A Cortex Transform Based Image Fusion Method

An image fusion method based on cortex transform is proposed.It simulates the response characteristics of simple cells in biological visual cortex,and decomposes one image into sub-images with different resolutions and orientations in which every pixel represents the response of some neuron in the visual cortex.This method is invertible through simple expansion and additional operations.The results show that when evaluated with the indexes of cross entropy and contrast,it performs better than the wavelet transform based fusion method with the fusion rules of maximum absolute coefficient and maximum directive contrast.Furthermore,it is more visually pleasing for some source images.