A single image dehazing model using total variation and inter-channel correlation

摘要

Outdoor images are often degraded by haze, causing a change of image contrast and color values. In this paper, we propose a novel variational model for the removal of haze in a single color image, by incorporating an inter-channel correlation term into the total variation based model in Wang et al. (Pattern Recognit 80:196–209, 2018). The proposed model enables both color and gray-valued transmission maps, contributing to its broad applications, and its convergence analysis is also provided. To realize the proposed model, we adopt an alternating minimization algorithm, and then the alternating direction method of multipliers is employed for solving subproblems. These result in an efficient iterative algorithm, with its convergence proven. Numerical experiments validate the outstanding performance of the proposed model compared to the state-of-the-art methods.