Robust Single Image Dehazing and Image Enhancement via Generative Adversarial Networks

Abstract

Haze removal is an essential factor where various research communities related to computer vision have conducted research with explicit outputs. Image de-hazing consumes more time, and the state-of-the-art solutions are less robust. In order to achieve this, a GAN-based approach is likely to be used, where de-hazing is performed unanimously and then retainment of the image is done accordingly.