This paper introduces a contrast-based multiresolution image fusion technique.Three steps can implement the method: First
the input multispectral images are decomposed by wavelet transform
meanwhile
the multiresolution contrast sequences of each input images can be obtained.Second
according to the contrast-based criterion
the multiresolution analysis of the fused image can be obtained on the corresponding levels of the multiresolution analysis of the input images.Finally
the output image can be obtained through inverse wavelet transform.This algorithm is tested by fusing the visual and infrared images from same scence.The experiment shows that the fused image can preserve the details of the each input images successfully.