Beast damage measures using intelligent capturing system
Chenghui LIU, Wenbang Dou, Naoyuki Kubota · The Proceedings of JSME annual Conference on Robotics and Mechatronics (Robomec) · 2021
In Japan, damage caused by vermin such as deer and wild boar in farmland is serious. Measures such as extermination with firearms and traps have been taken, and the amount of damage has been gradually decreasing in recent years, but there is a need for more efficient extermination. In this paper, we develop an intelligent capture system using off-the-shelf net launchers and confirm its effectiveness. The proposed system includes an image recognition system and a tracking system. The image recognition system uses a method called Single Shot Multibox Detection (SSD) to recognize animals, and the tracking system uses a pan-tilt mechanism to track animals. We show that the proposed system can capture animals by making these two systems work effectively.