Remote sensing image multi-feature fusion based on NSCT transform and compressive sensing

For the remote sensing image fusion algorithm based on non-subsampled Contourlet transform (NSCT) calculation complexity and cannot extract details from source images effectively. So a new remote sensing image multi-feature fusion algorithm based on NSCT transform and compressive sensing is proposed. Firstly, luminance component V of multi-spectral image is extracted via HSV transform, and decompose the luminance component V and PAN image by NSCT; Then, Apply PCNN-based fusion rules to fusing low frequency sub-band, and the equivalent number is utilized as the linking strength; For the high frequency sub-band, a fusion rule based on the multi-feature in compressive sensing domain is presented; Finally, the fused images are obtained by inverse NSCT transform and inverse HSV transform, respectively. The experimental results show that the proposed fusion algorithm substantially outperforms the best-known remote sensing image fusion algorithms in terms of calculation complexity and the quality of the fused image, and has the better performance in visual effect and objective evaluation metrics