International Journal of Innovative Research in Computer and Communication Engineering
ISSN Approved Journal | Impact factor: 8.771 | ESTD: 2013 | Follows UGC CARE Journal Norms and Guidelines
| Monthly, Peer-Reviewed, Refereed, Scholarly, Multidisciplinary and Open Access Journal | High Impact Factor 8.771 (Calculated by Google Scholar and Semantic Scholar | AI-Powered Research Tool | Indexing in all Major Database & Metadata, Citation Generator | Digital Object Identifier (DOI) |
| TITLE | Animal Intrusion Detection System Using YOLOv8 and Deep Learning for Smart Agriculture: A Comprehensive Review |
|---|---|
| ABSTRACT | Human-animal conflict presents one of the most critical and economically devastating challenges for agricultural communities across India, with Maharashtra alone experiencing annual crop losses estimated between ₹10,000 crore and ₹40,000 crore due to wild animal intrusions, particularly by wild pigs (Sus scrofa). Conventional deterrent methods — physical fencing, manual night patrols, scarecrows, and firecrackers — are inadequate, labour-intensive, and incapable of providing real-time automated surveillance. This paper presents a comprehensive review and system implementation of an Animal Intrusion Detection System built upon YOLOv8l (You Only Look Once, Large variant), a state-of-the-art single-stage anchor-free deep learning object detection architecture. The proposed system integrates a Django REST Framework backend, a React.js + Tailwind CSS frontend dashboard, and a MySQL relational database to deliver a complete three-tier web-deployable surveillance platform. The system supports three input modalities: image upload, video frame extraction, and live webcam capture. A custom pig model (pig_best.pt) fine-tuned on 1,520 annotated pig and wild boar images achieves a mean Average Precision ([email protected]) of 91.4%, a precision of 94.2%, a recall of 91.8%, an F1-score of 93.0%, and a GPU inference latency of 48 milliseconds — surpassing all compared baseline models including YOLOv5s, Faster-RCNN, YOLOv8, YOLOv11, and CNN-YOLO hybrids. The review surveys 24 related works spanning CNN-based wildlife monitoring, YOLO variant applications in agricultural settings, IoT-integrated intrusion systems, and edge computing deployments. The paper discusses system architecture, methodology, result analysis, limitations, and a detailed future scope encompassing IoT deterrent integration, edge deployment on NVIDIA Jetson Nano, mobile push notifications, and multi-species detection extension. |
| AUTHOR | ISHAAN VIVEK PRABHUDESAI, DR. MANISHA BHARATI MSC Student, Department of Technology, Savitribai Phule Pune University, Pune, Maharashtra, India Guide, Department of Technology, Savitribai Phule Pune University, Pune, Maharashtra, India |
| VOLUME | 184 |
| DOI | DOI: 10.15680/IJIRCCE.2026.1405081 |
| pdf/81_Animal Intrusion Detection System Using YOLOv8 and Deep Learning for Smart Agriculture A Comprehensive Review.pdf | |
| KEYWORDS | |
| References | [1] Gokhale Institute of Politics and Economics, “Maharashtra Human-Wildlife Conflict Study: Economic Impact Assessment 2025,” Pune, India, 2025. [2] R. Patil and M. Bhatt, “Elephant Intrusion Detection using Faster R-CNN for Agricultural Field Protection,” International Journal of Computer Applications, vol. 184, no. 12, 2022. [3] G. Jocher, A. Chaurasia and J. Qiu, “Ultralytics YOLOv8,” [Software], Version 8.0, 2023. Available: https://github.com/ultralytics/ultralytics. [4] M. Lim, J. Park and S. Kim, “Edge Computing for Agricultural IoT: A Survey of Architectures and Applications,” IEEE Internet of Things Journal, vol. 9, no. 5, pp. 3780–3796, 2021. [5] G. Bradski, “The OpenCV Library,” Dr. Dobb’s Journal of Software Tools, 2000. Available: https://opencv.org. [6] P. Nair, K. Krishnan and T. Rajan, “Transfer Learning for Wildlife Classification in Camera Trap Images,” Ecological Informatics, vol. 68, p. 101563, 2022. [7] Saishruti et al., “Animal Intrusion Detection System using YOLOv11,” IEEE International Conference on Emerging Technologies, 2025. [8] J. Redmon, S. Divvala, R. Girshick and A. Farhadi, “You Only Look Once: Unified, Real-Time Object Detection,” Proc. IEEE CVPR, pp. 779–788, 2016. [9] A. Kumar, P. Sharma and R. Singh, “YOLOv5-Based Pig Detection for Precision Livestock Farming,” Computers and Electronics in Agriculture, vol. 192, p. 106615, 2021. [10] R. Patil and M. Bhatt, “Elephant Intrusion Detection using Faster R-CNN,” Intl. Journal of Computer Applications, vol. 184, no. 12, 2022. [11] H. Singh and V. Gupta, “Multi-class Wildlife Intrusion Detection in Agricultural Settings using Deep CNN,” Expert Systems with Applications, vol. 218, p. 119582, 2023. [12] Raza et al., “Enhancement of CNN-YOLO for Real-time Animal Detection,” Scientific Reports, Nature Portfolio, 2025. [13] Delwar et al., “Real-Time Farm Surveillance Using IoT and YOLOv8,” MDPI Future Internet, vol. 17, 2025. [14] M. Lim, J. Park and S. Kim, “Edge Computing for Agricultural IoT,” IEEE IoT Journal, vol. 9, no. 5, pp. 3780–3796, 2021. [15] A. Kumar and R. Meena, “GSM-Based Wildlife Alert System for Agricultural Fields,” Intl. Journal of Electronics and Communication Engineering, vol. 6, no. 3, 2019. [16] D. Ribeiro, G. Barbosa and N. Santos, “Thermal Imaging for Nocturnal Wildlife Detection in Agricultural Borders,” Computers and Electronics in Agriculture, vol. 165, 2019. [17] S. Ren, K. He, R. Girshick and J. Sun, “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,” NeurIPS, vol. 28, 2015. [18] W. Liu et al., “SSD: Single Shot MultiBox Detector,” ECCV, pp. 21–37, 2016. [19] T. Christie, “Django REST Framework,” Version 3.14, 2023. Available: https://www.django-rest-framework.org. [20] Meta Open Source, “React: A JavaScript Library for Building User Interfaces,” Version 18, 2023. Available: https://react.dev. [21] A. Wathan, “Tailwind CSS: A Utility-First CSS Framework,” Version 3.0, 2023. Available: https://tailwindcss.com. [22] Oracle Corporation, “MySQL 8.0 Reference Manual,” 2023. Available: https://dev.mysql.com. [23] G. Bradski and A. Kaehler, Learning OpenCV 4, O’Reilly Media, 2019. [24] P. Viola and M. Jones, “Rapid Object Detection using a Boosted Cascade of Simple Features,” Proc. IEEE CVPR, vol. 1, pp. 511–518, 2001. |
