The Research Of Mutli-exposure Color Image Fusion Approach

Image fusion is process of combining two or more images into an image. It can extractfeatures from source images, and provide more information than one image can. This fusion process is helpful for scene understanding not only by humans but by computer vision systems.In this paper, we proposed a method based on genetic algorithms (GA) for fusing multi-exposure images. It partitions the image domain into uniform blocks and for each block selects the image that contains the most information within that block. The selected images are then biended iogether using rationai Gaussian blending functions that are ceniered at the blocks. The choice both of the block size and width of the blending functions will influence the result of fusion image, so we employ GA for optimizing the parameters which are the block size and width of the blending functions. In order to confirm the evaluate function of the best fusion image that we carried on a great deal of data analysis from many facets of the information content and visual effect etc. The evaluation function can carry on regulate according to the different purpose with the fusion images application, making the textual method get a more extensive application. The experimental results show that proposed algorithms have good effect in dealing with the images in different scene irradiance background or multi-exposure images.