Image fusion system and method for parallelization optimization of image fusion algorithm

An image fusion system and a method for parallelization optimization method of an image fusion algorithm are provided. After different images of the same scene are aligned to form a superimposed image, the superimposed image is filtered to obtain a feature image. Color depths of a plurality of pixel points in each of a plurality of pixel units of the feature image are compared to each other to determine a maximum color depth among the color depths in each of the plurality of pixel unit. When a relationship between the color depth of any one of the plurality of pixel points and the maximum color depth in the same pixel unit does not meet a preset relationship, the color depth of the one of the plurality of pixel points is adjusted according to the maximum color depth thereof. Finally, a fused image is generated according to the adjusted feature image.