm. Experimental results on AVIRIS data show that the proposed method not onlyprovides high compression performance, but also preserves targets interested effectively.