AI-Based Drone System for Pest and Stray Animals in Farms

Abstract

Pests and stray animals are a major hazard to crops, particularly to small and marginal farmers, causing tremendous economic losses. Conventional methods of insect control are time consuming, costly and hard to amplify. This research proposes an AI manual drone system that combines real-time imaging and future analysis to an early detection of pests and stray animals. The system employs a low-cost unmanned aerial vehicle with a digital camera to photograph the area, avoid crop damage and cut down unnecessary pesticides. Images collected to identify pests and lost animals are treated using laptop vision strategies. A convolutional neural network (CNN) is trained upon a dataset including healthy crops, pest-infested plants and lost animals to verify the precision of high detection. The system controls the parameters by adjusting them based on environmental situations in order to dynamically minimize false positivity and make the detection more reliable. On top of this, real-time environmental data are used to predict potential insect invasions based on climatic factors so that prevention measures can be initiated. The system provides instant notice, which alerts farmers via mobile devices to allow timely action. Experimental results indicate 92.5% accuracy in detection, 7.5% false positive rate and average response time of 2.3 seconds, and provide quicker and accurate identification than conventional methods. In contrast to current single function solutions, this allows the double-turning system automated agricultural monitoring and efficient insect control, and is a scalable and effective agricultural safety solution.

Introduction

The text discusses the critical role of agriculture in food security and economic stability, highlighting challenges from insect pests and stray animals that cause significant crop losses. Traditional pest and animal control methods like fencing, manual inspections, and chemical pesticides are often inefficient, costly, and labor-intensive, especially for large-scale farms.

To address these issues, the study proposes an AI-powered autonomous drone system equipped with high-resolution cameras and deep learning models (notably YOLOv8 and CNN architectures) for real-time detection of both insect pests and stray animals. The system integrates advanced image processing techniques (Gaussian filtering, edge detection, histogram equalization) and uses thermal imaging to improve detection accuracy, even in low-light conditions.

This drone platform includes onboard AI processing on a Raspberry Pi, reducing dependence on cloud computing and minimizing latency and energy consumption. It also features a hybrid power management system to extend flight time. When pests or animals are detected, the system instantly alerts farmers via Bluetooth and GSM communication.

The system achieves high accuracy (~92.5%), precision (~91.2%), and fast response times (~2.3 seconds), outperforming traditional IoT sensor-based methods and manual monitoring. It offers a scalable, cost-effective, and automated solution for comprehensive farm surveillance, minimizing pesticide overuse and crop losses while enhancing agricultural safety and productivity.

Conclusion

This study offers an AI-powered entirely drone to identify pests and missing animals while solving a main firm in contemporary agriculture. Conventional approaches demand guide interventions and chemical solutions, harmful to time building and the environment. Contrarily, our device is advantaged by photography processing and AI-run class to provide real-time computer-controlled identification, taking reactiveness to assist farmers to defend their crops. In contrast to existing methods that rely on expensive thermal cameras, this method employs cost-effective photography-based detection methods. Through deep learning fashion trained on various datasets, the device achieves high precision in identifying pests and lost animals under special environmental situations. The inclusion of the drone time image ensures scalability to the massive fields. In the period of being efficient, the device has certain limits, which incorporates the state of high quality lighting, digital camera and data set variety. Upgrade in the future needs to be identified by reinforcing the model power, incorporating multi -sensor realities and processing actual -time choice for improved general performance. Lastly, this inspection adds value to smart agriculture by providing an enduring and exceptional AI driven detection unit. In the future, further advancements in AI and drones will make things more beautiful at performance, thereby creating an additional smart and self-driven agricultural habitat.

References

[1] N. M. Prasad, R. S. Hegde, and R. R. Rao, \"Smart pest detection for an agricultural field crop based on deep learning,\" Int. J. Agric. Environ. Inf. Syst., vol. 10, no. 2, pp. 55–69, 2024. [2] M. A. Pawar, S. S. Kulkarni, and P. R. Deshmukh, \"Farmatron-pest detection and treatment using AI-based drones,\" Int. J. Res. Appl. Sci. Eng. Technol., vol. 8, pp. 19–24, 2020. [3] M. H. M. Ghazali and W. Rahiman, \"Vibration-based fault detection in drones using artificial intelligence,\" IEEE Sensors J., vol. 22, no. 9, pp. 8439–8448, 2022. [4] T. Özer and Ö. Türkmen, \"Low-cost AI-based solar panel detection drone design and implementation for solar power systems,\" Robotic Intell. Autom., vol. 43, no. 6, pp. 605–624, 2023. [5] J. Zhang, \"Towards a high-performance object detector: insights froY43m drone detection using ViT and CNN-based deep learning models,\" in Proc. IEEE Int. Conf. Sens. Electron. Comput. Eng. (ICSECE), Aug. 2023, pp. 141–147. [6] A. Carrio, S. Vemprala, A. Ripoll, S. Saripalli, and P. Campoy, \"Drone detection using depth maps,\" in Proc. IEEE/RSJ Int. Conf. Intell. Robots Syst. (IROS), 2018, pp. 1034–1037. [7] S. Kumar, R. Srivastava, A. K. Yadav, S. Verma, B. Kusvaha, and S. Verma, \"Design optimization and performance analysis of autonomous drone systems,\" J. Aerosp. Eng. Technol., vol. 14, no. 3, pp. 21–30, 2024. [8] D. P. Sai, A. Farooq, G. Kumari, K. S. Priyanka, and D. Govardhan, \"Computational fluid dynamics analysis for enhancing the performance of drone frames,\" Int. J. Eng. Res. Technol. (IJERT), vol. 10, no. 5, 2021. [9] W. Q. Li, X. X. Han, Z. B. Lin, and A. Rahman, \"Enhanced pest and disease detection in agriculture using deep learning-enabled drones,\" Acadlore Trans. Mach. Learn., vol. 3, no. 1, pp. 1–10, 2024. [10] K. K. Shaw and R. Vimalkumar, \"Design and development of a drone for spraying pesticides, fertilizers, and disinfectants,\" Int. J. Eng. Res. Technol. (IJERT), vol. 9, 2020. [11] C. Akdo?an, T. Özer, and Y. O?uz, \"Design and implementation of an AI-controlled spraying drone for agricultural applications using advanced image preprocessing techniques,\" Robotic Intell. Autom., vol. 44, no. 1, pp. 131–151, 2024. [12] L. Nanni, A. Manfè, G. Maguolo, A. Lumini, and S. Brahnam, \"High performing ensemble of convolutional neural networks for insect pest image detection,\" Ecol. Inform., vol. 67, p. 101515, 2022. [13] L. Crupi, L. Butera, A. Ferrante, and D. Palossi, \"A deep learning-based pest insect monitoring system for ultra-low power pocket-sized drones,\" in Proc. IEEE Int. Conf. Distrib. Comput. Smart Syst. Internet Things (DCOSS-IoT), 2024, pp. 323–330. [14] Y. R. G. V., \"Internet of Things assisted unmanned aerial vehicle for pest detection with optimized deep learning model,\" Web Intell., vol. 22, no. 2, pp. 269–290, 2024. [15] D. Kapetas, P. Christakakis, S. Faliagka, N. Katsoulas, and E. M. Pechlivani, \"AI-driven insect detection, real-time monitoring, and population forecasting in greenhouses,\" AgriEng., vol. 7, no. 2, p. 29, 2025. [16] M. Ibraheam, K. F. Li, and F. Gebali, \"An accurate and fast animal species detection system for embedded devices,\" IEEE Access, vol. 11, pp. 23462–23473, 2023. [17] M. A. Alanezi, A. Mohammad, Y. A. Sha’aban, H. R. E. H. Bouchekara, and M. S. Shahriar, \"Auto-encoder learning-based UAV communications for livestock management,\" Drones, vol. 6, no. 10, p. 276, 2022. [18] M. K. Rajagopal and B. M. MS, \"Artificial intelligence-based drone for early disease detection and precision pesticide management in cashew farming,\" arXiv preprint arXiv:2303.08556, 2023. [19] P. K. Singh and A. Sharma, \"An intelligent WSN-UAV-based IoT framework for precision agriculture application,\" Comput. Electr. Eng., vol. 100, p. 107912, 2022. [20] N. R. Dr., S. N. S., S. K. K., S. S. Bedre, and S. C. S., \"AI-based drone for crop disease detection in precision agriculture,\" Int. J. Eng. Res. Technol. (IJERT), vol. 13, no. 4, 2024. [21] B. Martinez, J. K. Reaser, A. Dehgan, B. Zamft, D. Baisch, C. McCormick, A. J. Giordano, R. Aicher, and S. Selbe, \"Technology innovation: advancing capacities for the early detection of and rapid response to invasive species,\" Biol. Invasions, vol. 22, no. 1, pp. 75–100, 2020. [22] E. Larson et al., \"From eDNA to citizen science: emerging tools for the early detection of invasive species,\" Front. Ecol. Environ., vol. 18, pp. 194–202, 2020. [23] Z. J. Junyang, \"Towards a high-performance object detector: insights from drone detection using ViT and CNN-based deep learning models,\" in Proc. IEEE Int. Conf. Sens. Electron. Comput. Eng. (ICSECE), 2023, pp. 141–147.

Copyright

Copyright © 2025 Sumit Rathod, Nirmiti Puslekar, Tanush Polaswar. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.