Particularly in bad weather, atmospheric events including rain, fog, and haze greatly lower visibility. These kinds of weather seriously compromise road safety as they could obscure drivers' eyesight and cause collisions. By means of rain, fog, and haze detection and removal from traffic camera or vehicle-mounted sensor-captured images, visibility may be greatly improved, therefore facilitating driver navigation and accident avoidance. Often based on image processing methods and filters, traditional systems These techniques try to improve vision by lowering the effect on photos of rain streaks, fog, or haze. Still, these conventional methods are sometimes less successful, particularly in real-time applications and in changing weather. Therefore, in the framework of traffic safety, the demand for a strong system for rain, fog, and haze elimination takes front stage. Together with efficient removal methods, accurate and real-time identification of these meteorological events can improve driver vision, lower accidents, and even save lives. Autonomous cars notably depend on these systems as safe navigation depends on good, unhindered eyesight. Thus, this work intends to construct a system using the well-known open-source computer vision library OpenCV. With its extensive suite of tools and algorithms, OpenCV offers a strong basis for creating systems of rain, fog, and haze identification and removal. It provides several image processing methods to handle these difficulties: machine learning algorithms, morphological procedures, and filtering among other ones. Intelligent algorithms combined with OpenCV allow one to generate precise and quick solutions for real-time applications, therefore helping to reduce traffic accidents under unfavorable weather