This work is an anti-theft and warning system that combines artificial intelligence, object detection, computer vision and virtual fence algorithms. This technology allows users to select and draw virtual fence protection areas according to site requirements and create three warning lines to detect objects such as vehicles and underwater divers approaching the warning area. Based on the degree of intrusion, corresponding warning signals will be immediately sent to the management personnel.

"1. Polygonal virtual fences can be customized according to the areas that need to be warned;
 
 2. Design a deep learning model that can accurately detect and identify targets in images, including vehicles, divers, ships, etc. The system’s day and night accuracy mAP averages over 93%;
 
 3. The intelligent alarm system can automatically trigger different levels of alarm functions based on target behavior;
 
 4. It can be quickly installed and used in different fields (above ground and underwater)."

"Industrial applications have been successfully implemented in:
 
 1.Five large-scale offshore cage net farms in Penghu have the system in place with the goal of lowering annual theft losses of more than NT$10M for farmers;
 
 2.A large shipyard is responsible for omni-security tasks on the dock and underwater;
 
 3.It was installed at the gate of underground parking lot at Taipei Tech to prevent teachers and students from deliberately not going to the gate to have their forehead temperatures taken."