High Dynamic Range Reduction Via Maximization of Image Information
Arthur Ardeshir Goshtasby · 2003
An algorithm for blending multiple-exposure images of a scene into an image with maximum information content is introduced. The algorithm partitions the image domain into subimages and for each subimage selects the image that contains the most information about the subimage. The selected images are then blended together using monotonically decreasing blending functions that are centered at the subimages and have a sum of 1 everywhere in the image domain. The optimal subimage size and width of blending functions are determined in a gradient-ascent algorithm that maximizes image information. The proposed algorithm reduces dynamic range while preserving local color and contrast.