Image Generation with Smooth Gradations from Plural Gray-scaled Images Using Genetic Algorithm

Abstract

This paper proposes a method to generate an output image with good contrast and smooth gradations from plural gray-scaled input images that have different exposures. The proposed method generates an output image with a wide gray-scaled dynamic range and delicate gradations from an accumulated image of plural input images using the genetic algorithm. The chromosome of an individual is represented by the array of genes that have all gray-scales included in the accumulated image. The fitness of each individual is evaluated by the liner differential value that is obtained in the output image. The experimental results show that the output images with delicate and smooth gradations were generated by the proposed method. The qualities of the output images were also evaluated using the method based on the delta-histograms. The output images by the proposed method recorded the good sharpness factors.