Design and Implementation of an Embedded UAV Visual Detection System for Agricultural Inspection

Chang Luo; Jingjing Song; Huan Gao; Mingxuan He; Qian Wang

doi:10.63313/AJET.9041

Authors

Chang Luo School of Information Science and Engineering, Dalian Polytechnic University, Dalian, China Author
Jingjing Song School of Information Science and Engineering, Dalian Polytechnic University, Dalian, China Author
Huan Gao School of Information Science and Engineering, Dalian Polytechnic University, Dalian, China Author
Mingxuan He School of Information Science and Engineering, Dalian Polytechnic University, Dalian, China Author
Qian Wang School of Information Science and Engineering, Dalian Polytechnic University, Dalian, China Author

DOI:

https://doi.org/10.63313/AJET.9041

Keywords:

Quadrotor, UAV, Machine, vision, perception, Embedded, edge, computing, Target, detection

Abstract

To address the problems of low efficiency in manual pest identification, complex field environments, and other challenges in agricultural inspection scenarios, this paper designs and implements a low-cost embedded UAV visual inspection system. Based on a quadrotor UAV platform, the system integrates a flight control system, an onboard camera, and an embedded computing unit to achieve automatic image ac-quisition, edge-side processing, and result output. In the target detection module, an improved Global Context Perception YOLO(GCP-YOLO) model is introduced to enhance the recognition ability for pest targets under complex backgrounds. Exper-imental results show that the proposed system achieves good performance in both detection accuracy and real-time processing, and verifies its engineering feasibility in low-cost agricultural inspection applications.

References

[1] YIN J, ZHU J, CHEN G, et al. An intelligent field monitoring system based on enhanced YOLO-RMD architecture for real-time rice pest detection and management[J]. Agriculture, 2025, 15(8): 798. DOI:10.3390/agriculture15080798.

[2] CEN X, LU S, QIAN T. YOLO-LCE: A lightweight YOLOv8 model for agricultural pest detection[J]. Agronomy, 2025, 15(9): 2022. DOI:10.3390/agronomy15092022.

[3] HOU S, LIU Z, ZHANG Y, et al. RS-YOLO: A highly accurate real-time detection model for small-target pests in complex environments[J]. Smart Agricultural Technology, 2025, 11: 100637. DOI:10.1016/j.atech.2025.100637.

[4] HE Z, ZHANG Y, LIU C, et al. SSCW-YOLO: A lightweight and high-precision model for UAV-based small object detection[J]. Drones, 2026, 10(1): 41. DOI:10.3390/drones10010041.

[5] ZHOU Y, HUANG W, LIU B, et al. Lightweight GAFNet model for robust rice pest detection in complex agricultural environments[J]. AgriEngineering, 2026, 8(1): 26. DOI:10.3390/agriengineering8010026.

[6] HUANG Y, LIU Z, ZHAO H, et al. YOLO-YSTs: An improved YOLOv10n-based method for real-time field pest detection[J]. Agronomy, 2025, 15(3): 575. DOI:10.3390/agronomy15030575.

[7] LU Y, ZHANG X, LIU H, et al. MA-YOLO: A pest target detection algorithm with multi-scale fusion and attention mechanisms[J]. Agronomy, 2025, 15(7): 1549. DOI:10.3390/agronomy15071549.

[8] SUN F, GUAN Z, LYU Z, et al. Efficient and lightweight deep learning model for small-object detection in stored grain pest management[J]. Journal of Stored Products Research, 2026, 115: 102890. DOI:10.1016/j.jspr.2025.102890.

[9] ZHU L, PENG W, ZHOU L, et al. YOLO-LSD: A lightweight scale dynamic model for tiny pest detection in tea plantations[J]. Smart Agricultural Technology, 2025, 12: 101474. DOI:10.1016/j.atech.2025.101474.