Single image deraining via decorrelating the rain streaks and background scene in gradient domain. (July 2018)

• Record Type: Journal Article
• Authors: Du, Shuangli; Liu, Yiguang; Ye, Mao; Xu, Zhenyu; Li, Jie; Liu, Jianguo
• Is Part Of: Pattern recognition. Volume 79(2018:Jul.)
• Page Start: 303
• Page End: 317
• Publication Date: 2018-07
• Subjects: Rain removal -- Gradient domain -- Decomposition model; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2018.02.016
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: We propose a single image de-raining method in gradient domain based on the observation that the influence of rain streaks on X- and Y- gradients shows different statistical properties. A rain-free direction is first proposed, along which a rain image or a block is least-affected in gradient domain. It is effective in recovering image texture details. To extract the rain-free gradient along the direction perpendicular to the rain-free one, a novel decomposition model is proposed, which is formulated as an objective function with three terms: an anisotropic total variation term, a low-rank constraint and a de-correlation term. Abstract: Single image based rain removal is very challenging due to the lack of temporal and context information, and the existing techniques are usually unpractical in real-time applications as they are time-consuming, and make images blurred in varying degrees. To tackle this issue, this paper proposes a novel framework, based on a new observation that the background has a reasonably low correlation with rain streaks in gradient domain. The framework mainly contains three steps: 1) a rain-free direction with respect to a rain image or a block therein is proposed, describing the fact that there exists a direction along which the image is least-affected in gradient domain; 2) by combing total variation, low-rank constraint and a de-correlation term, a novel decomposition model is proposed to explicitly extract the rain and rain-free gradient components along the direction perpendicular to the just calculated rain-free direction; 3) the rain-free image is reconstructed using Poisson equation, which effectively resists the sparse noise contained in gradients. The favorable performance of the proposed framework has been confirmed by many experimental results, and especially the computational complexity is low.