MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202641075334 A) filed by Hindusthan Institute Of Technology on June 18, 2026, for Smart Agriculture Wildlife Intrusion Detection And Repellent System Using Machine Learning.
Inventors include Dr. C. Natarajan; Dr. D. Jeyakumari; Dr. B. Paulchamy; D. Suganthi; S. Mooventhan; E. Om Prakash; G. Sri Kumaran; and J. Susilkumar.
The application for the patent was published on July 03, 2026, under issue no. 27/2026.
Abstract: Wildlife intrusion into agricultural fields remams a persistent challenge that causes significant crop loss and economic stress for farmers. This project presents an intelligent monitoring and response system that uses machine learning to detect animal presence and initiate timely deterrent actions. A camera continuously captures field images, which are processed using a YOLO-based detection model capable of identifYing multiple wildlife classes in real time. Upon confirmation of intrusion, the system activates non-harmful repellents such as flashing lights and • sound signals to drive animals away without causing injury. Simultaneously, alerts along with captured images. are transmitted to farmers through wireless communication for immediate awareness and decision-making. · The hardware platform integrates an ATmega328 microcontroller with sensors and actuation modules, ensuring low-power operation and field suitability. The trained model demonstrates reliable detection accuracy under varied environmental conditions, making it practical for deployment in rural settings. By reducing dependence on manual surveillance, the system enhances safety and efficiency. Overall, this solution offers a scalable and cost-effective approach to protecting crops while promoting sustainable coexistence between agriculture and wildlife. Keywords: wildlife detection, machine learning, YOLO, smart agriculture, intrusion monitoring, automated repellents, loT-based alerts, crop protection, embedded systems, precision farming. The design supports modular upgrades, enabling integration with solar power and cloud analytics for extended functionality. Its adaptability allows deployment across diverse crop types and terrains, ensuring broad usability. Future refinements can improve classification accuracy and reduce latency, further strengthening real-time response capabilities in dynamic outdoor environments. The system remains robust under challenging lighting and weather.
Disclaimer: Curated by HT Syndication.
