Image Dehazing Using Near-Infrared Information Based on Dark Channel Prior

Images collected in haze or other weather have problems of quality degradation, blurred details, and low recognition, which seriously affect the application and development of related image application fields. The dark channel prior method is the typical method of image dehazing, but it is not suitable for sky region. The wavelength of near-infrared is longer than that of visible light, which makes it more penetrating, less affected by the scattering of suspended particles in the air, and carries more detailed information. This paper uses near-infrared information to distinguish between sky region and non-sky region, improve the dark channel prior method, enhance the visibility of the image, and restore the contrast of the image. A large number of experiments have proved that our method has achieved satisfactory results in solving problems such as sky region and darker scene. This method has strong pertinence and stability for image dehazing, and the effect is relatively natural and efficient. Therefore, our method is better than the existing method of image dehazing. (C) 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the scientific committee of the International Conference on Identification, Information and Knowledge in the internet of Things, 2020.