A constrained total variation model for single image dehazing. (August 2018)

• Record Type: Journal Article
• Authors: Wang, Wei; He, Chuanjiang; Xia, Xiang-Gen
• Is Part Of: Pattern recognition. Volume 80(2018:Aug.)
• Page Start: 196
• Page End: 209
• Publication Date: 2018-08
• Subjects: Dehazing -- Total variation -- Variational method -- Gradient projection algorithm; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2018.03.009
• Languages: English
• ISSNs: 0031-3203

Abstract

Abstract : highlights: Propose a new formulation to describe a hazy image by combining the Koschmieder's law and the Retinex theory. Propose a variational model to convert the problem of estimating the depth of scene to a constrained minimization problem. Prove the existence and uniqueness of solution of the proposed model; Develop an algorithm for numerical solution of our model by combining alternating minimization with fast gradient projection. Experiments show that our model has the best visual effect and the highest average PSNR compared to six relevant models in the literature. Abstract: Haze removal (or dehazing) is very important for many applications in computer vision. Because depth information and atmospheric light are usually unknown in practice, haze removal is a challenging problem, especially for single image dehazing. In this paper, we propose a new variational model for removing haze from a single input image. The proposed model combines Koschmieder's law with Retinex assumption that an image is the product of illumination and reflection. We assume that scene depth and surface radiance are spatially piecewise smooth, total variation is thus used for regularization in our model. The proposed model is defined as a constrained optimization problem, which is solved by an alternating minimization scheme and a fast gradient projection algorithm. Theoretical analyses are given for the proposed model and algorithm. Some numerical examples are presented, which have shown that our model has the best visual effect and the highest average PSNR (Peak Signal-to-Noise Ratio) compared to six relevant models in the literature.