Automated Motorcycle Helmet Detection Using The Combination of YOLO AND CNN

Motorcycle accidents are a significant public concern worldwide. Accidents have been increasing rapidly which leads to the loss of several lives. One of the main contributing factors to the severity of injuries in these accidents is the lack of helmet use. Though the usage of helmets is necessary, it is also considered a piece of mandatory safety equipment among drivers, but they do not make use of it which causes unfortunate incidents to take place. Traffic police do try to bring awareness among the people but most of the population refuses to follow them. To avoid such incidents and to detect people from breaking the rules, this research was introduced. Here, we present a study based on an automated helmet detection system for motorcyclists using the combined techniques of YOLO and CNN. The combination of these image processing techniques such as You Only Look Once (YOLO) and Convolutional Neural Network (CNN) helps in the identification of helmets from image datasets. Because of this combination, the model can accurately identify helmets with a high degree of reliability due to the use of image datasets. And due to this combination, our system can accurately detect helmets with an accuracy of 94.29%.