High-altitude remote sensing image dehazing via transmission optimization and detail enhancement
Guohou Li, Kaijie Jin, Ying Zheng, Junpeng Xu, Wenyi Zhao, Weidong Zhang · Remote Sensing Letters · 2025
Aerosols and water vapour in the atmosphere cause scattering and absorption of light during propagation, making remote sensing images hazy, blurred and low contrast. To address these issues, we propose a high-altitude remote sensing image dehazing method via transmission optimization and detail enhancement, termed TODE. We first apply transmission optimization-based image dehazing to refine the input image, improving the accuracy of the transmission map and enhancing global contrast to obtain an initial dehazed image. Meanwhile, a contrast enhancement method with adaptive atmospheric light estimation is employed to optimize local brightness and accurately compute the atmospheric light component, avoiding bright spot interference and producing a contrast-enhanced image. Finally, a multi-layer advantage fusion strategy is applied to decompose and integrate multi-level features from the images, achieving superior visual quality. Experiments show that the proposed TODE method maintains structural details and improves the clarity of remote-sensing images.