Design of wavelet based image fusion using A Trous algorithm

Image fusion presents a technique based on wavelet fusion for change detection in Synthetic Aperture Radar images. The image fusion technique is introduced to generate a difference image by using complementary information from a mean-ratio image and a log-ratio image. Our objective is to analyse the performance of the A Trous algorithm for Synthetic Aperture Radar Images. In our work, an improved additive-wavelet fusion method is presented using the A Trous algorithm and it is modified to suit the Grey scale Multispectral Images. This method is applied to the wavelet coefficients that are formulated by the mean ratio and log ratio difference images and produce the fused image. To analyse the performance of the fused image obtained above two methods information entropy, correlation coefficient and Root Mean Square Error are computed and quantitative analysis is done for two set of input source images.