A color daytime and nighttime image fusion algorithm based on IHS and Multi-Wavelet transform

Abstract

A daytime and nighttime color image fusion method based on IHS transform and Multi-Wavelet transform is presented. The method can automatically merge images of a scene captured under different illumination and can be used for practical purposes. Specifically, the fused image can give more information of the context of night-time traffic videos, which makes them easier to be understood. The context is automatically captured by a camera in a fixed-position during daytime and then is fused into a nighttime image of the same scene. The advantages of this method are that it preserves the important local perceptual cues and it can avoid the problems with traditional methods such as aliasing, ghosting and haloing.