Journal Article

Single Image Haze Removal from Image Enhancement Perspective for Real-Time Vision-Based Systems

Kang, Bongsoon, Lee, Gi-Dong, Ngo, Tri Minh, Nguyen, Quoc-Hieu, Lee, Seungmin, Ngo, Dat

2020

Overview

Vision-based systems operating outdoors are significantly affected by weather conditions, notably those related to atmospheric turbidity. Accordingly, haze removal algorithms, actively being researched over the last decade, have come into use as a pre-processing step. Although numerous approaches have existed previously, an efficient method coupled with fast implementation is still in great demand. This paper proposes a single image haze removal algorithm with a corresponding hardware implementation for facilitating real-time processing. Contrary to methods that invert the physical model describing the formation of hazy images, the proposed approach mainly exploits computationally efficient image processing techniques such as detail enhancement, multiple-exposure image fusion, and adaptive tone remapping. Therefore, it possesses low computational complexity while achieving good performance compared to other state-of-the-art methods. Moreover, the low computational cost also brings about a compact hardware implementation capable of handling high-quality videos at an acceptable rate, that is, greater than 25 frames per second, as verified with a Field Programmable Gate Array chip. The software source code and datasets are available online for public use.

• Publisher: MDPI AG,MDPI
• Subject: adaptive tone remapping / Algorithms / Atmospheric aerosols / detail enhancement / field programmable gate array / haze removal / Light / Methods / multiple-exposure image fusion / real-time processing / Researchers