Enhancing Rider Safety - A Computer Vision based Approach for Helmet Detection

Bike riders have been rapidly rising in numerous countries throughout time. Motorcycles are popular among residents from many social strata for a variety of reasons, including their low cost. Helmets are required by law, yet the majority of people choose not to wear them. The primary purpose of the helmet is to protect the riders. This study collects and annotates a variety of datasets, including photographs and video frames with and without helmets. Following that, this study modifies a trained object identification model such that it can recognise helmets in a variety of locations and lighting conditions. This study uses the trained model to analyse real-time video data from cameras positioned in locations. When helmets are recognised, they are marked in the video feed with bounding boxes, allowing spectators to easily identify persons, who follow safety requirements. This study incorporates processing techniques such as non-maximum suppression to improve accuracy. Furthermore, this technology allows you to establish a confidence threshold which will activate warnings or messages if helmets do not achieve the specified degree of confidence or are not identified at all. This study employs cutting-edge deep learning approaches for object detection, allowing persons wearing helmets in video feeds to be identified. To build a customised object identification model particularly built for helmet recognition, we use transfer learning with trained convolutional neural networks (CNNs). The goal of this research work is to create a deep learning-based helmet detection system using trained models and datasets, which will be useful for the police branch in implementing regulations for the good of society. The main aim of this research work is to detect the biker's helmets to enhanced road safety by using deep learning and CNN based algorithms. The proposed model enhances biker's safety and its helpful for the traffic police to check whether they wear the helmet or not while driving.