Unveiling Clarity: A Survey on Haze Removal Techniques Using Deep Learning Approaches

Image dehazing remains a continuously ever-evolving subject of study in computer vision. The existence of airborne elements in the environment, such as haze, fog, and mist, can substantially degrade the quality of pictures. Furthermore, applying these dehazing approaches is extremely significant in various domains, including urban mobility, motion analysis, sight surveillance, visual data manipulation, machine perception, open-air visual capture, object identification, and entity recognition. Therefore, effective haze removal methods are essential for obtaining clear images. Consequently, numerous dehazing approaches have emerged over the last few decades. This review aims to provide invaluable insights for emerging researchers. It also meticulously explores the historical evolution and current state of dehazing techniques that harness the power of deep learning, offering a comprehensive survey for those delving into this dynamic and pivotal field. In a world where visuals are paramount, tackling the haze is no less than a quest for clarity and precision.