Fusion of Remote Sensing Image with Compressed Sensing Based on Wavelet Sparse Basis

Because of its compressive sample feature that the sampling rate is far lower than the Nyquist/Shannon, a large number of sampled data are reduced by compressed sensing. In this paper, an efficient image fusion framework is proposed based on compressed sensing. This method firstly extracts panchromatic image and multispectral image separate R, G, B components, secondly wavelet transform is performed on these components, thirdly, compressed sensing domain data are got by using the Gaussian random matrix to sample the sparse data. Fourthly, the compressed data are fused by taking different weights. Finally, the fusion image is reconstructed by OMP algorithm. The experimental results prove that the less data needed to be processed and show the better fusion effect than the traditional methods.