A Novel Image Fusion Method Based on Contourlet Decomposition

Since multisensor image fusion is becoming a promising technique, many image fusion techniques based on the multiresolution wavelet decomposition have been developed. However, wavelets have been proved to have disappointing behaviors in higher dimensions. So, we proposed a novel multisensor image fusion scheme based on contourlet transform, which can capture the intrinsic geometrical structure that is key in visual information. We have estimated the validity of the fusion method by analyzing, visually and quantitatively, and the approaches proposed in this paper result in fused images with improved quality with respect to those obtained by wavelet-based fusion methods.