Guided Real Image Dehazing using YCbCr Color Space

2024-12-23

AI Summary

To address two key challenges in real-scene image dehazing—weak texture representation in the RGB space and scarcity of ground-truth paired data—this paper proposes the Structure-Guided Dehazing Network (SGDN). Methodologically, it introduces a novel YCbCr color-space guidance mechanism that leverages strong structural priors to supervise RGB-domain feature restoration; designs a Bichromatic Guidance Bridge (BGB) to fuse phase-aware information with interactive attention; and incorporates a Chroma Enhancement Module (CEM) to improve chrominance fidelity. Furthermore, we construct RW²AH—the first real-world dehazing dataset featuring geographic and climatic diversity, with rigorously pixel-aligned hazy–clean pairs. Extensive experiments demonstrate that SGDN achieves significant improvements over state-of-the-art methods across multiple real-world haze and smoke datasets. Both the source code and the RW²AH dataset are publicly released.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Structured Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Image dehazing, particularly with learning-based methods, has gained significant attention due to its importance in real-world applications. However, relying solely on the RGB color space often fall short, frequently leaving residual haze. This arises from two main issues: the difficulty in obtaining clear textural features from hazy RGB images and the complexity of acquiring real haze/clean image pairs outside controlled environments like smoke-filled scenes. To address these issues, we first propose a novel Structure Guided Dehazing Network (SGDN) that leverages the superior structural properties of YCbCr features over RGB. It comprises two key modules: Bi-Color Guidance Bridge (BGB) and Color Enhancement Module (CEM). BGB integrates a phase integration module and an interactive attention module, utilizing the rich texture features of the YCbCr space to guide the RGB space, thereby recovering clearer features in both frequency and spatial domains. To maintain tonal consistency, CEM further enhances the color perception of RGB features by aggregating YCbCr channel information. Furthermore, for effective supervised learning, we introduce a Real-World Well-Aligned Haze (RW$^2$AH) dataset, which includes a diverse range of scenes from various geographical regions and climate conditions. Experimental results demonstrate that our method surpasses existing state-of-the-art methods across multiple real-world smoke/haze datasets. Code and Dataset: extcolor{blue}{url{https://github.com/fiwy0527/AAAI25_SGDN.}}

Problem

• Dehazing
• RGB Color Mode
• Machine Learning

Innovation

• SGDN
• YCbCr color space
• Dehazing performance