Detail Recovery and Color Enhancement for Single Image Dehazing

Abstract

Due to the complexity of taking photos on foggy days, images captured in the real world are prone to problems of detail loss, color dimness, and low saturation. Considering image dehazing as a problem of detail recovery and color enhancement, a two-stage progressive fusion method of single image dehazing is proposed, which consists of a detail recovery branch, a color enhancement branch, and a wavelet fusion module. Firstly, to solve the problem of hazy image details and detail loss after image defogging, we introduce gamma transform, block homomorphic filtering, and multi-scale exposure fusion in the detail recovery branch for the first fusion stage to restore image details. Secondly, aiming at the problem that the overall color of the image is dark and the image saturation is poor after defogging, the color-preserving CLAHE(limited contrast adaptive histogram equalization) of contrast enhancement method is applied in the color enhancement branch to enhance the color. Finally, to fully use the complementary information between the two enhanced images, a wavelet fusion module is designed for pixel-level fusion in the second stage, which can recover the details and improve color information while removing the fog to the greatest extent. The experimental results show that the image's color can be restored naturally, and the details can be preserved completely. The experimental results show that the image's color can be restored naturally and the details can be fully preserved.