License Plate Detection Using YOLOv7 and Optical Character Recognition

In light of the rising number of accidents involving reckless driving, effective number plate detection is essential for road safety. To enforce traffic laws and guarantee a safe driving environment, number plates must be accurately identified. However, current automatic number plate detection systems face efficiency problems, especially when dealing with bad weather, situations involving high speeds, or fuzzy images-common occurrences on the roadways. In response, a novel strategy using image augmentations that mimic various environmental situations is suggested. The detecting technology is strengthened against these difficulties by simulating the settings known to obstruct number plate clarity. This paper precisely extract numbers from the detected plates by utilising the YOLO V7 architecture's capabilities and effective Optical Character Recognition (OCR) approaches. This papers thorough testing of various image sizes resulted in improved findings and a detection method that was optimised. By improving automatic number plate recognition, addressing the shortcomings of current systems, and promoting a safer driving environment for both motorists and pedestrians, this research greatly improves road safety.