Rain, fog, and haze are atmospheric phenomena that significantly reduce visibility, especially during adverse weather conditions. These weather conditions pose a substantial risk to road safety, as they can impair the vision of drivers and lead to accidents. Detecting and removing rain, fog, and haze from images captured by traffic cameras or vehicle-mounted sensors can significantly enhance visibility, helping drivers navigate safely and prevent accidents. Traditional systems often rely on image processing techniques and filters. These methods attempt to enhance visibility by reducing the impact of rain streaks, fog, or haze on images. However, these traditional techniques are often limited in their effectiveness, especially in real-time applications and varying weather conditions. Hence, the need for a robust system for rain, fog, and haze removal in the context of traffic safety is paramount. Accurate and real-time detection of these weather conditions, coupled with effective removal techniques, can enhance driver visibility, reduce accidents, and save lives. These systems are especially crucial for autonomous vehicles, where clear and unobstructed vision is vital for safe navigation. Therefore, this research aims to build a system with OpenCV, a popular open-source computer vision library. OpenCV, with its rich set of functions and algorithms, provides a robust platform for developing rain, fog, and haze detection and removal systems. It offers various image processing techniques, including filtering, morphological operations, and machine learning algorithms that can be utilized to address these challenges. By implementing intelligent algorithms in conjunction with OpenCV, it's possible to create efficient and accurate solutions for real-time applications, aiding in the prevention of traffic accidents during adverse weather conditions.