An Innovative Approach for Detection of Motorcyclists Helmet Non-Compliance using Edge Computing

Using innovative technologies and creative strategies can help reduce fatalities and lessen the financial impact of accidents. This suggests a significant improvement in global traffic law enforcement and road safety advocacy. The study proposed in this paper uses an innovative approach to address the challenge of motorcyclists neglecting traffic regulations through a deep learning-driven framework. Using advanced pre-trained object detection models, the system accurately identifies helmetless motorcyclists and number plates. In addition, the proposed model does the Optical Character Recognition (OCR) to extract the license plate information from the identified motorcyclists who are not wearing helmets. The architecture was tested using an NVIDIA Jetson as the edge device, delivering quick performance, efficiently managing large data volumes, and optimizing resource use. Extensive testing and in-depth analysis show that the proposed approach is robust and highly beneficial. The proposed system integrates OCR algorithms, edge computing capabilities, and advanced deep learning techniques to offer a scalable solution for assisting law enforcement entities and enhancing traffic safety initiatives.