A PDE-Based Image Dehazing Method via Atmospheric Scattering Theory

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hu, Liubing, Wang, Pu, Gao, Guangwei, Wang, Chunyan, Zheng, Zhuoran
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908588561137664
author Hu, Liubing
Wang, Pu
Gao, Guangwei
Wang, Chunyan
Zheng, Zhuoran
author_facet Hu, Liubing
Wang, Pu
Gao, Guangwei
Wang, Chunyan
Zheng, Zhuoran
contents This paper introduces a novel partial differential equation (PDE) framework for single-image dehazing. We embed the atmospheric scattering model into a PDE featuring edge-preserving diffusion and a nonlocal operator to maintain both local details and global structures. A key innovation is an adaptive regularization mechanism guided by the dark channel prior, which adjusts smoothing strength based on haze density. The framework's mathematical well-posedness is rigorously established by proving the existence and uniqueness of its weak solution in $H_0^1(Ω)$. An efficient, GPU-accelerated fixed-point solver is used for implementation. Experiments confirm our method achieves effective haze removal while preserving high image fidelity, offering a principled alternative to purely data-driven techniques.
format Preprint
id arxiv_https___arxiv_org_abs_2506_08793
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A PDE-Based Image Dehazing Method via Atmospheric Scattering Theory
Hu, Liubing
Wang, Pu
Gao, Guangwei
Wang, Chunyan
Zheng, Zhuoran
Computer Vision and Pattern Recognition
Image and Video Processing
This paper introduces a novel partial differential equation (PDE) framework for single-image dehazing. We embed the atmospheric scattering model into a PDE featuring edge-preserving diffusion and a nonlocal operator to maintain both local details and global structures. A key innovation is an adaptive regularization mechanism guided by the dark channel prior, which adjusts smoothing strength based on haze density. The framework's mathematical well-posedness is rigorously established by proving the existence and uniqueness of its weak solution in $H_0^1(Ω)$. An efficient, GPU-accelerated fixed-point solver is used for implementation. Experiments confirm our method achieves effective haze removal while preserving high image fidelity, offering a principled alternative to purely data-driven techniques.
title A PDE-Based Image Dehazing Method via Atmospheric Scattering Theory
topic Computer Vision and Pattern Recognition
Image and Video Processing
url https://arxiv.org/abs/2506.08793