A novel fuzzy approach for low contrast color image enhancement

Low contrast affects color images which are captured and transferred digitally. To tackle this challenge, the contrast must be improved with the least amount of information loss possible, so that the enhanced images may be used in both human visual systems and automated systems. The paper introduces the LCFIE framework, which uses fuzzy set theory to increase the color images’ contrast. It automatically recognizes the images that need to be enhanced and classifies them as dark, bright, or pleasant. Fuzzification and membership value modification are accomplished using a modified Gaussian function and a Sigmoid function, respectively. The required parameters are optimized by dividing the optimization problem into single-variable optimization problems, which take less time to solve. The parameters have been chosen to ensure that no information is lost. Observers’ Mean Opinion Score is utilized to grade the visual quality of images. The image quality is quantified using the mean, standard deviation, colorfulness index, fitness function, NR-CDIQA, and CQE. Extensive experiments revealed the supremacy of the proposed method in increasing the contrast of the image, both in qualitative and quantitative terms.