In general, license plates are recognized under normal circumstances but it becomes very critical to track in hazy and foggy environment. Such conditions are often faced at hilly or terrain area where foggy weather and hazy environment creates challenges for safety and security. Mainly tourists are the soft target at such types of tourist destinations looted and theft through vehicles. Safety and security is a prime concern for an international tourist while explore any other country. Despite this, recognizing vehicle license plates is very difficult, especially when there is fog present in a particular global environment. Most of the time, fog or haze blurs the boundaries and characters of license plates, making them difficult to detect or identify. The purpose of this paper is to propose a hybrid machine learning algorithm to remove haze, improve image quality and identify a vehicle. To make license plate recognition more accurate in foggy weather, this algorithm uses two hybrid machine learning approach, which will be conducive to ensuring the safe operation of vehicles in foggy weather.