Animal Intrusion Detection using YOLOv9
M. Nithya Sree, K. Tejaswi, Shruti Shalini, M S Arunkumar · 2025
Animal intrusion in human-populated areas poses significant threats to both wildlife and human safety, necessitating efficient monitoring and detection systems. This research focuses on developing an animal intrusion detection model using YOLOv9, a deep learning object detection algorithm. The suggested approach uses a specially trained YOLOv9 model to precisely identify and categorize different animal species. To ensure robust and accurate results, the training involves a diverse dataset of images including different types of animals in varied environments. The overall aim is to facilitate high accuracy and real-time detection for proactive action in preventing human-wildlife conflicts. Experimental results confirm that the proposed model successfully detects animals with state-of-the-art accuracy and significantly lower inference time than its predecessors of YOLO. This work further provides a groundwork for future connection with hardware systems (for instance, surveillance cameras and drones) to automate monitoring of close ecosystems in forest edges, farming fields, and urban settlements.