Weak illumination image enhancement algorithm based on cyclic generation countermeasure network

Abstract

To solve the problem of missing or distorted detail texture when manually adjusting image parameters, a weak illumination image enhancement algorithm based on cyclic generation game network is proposed. The image features are normalized by Gaussian distribution. Combined with homomorphic filtering theory and defogging operation, the image is generated and denoised according to the network brightness to enhance the weak illumination image. The experimental results show that after using this method to process the image, the image entropy increases by 6.8%, the contrast increases by 27.5%, and the noise content decreases by 24.1%. It has better contrast. It can not only meet the enhancement effect of weak light, but also ensure the details of the image, so that the image has richer details and good visual appearance.