A variational framework for single low light image enhancement using bright channel prior

Xueyang Fu, Delu Zeng, Yue Huang, Xinghao Ding, Xiao-Ping Steven Zhang

Abstract: Low light image enhancement is prerequisite in many fields, such as surveillance systems, road safety and waterway transport. In this paper, a new variational framework using bright channel prior is proposed to address the low light image enhancement problem within a single image. An alternating direction optimization method is employed to solve the variational problem. Experiment results show that the new method can better eliminate the black halo and suppressing the issues of over-enhancement and color distortion when compared with other existing methods.