Deep Learning-Based System for Automatically Detecting and Panelizing Helmet Rule Violations

In order to reduce the likelihood of brain damage sustained while motorcycling, a helmet should be worn at all times. Even in jurisdictions where it is the law, fewer people are opting to wear helmets. This system’s goal is to automate the process of recognizing motorcycle license plates and riders, whether they are wearing helmets or not. Currently, helmets worn by motorcyclists are physically examined by law enforcement. Due to human error and the slow rate of manual assessment, there are too many limits to justify it. Due to the lack of complete automation in CCTV-based systems, human interaction in monitoring is still necessary in many non- and semi-urban settings. Due to the increasing number of fatalities caused by motorcyclists who did not wear helmets, there has been an increase in research on the topic of road safety monitoring. To achieve the goal of automatic helmet recognition and analysis of the number plate of the helmet regulation infringement, this strategy proposes using YOLO V8 and decision trees to construct an effective image processing solution. Extensive measures have been used to test the effectiveness of this technique, and the findings have been quite good.