YOLO and BiLSTM Fusion for Smart Motorcycle Ignition

Motorcycle accidents remain a leading cause of severe injuries and fatalities worldwide, underscoring the critical need for effective helmet enforcement mechanisms. This paper presents a novel and intelligent helmet recognition system designed to enhance road safety through accurate, real-time detection and compliance verification. Traditional systems based on Single Shot detection (SSD) and Convolutional neural networks(CNNs) often suffer from limitations in detection accuracy and latency, making them less viable for dynamic realworld applications. To overcome these challenges, the proposed system integrates the YOLO (You Only Look Once) architecture for fast and precise object detection with a Bidirectional Long Short-Term Memory (BiLSTM) network to capture and classify temporal features, enhancing decision-making over time. The Raspberry Pi receives the output of the helmet detection and uses it to control the motor cycle's ignition and trigger the relay. A key innovation of this system is a 60-second continuous verification module that ensures the motorcycle ignition activates only of a helmet is consistently detected, thereby minimizing false positives and promoting genuine compliance. Extensive experimental evaluation demonstrates substantial improvements in accuracy, processing speed and robustness across varying conditions, outperforming conventional approached. The proposed solution offers a practical, scalable, and efficient framework for integration into intelligent transportation system, with the ultimate goal of reducing head injuries and saving lives through proactive safety enforcement.