改进投影梯度NMF的NSST域多光谱与全色图像融合

Abstract

For the effective combination of spectral information of multispectral image with spatial detail information of panchromatic image, and further improvement in the quality of fused multi-spectral image, a fusion method of multispectral and panchromatic images based on improved projected gradient non-negative matrix factorization(NMF)and improved pulse coupled neural network(PCNN)in non-subsampled shearlet transform (NSST)domain is proposed. Intensity hue saturation (IHS) transform is performed for multispectral image. The histogram of panchromatic image is matched to the histogram of intensity component of multispectral image, and the contrast of panchromatic image is enhanced. The intensity component of multispectral image and panchromatic image is decomposed by NSST, respectively. The low frequency image is obtained by fusion of two low frequency coefficients using improved projected gradient NMF. The spatial information of the fused image is further improved. For the fusion of high frequency sub-band coefficients, improved PCNN is adopted to enhance the image details. The fused intensity component is reconstructed by inverse NSST and the final fused image is obtained by inverse IHS transform. A large number of experimental results show that the proposed method enhance the spatial details in the fused image, while preserving spectral information of the multispectral image. And it is superior to the existing fusion methods such as the method based on IHS transform, the method of non-subsampled Contourlet transform (NSCT) combined with PCNN, and the method of NSCT combined with NMF. ©, 2015, Chinese Optical Society. All right reserved.