Pyramid Fusion Dark Channel Prior for Single Image Dehazing
This work addresses image quality enhancement for computer vision applications by providing an incremental improvement to a well-known prior-based method.
The paper tackles the problem of single image dehazing by proposing the pyramid fusion dark channel prior (PF-DCP), which improves upon the Dark Channel Prior by using multi-scale image pyramids to avoid patch size selection issues, resulting in a high-quality haze-free image with fewer artifacts and outperforming traditional methods while matching deep learning approaches on the RESIDE SOTS dataset.
In this paper, we propose the pyramid fusion dark channel prior (PF-DCP) for single image dehazing. Based on the well-known Dark Channel Prior (DCP), we introduce an easy yet effective approach PF-DCP by employing the DCP algorithm at a pyramid of multi-scale images to alleviate the problem of patch size selection. In this case, we obtain the final transmission map by fusing transmission maps at each level to recover a high-quality haze-free image. Experiments on RESIDE SOTS show that PF-DCP not only outperforms the traditional prior-based methods with a large margin but also achieves comparable or even better results of state-of-art deep learning approaches. Furthermore, the visual quality is also greatly improved with much fewer color distortions and halo artifacts.