Multi-Branch Non-Homogeneous Image Dehazing via Concentration Partitioning and Image Fusion

TL;DR

This paper introduces CPIFNet, a multi-branch deep neural network that effectively dehazes non-homogeneous images by decomposing the problem into homogeneous sub-problems and intelligently fusing the results.

Contribution

The paper proposes a novel multi-branch neural network framework that decomposes non-homogeneous dehazing into homogeneous sub-problems and fuses the results for improved performance.

Findings

CPIFNet outperforms existing methods on non-homogeneous haze images.

The two-stage architecture effectively handles spatially varying haze concentrations.

The comprehensive loss function improves dehazing quality across multiple metrics.

Abstract

Existing single image dehazing methods have demonstrated satisfactory performance on homogeneous thin-haze images; however, they often struggle with non-homogeneous hazy images that exhibit spatially varying haze concentrations and abrupt density transitions across different regions. To address this fundamental limitation, we propose a novel multi-branch deep neural network framework, termed Concentration Partitioning and Image Fusion Network (CPIFNet), which decomposes the challenging non-homogeneous dehazing problem into a set of tractable homogeneous sub-problems. Our key insight is that a single non-homogeneous hazy image can be viewed as a composite of multiple local regions, each exhibiting approximately homogeneous haze characteristics. CPIFNet employs a two-stage architecture consisting of an Image Enhancement Network (IENet) stage and an Image Fusion Network (IFNet) stage. In the first stage, multiple IENet branches are independently trained on homogeneous haze datasets of different concentration levels, producing enhancement models that excel at restoring regions matching their respective haze densities. In the second stage, the IFNet intelligently aggregates the advantageous regions from all enhancement outputs through deep feature stacking and merging, yielding a unified high-quality dehazed result. Furthermore, we introduce a comprehensive loss function incorporating reconstruction, perceptual, structural, and color losses to jointly supervise both stages.