Dehazing for Real-World Scenarios through Adaptive Perturbation Injection and Local-Global Dual-Stage Defense

Recovering high-quality clear images from hazy images is a core task in the field of image processing. Although existing dehazing models have achieved success on synthetic hazy images, they still face significant challenges when dealing with hazy images in real-world scenarios. To address this issue, this paper proposes a real-world image dehazing network with adaptive perturbation injection and dual-layer defense mechanisms. The network primarily consists of a perturbation generation module, a local-global dual-layer defense module, and a semantic guidance module. Specifically, the perturbation generation module adaptively generates perturbation cues to simulate the degradation factors of real-world hazy image generation. These perturbations are then injected into synthetic hazy images to reduce the distribution gap between synthetic and real-world hazy images. The local-global dual-layer defense module mitigates the impact of perturbations in hazy images at both local and global levels, enhancing the dehazing performance in real-world scenarios. However, the dual-layer defense mechanism may inadvertently degrade useful features in the image during dehazing. To address this, a semantic guidance module is introduced to incorporate semantic information into the image reconstruction process, further improving the quality of the dehazed results. Experimental results demonstrate that the proposed method outperforms existing methods in both quantitative metrics and visual quality on real-world hazy images. The code for the method in this paper can be downloaded from https://github.com/songshuaitian/API-Net.