Remote sensing image fusion based on Curvelet transform

Compared with wavelet, Curvelet has much better identification ability of directions, so it is more appropriate for the analysis of the image edges such as curve or line characters than wavelet. When introducing curvelet transform to remote sensing image fusion, we can take the characteristics of original images well, The ability of noise suppression is also better than wavelet transform. So the method of curvelet transform based image fusion is proposed. Firstly, the original images are decomposed using curvelet transform, then the coefficients are fused with the different fusion rules in the different frequency bands, finally the fused coefficients are reconstructed to obtain fusion results. Mean square error, difference coefficient etc. are used to evaluate the results. The fusion images are compared with that based wavelet transform. The results show that this method can get much better fusion results than wavelet in two respects: preserving original images' important information and noise suppression.