AOD-Net: All-in-One Dehazing Network

TLDR

This work introduces AOD‑Net, an end‑to‑end convolutional neural network that directly produces a clean image from a hazy input and can be embedded into higher‑level vision models. AOD‑Net is built on a reformulated atmospheric scattering model and uses a lightweight CNN to generate the clean image without separately estimating transmission or atmospheric light. Experiments on synthetic and real hazy images show that AOD‑Net outperforms state‑of‑the‑art methods in PSNR, SSIM, and visual quality, and its integration with Faster R‑CNN yields a significant boost in object detection performance on hazy scenes.

Abstract

This paper proposes an image dehazing model built with a convolutional neural network (CNN), called All-in-One Dehazing Network (AOD-Net). It is designed based on a re-formulated atmospheric scattering model. Instead of estimating the transmission matrix and the atmospheric light separately as most previous models did, AOD-Net directly generates the clean image through a light-weight CNN. Such a novel end-to-end design makes it easy to embed AOD-Net into other deep models, e.g., Faster R-CNN, for improving high-level tasks on hazy images. Experimental results on both synthesized and natural hazy image datasets demonstrate our superior performance than the state-of-the-art in terms of PSNR, SSIM and the subjective visual quality. Furthermore, when concatenating AOD-Net with Faster R-CNN, we witness a large improvement of the object detection performance on hazy images.