Hailiang Shi and Xiangjun Xin
Remote sensing, PCA, NSCT, correlation coefficient, local gradient, Sobel
A new fusion method is proposed for panchromatic (PAN) and multispectral (MS) remote sensing images based on nonsubsampled contourlet transform (NSCT) and principal component analysis (PCA). The method applies PCA to MS first to get the first principal component, and then performs NSCT on this component and PAN to obtain approximation and detail coefficients. For the low-frequency coefficients, to balance the spatial and spectral information PCA is used again; for the subbands coefficients, correlation coefficient and local Sobel gradient are used. Experimental results demonstrate that this method can improve the spatial quality while preserving spectral information effectively, and it is superior to several traditional methods including the PCA- and discrete wavelet transform-based methods.
Important Links:
Go Back
