A cascaded approach for image defogging based on physical and enhancement models

In this paper, we propose a novel cascade strategy approach for visibility restoration in foggy images. The proposed cascade strategy is based on the combination of enhancement and physical models, the contrast limited adaptive histogram equalization (CLAHE) and no-black pixel constraint with planar assumption (NBPC \(+\) PA) methods. The use of CLAHE enhances the visibility of foggy image, but it produces color and edge distortion, boosts noise and creates halo effects. We overcome these shortcomings of CLAHE by feeding its output to the NBPC \(+\) PA. In order to improve the cascaded performance of the two methods, we determine the suitable parameters. The proposed cascading utilizes the individual strengths of the two approaches which in turn provides better defogging results for homogeneous as well as inhomogeneous fog. We present objective quality assessment and visibility enhancement of various foggy images. The experimental results verify the enhanced defogging capabilities of the proposed cascade strategy compared to the existing defogging algorithms.