A novel fusion scheme for infrared and visual images based on wavelet and color transfer algorithm

Abstract

The pseudo-color processing for infrared (IR) image is useful for object detection and tracking. Image fusion of IR and visual images is an effectual scheme for pseudo-color processing. But in some conditions, we only have the IR images of the scene. How to perform a pseudo-color processing for these IR images is a difficult and interesting problem. In the paper, a novel image fusion method based on wavelet and color transfer is proposed. With wavelet and color transfer, the chromaticity values of the color visual image are assigned to the IR image. By the method, the IR and visual images can be fused even if the scenes of two images are not the same one. The only requirement of the method is that the compositions of the source and target scenes resemble each other. The experimental results show that the algorithm can give IR image a natural color appearance. Such a full color representation of nighttime scenes may be of great ergonomic value by making the interpretation of the displayed scene easier (more intuitive) for the observer.