Comparisons of Several Pixel-Level Image Fusion Schemes for Infrared and Visible Light Images

Image fusion provides a useful tool to integrate multiple images into a composite image that is more suitable for the purposes of human visual perception and can help us to extract more features. In this paper we compare the wavelets-based schemes which are based on ordinary wavelet transform, wavelet packet analysis and the nonlinear fusion scheme. It is concluded that the nonlinear fusion scheme has the best performance and more wavelet decomposition levels will give better results during the fusion process