Weapon Detection for Smart Surveillance System using YOLOV6

In recent years, many parts of the world recorded an increase in weapon violence, which led to a lot of innocent
killings. The purpose of this work is to develop a computer vision-based automated smart surveillance system that
detects visible weapons (a variety of knives, guns, and rifles) and sends alerts to the appropriate authorities for
necessary action. Deep learning has provided a lot of solutions across the various sectors of human endeavor. In
our proposed system, we used a single-stage target deep learning algorithm called YOLOv6 (You Only Look Once
version 6) to detect visible weapons in real-time. Early detection of visible weapons increased response time by
providing authorities and law enforcement with precious time to address the situation.