Experimental evaluation demonstrates enhanced image clarity and a 5.66-fold speedup in hazy scenes, highlighting viable real-time preprocessing for computer vision.
Many computer vision applications demand the quality input image to act further. However, if there is haze, the image gets deteriorated. Therefore, it needs to be dehazed before being used as an input. In this article, we propose a Hash-Based-Single-Image-Dehazing (HBSID). It utilizes the Atmospheric Scattering Model (ASM). The depth of the pixels is identified using the DepthAnything model. It estimates the transmittance using hashing. Comprehensive experiments demonstrate the efficacy of the proposed modules. The average PSNR value improves from 135.19% (maximum) to 2.03% (minimum). Similarly, the average SSIM value is improved between 41.29% (maximum) and 1.33% (minimum). The NIQE and BRISQUE values are also reduced to 8.73% and 11.75%, respectively. Further, the processing speed is noticed 5.66 times faster compared to the model without the hashing module. It will be highly useful in providing a clean input to computer vision tasks, even in a hazy environment.
No takes yet. Share an insight, caveat, or question.
Bhatia et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: