Medical image fusion method based on wavelet multi-resolution and entropy

After studying the characteristics of DWT and reflecting the statistical features by information entropy, we propose a medical image fusion method on the basis of wavelet entropy decomposition transform. Firstly, we use wavelet transform to decompose the multi-resolution signal from each image. Then we adapt different rules to fuse sub-images with different bandwidths at each resolution level. Finally we could gain fused image through the inverse transform. The method by calculating wavelet entropy can discover slight changes of signal, keep details of tiny regions in the procedure effectively. It also provides favorable evidence for doctors to detect pathological changes in early stage. The experimental results have proved that it is an efficient way to make fused image more adaptable to human vision.