Published Paper Details:

This paper solves a real-world problem of rider safety by ensuring helmets are worn through the use of the machine-learning technology, YOLO v8, used for helmet detection. The goal is to develop a system that can accurately detect helmets in real-time video feeds. starting from the high-resolution recording of video captures made from the camera on the rider, image processing is deployed to enhance the quality of the frames and make it easier for features to be identified. The real-time processing system efficiently detects whether the rider is wearing a helmet. several processes are involved namely, collecting data, preparing images, choosing the correct architecture, and building up the model. Training the model is essential, and during this stage, the enhancement of the model by the addition of optional data augmentation can significantly help improve the model's performance and make it less sensitive to certain variations, like lighting and pose of the rider. the YOLO v8 model consists of very efficient layers that help with the quick identification of helmets on riders. This has been trained on a lot of data, and its performance has been reported regularly to find efficiency in classifying any object as a rider with/without a helmet. if a helmet is detected, the servo motor, in conjunction with the buzzer, will be activated to alert the rider. The feedback here is instantaneous and encourages them always to buckle up. This is a so-called proactive attempt to better rider safety with the addition of smart technology with the sole aim of proper wearing of helmets.