Haze Removal of Pedestrian Monitoring Images Using Dark Channel Prior Technique

Computer vision techniques are increasingly finding practical applications, especially in scenarios that rely heavily on outdoor image processing. Image visibility plays a key role in these contexts. The quality of captured scenes, especially outdoors, is significantly affected by adverse weather conditions such as fog or mist. These atmospheric conditions lead to degradation of visibility, resulting in reduced color vibrancy, reduced contrast and low saturation. Differences between objects and their background become less pronounced, leading to perceptible changes in color. To reduce the negative impact of haze and fog, several dehazing methods have been identified with the goal of restoring visual clarity, thereby extending the usefulness of computer vision. Additionally, achieving real-time dehazing capabilities is becoming increasingly imperative in response to requirements for computer vision applications that operate in real-time.