Perceptual-Based Fusion of Infrared and Visual Images in Nonsubsampled Contourlet Transform

Abstract

In this paper, a perceptual-based fusion method for infrared and visual images is proposed. Many methods based on gradient and energy have been used in the literature, but these methods are not considered the human visual system (HVS) and their experiment results are less than ideal. So a method depending on contrast sensitivity function (CSF) based on the respond of HVS as a criteria to determine the salient features of infrared and visual images to decide how to fusion is proposed. In this method Fourier transform is taken, which is sensitive to contrast change and is as an important criteria to form a salient metric. Comparing with other fusion methods, experiment results also prove that the proposed method is more outstanding.