A Hybrid Approach with CLAHE and Dark Channel Prior for Enhancing Underwater Images

The underwater images (UIM) are rigorously degraded by colour deviations, scattering, and absorption. This can be addressed by enhancing the UIM. Prior knowledge of the degradation model is combined with enhancement methods to achieve superior performance. We have used a hybrid approach that incorporates the White Balancing (WB) technique for colour correction, CLAHE for enhancing contrast, and the Dark Channel Prior (DCP) method to decrease haziness in the image. Firstly, the UIM is colour corrected using WB. CLAHE is applied on a white-balanced image. DCP is also applied on white-balanced images to remove haziness. Finally, CLAHE output and DCP output are fused at a multi-scale by considering different weight maps to obtain the enhanced UIM. The U45, RUIE, and UIEB datasets are used to validate the effectiveness of the devised technique. A comparison of the developed technique with current methods reveals that the developed technique is superior.