Comprehensive morphology and non-separable wavelet edge protection image fusion algorithm

In view of this situation that image fusion algorithm of separable wavelet is unable to retain the fusion image edge information effectively, taking the advantage of non-separable wavelet transformation and morphology merit, an image fusion algorithm that emphasizing edge preserving is proposed. Under the non-separable wavelet decomposed frame, this algorithm uses the fusion rules that combine morphological edge detection maximum value and the variance weighting analysis for the high frequency component, use local energy weighting method to carry on the fusion for the low frequency component. Finally through an inverse transformation of non-separable wavelet reconstructs fusion image. The experimental results indicate that, the fusion image obtained by the proposed algorithm can better reappear the edge information of the image, fuse the mutual complement information between the images organically, maximum reflect the characteristic information of source images, provide strong support for the future information analysis and extraction capacity of the images.