Agri-guard: An Ai-based Smart Agricultural Surveillance System Using Yolov8 For Real-time Object Detection

Abstract

Crop Damage Caused By Wild Animals, Birds, And Unauthorized Human Intrusion Remains A Major Challenge In Modern Agriculture. Traditional Monitoring Approaches Such As Manual Supervision, Fencing, And Static Alarm Systems Are Often Inefficient, Labor-intensive, And Unable To Provide Continuous Protection. This Paper Presents Agri-Guard, An AI-powered Smart Agricultural Surveillance System That Leverages Real-time Computer Vision And Deep Learning For Automated Detection And Deterrence Of Threats In Farmland Environments. The Proposed System Utilizes The YOLOv8 Object Detection Model For Identifying Humans, Animals, And Birds From Live Camera Feeds. Video Streams Are Processed Using OpenCV, And Detections Are Handled Through A FastAPI-based Backend Architecture. Upon Detecting A Threat, The System Triggers Intelligent Alert Mechanisms, Records Evidence, And Logs Detection Data Into A Structured Database For Monitoring And Analysis. Experimental Evaluation Demonstrates That The System Achieves High Detection Accuracy With Low Latency, Making It Suitable For Real-time Agricultural Deployment. The Modular Architecture Ensures Scalability, Hardware Extensibility, And Integration With IoT-based Deterrent Mechanisms Such As Lights, Alarms, And Laser Systems.

Keywords