Locally adaptive contrast enhancement using convolutional neural network

Abstract

Law contrast images obtained from smart phones or any imaging devices are difficult to process by image processing systems and general users. Therefore, an enhancement of low contrast image is an important task in image processing and for device users. In this paper, we propose a method to detect low contrast regions using CNN(Convolutional Neural Network) and to improve the image quality by using chromatic contrast weight. Experiments show that the proposed method reduces over enhancement, while recovering details of low contrast regions. © 2018 IEEE.