DESIGN OF WEAPON DETECTION SYSTEM FOR CONCEALED WEAPONS USING MULTI SENSOR INTEGRATION

ABSTRACT

The proliferation of concealed weapons poses a significant challenge to public safety and security, necessitating advanced detection systems capable of identifying potential threats in various environments. This paper addresses the need for an enhanced concealed weapons detection model by proposing a comprehensive integration of millimeter-wave and LiDAR sensors with advanced algorithms. The aim is to develop a system capable of accurately detecting and localizing concealed weapons in real-time, thereby mitigating the risks associated with potential acts of violence. To achieve this objective, the proposed system leverages state-of-the-art sensor technologies, including millimeter-wave radar and LiDAR sensors, to capture detailed environmental data. Advanced algorithms, such as YOLOv8 and EfficientNet, are employed for object detection and classification, enabling the system to identify concealed weapons with high precision and recall. Additionally, sensor fusion techniques are utilized to integrate data from multiple sources, further enhancing the system's detection capabilities. The outcomes of this research demonstrate promising results in the field of concealed weapons detection. Through comprehensive evaluations and experiments, the proposed system consistently achieves high performance metrics, including precision, recall, and mean average precision (mAP). The system exhibits rapid response times and robust detection capabilities across diverse scenarios, underscoring its effectiveness in enhancing security measures and safeguarding public spaces. Moving forward, further refinement and optimization of the system are warranted to address specific challenges and enhance its practical applicability in real-world settings.