An Image Dehazing Algorithm for Underground Coal Mine Environments Based on Dual‐Channel Prior and Adaptive Contrast Enhancement

Authors:
Open Access
Publisher website
Google Scholar
Abstract
Underground coal mine images often suffer from severe haze, colour distortion and detail loss due to suspended particles, hindering visual clarity and safety monitoring. This paper proposes a novel dehazing algorithm tailored for this scenario, with three core contributions: 1. dual‐channel atmospheric light fusion combined with adaptive gamma correction to address uneven lighting; 2. gradient‐domain guided filtering for edge‐preserving transmittance map refinement; 3. LAB‐space contrast‐limited adaptive histogram equalisation (CLAHE) to enhance detail visibility. The method integrates bright and dark channel priors, uses quadtree‐based atmospheric light estimation, and constructs a fused transmittance map via adaptive weighting. This algorithm provides clearer, more informative imagery for real‐time coal mine surveillance, enhancing operational safety in complex underground environments. Experimental results show that compared to the dark channel prior method, the average gradient (AG), information entropy (IE), standard deviation (STD) and weighted peak signal‐to‐noise ratio (WPSNR) are improved by 45.43%, 9.53%, 25.80% and 11.71%, respectively.
Funding Information
  • National Natural Science Foundation of China (52374156)

This publication has 21 references indexed in Scilit: