Image defogging algorithm based on metaheuristic dark channel prior

Abstract

Sky regions pose a significant challenge for defogging, often causing issues such as halos and color distortion. To address these issues, this paper proposes a novel defogging algorithm that integrates sky segmentation with an improved Red-billed Blue Magpie Optimization (RBMO) algorithm. Firstly, the sky region of the foggy image is obtained using a sky segmentation algorithm. Secondly, the principal component analysis is used to extract sky and non-sky regions features as weight parameters for fusion to calculate atmospheric light. Meanwhile, an improved RBMO algorithm is utilized to identify the ideal fusion weight parameters for calculating transmission. Finally, an enhancement step is applied to improve the visual quality of the defogged image. Experimental results on both synthetic and real-world datasets demonstrate that our proposed algorithm achieves superior performance in key evaluation metrics. This indicates its effectiveness in accurately segmenting sky regions, suppressing halos and artifacts, enhancing overall defogging quality, and preserving image details.