Volume - 6 | Issue - 4 | december 2024
Published
17 February, 2025
The rising global population and increasing food demand have placed immense pressure on agriculture, particularly in regions like the Marudhamalai foothills in Coimbatore district, where farmers face frequent crop damage caused by wildlife intrusions, such as wild boars and deer. To address these challenges, a wildlife intrusion detection system has been developed to safeguard crops, enhance agricultural productivity, and enable coexistence with wildlife. The system combines a laser detection setup with an AI-CAM that employs lightweight deep learning algorithms for real-time animal detection and classification. This system also ensures efficient animal deterrence and real-time monitoring for farmers, enabling them to assess the situation with the assistance of an intelligent rover built using IoT. The detection system consists of a processor, Light Dependent Resistor module, laser diode, and buzzer are used to detect intrusions with precision and provide an immediate response. The system includes an ESP32-CAM to monitor and deter animals effectively. It features live image processing through an OLED display and a USB-to-TTL adapter, ensuring reliable performance with minimal power consumption. By integrating IoT technology and advanced monitoring systems, farmers receive real-time updates and can remotely control a rover for necessary interventions. The system achieves 96.3% accuracy in real-time animal detection using an ESP32-CAM and a YOLO v8 model, with results displayed on an OLED screen and bounding boxes for classification.
KeywordsEdge AI Animal Intrusion Detection Internet of Things TinyML YOLOv8 Model ESP32 Microcontroller Edge Impulse and Machine learning
