Underwater Image Dehazing Using a Novel Color Channel Based Dual Transmission Map Estimation

Underwater images typically suffer from color distortion and low contrast, owing to the light absorption and scattering. To handle the visual manifestation of such degraded images, numerous underwater image dehazing algorithms have been presented. However, most of existing dehazing methods still have room for improvement in terms of preserving more details in the restored results. In this paper, we propose a simple yet effective dehazing method. The core idea is to produce the high-quality underwater images with rich detail information and vivid color. Firstly, an innovative color correction is designed to compensate the information of each color components. This operation is a pre-processing procedure, in which, selective absorption is fully considered. Then a dual transmission map-based haze removal method is introduced. Different from previous methods, a novel color channel with two terms is constructed to accurately estimate the transmission maps. The one is designed as the sharpened term to reveal more image detail and edges. The other is a difference of channel intensity prior term to remove the influence of light scattering. By using this strategy, our method can generate a natural appearance of the restored image with more detail information and higher color contrast. Experiments on representative images have proven that the performer of our method is 8.6% and 2.7% better than the second best on the average scores of patch-based contrast quality index (PCQI) and Entropy metrics, respectively.