The intrusion of wild animals into agricultural areas significantly threatens crop productivity and farmer safety. Conventional deterrent techniques frequently prove inefficient, require excessive manual effort, or risk harming wildlife and ecosystems. This study introduces an intelligent system combining IoT and artificial intelligence (AI) to detect wildlife in real time and protect crops sustainably. The proposed framework employs a camera module for continuous field surveillance, utilizing a Convolutional Neural Network (CNN) powered by the YOLO algorithm to identify animals. Once detected, an IoT-connected ultrasound emitter activates, humanely repelling the animals without physical harm. Additionally, the system captures the animal’s image and transmits an instant alert via GSM to the farmer, including the visual data for remote monitoring. This non-invasive approach enhances eco-friendly agriculture, minimizes crop damage, and provides farmers with real-time alerts, supporting smarter and more sustainable farming
Introduction
Agriculture has undergone several revolutions, from early domestication to the Green Revolution, and now a fourth revolution fueled by information and communication technologies (ICT). Modern innovations like autonomous robots, drones with hyperspectral imaging, AI disease diagnosis, and virtual fencing are transforming farming by providing precision, real-time monitoring, and data-driven insights worldwide, including in developing regions.
The text focuses on an AI- and IoT-based intelligent wildlife management system designed to protect crops from animal intrusions non-lethally. Current methods (fencing, repellents, hunting) have limitations regarding ethics, cost, and effectiveness. The proposed system uses field cameras combined with a YOLO-enhanced convolutional neural network to detect and classify animals in real time. Upon detection, it activates ultrasonic emitters and other deterrents while sending alerts with images to farmers via GSM cellular networks.
The system architecture includes modular components for wildlife detection, humane deterrence, automated alerts, and a centralized cloud dashboard. It uses edge computing for fast, local processing, solar power for energy efficiency, and supports remote monitoring in rural areas. The AI model is trained on diverse animal datasets, achieving high accuracy in species recognition despite environmental challenges.
Field testing demonstrated reliable wildlife identification and effective deterrence with reduced need for manual monitoring, providing a scalable, sustainable, and ethical solution for crop protection that balances agricultural productivity and conservation.
Conclusion
This paper demonstrates an innovative IoT-AI integration that effectively addresses wildlife intrusions in agricultural settings. The developed solution combines YOLO-optimized convolutional neural networks for real-time animal recognition with humane ultrasonic deterrents, creating an environmentally conscious crop protection system. Integrated GSM alerts with visual documentation enable remote farm monitoring, significantly reducing both crop losses and manual supervision requirements. The implemented framework represents a sustainable approach to modern agriculture by merging technological innovation with ecological preservation.
For system optimization, future development could incorporate advanced image compression algorithms to accelerate alert transmission speeds. Additional improvements may include solar-powered operation expansion and machine learning-based deterrent customization based on animal behavior patterns. These enhancements would further increase the system\'s efficiency and adaptability across diverse agricultural environments.
References
[1] M. De Clercq, A. Vats, and A. Biel, \"Agriculture 4.0: The future of farming technology,\" in Proceedings of the World Government Summit, Dubai, UAE, Feb. 2018, pp. 11-13.
[2] Y. Liu et al., \"From Industry 4.0 to Agriculture 4.0: Current status, enabling technologies, and research challenges,\" IEEE Transactions on Industrial Informatics, vol. 17, no. 6, pp. 4322-4334, Jun. 2021.
[3] M. S. Farooq et al., \"A survey on the role of IoT in agriculture for the implementation of smart farming,\" IEEE Access, vol. 7, pp. 156237-156271, 2019.
[4] K. Kirkpatrick, \"Technologizing agriculture,\" Communications of the ACM, vol. 62, no. 2, pp. 14-16, Feb. 2019.
[5] A. Farooq, J. Hu, and X. Jia, \"Analysis of spectral bands and spatial resolutions for weed classification via deep convolutional neural network,\" IEEE Geoscience and Remote Sensing Letters, vol. 16, no. 2, pp. 183-187, Feb. 2018.
[6] M. Apollonio et al., \"Ungulates and their management in Italy,\" in European Ungulates and Their Management in the 21st Century. Cambridge, UK: Cambridge University Press, 2010, pp. 475-505.
[7] A. Amici et al., \"Increase in crop damage caused by wild boar (Sus scrofa L.): The \'refuge effect\',\" Agronomy for Sustainable Development, vol. 32, no. 3, pp. 683-692, Jul. 2012.
[8] S. Giordano et al., \"IoT solutions for crop protection against wild animal attacks,\" in 2018 IEEE International Conference on Environmental Engineering, Mar. 2018, pp. 1-5.
[9] M. O. Ojo, D. Adami, and S. Giordano, \"Network performance evaluation of a LoRa-based IoT system for crop protection against ungulates,\" in 2020 IEEE 25th International Workshop on Computer Aided Modeling and Design of Communication Links and Networks, Sep. 2020, pp. 1-6.
[10] H. E. Heefner and R. S. Heffner, \"Auditory perception,\" in Farm Animals and the Environment, C. Phillips and D. Piggins, Eds. Wallingford, UK: CAB International, 1992.