Enhanced Locust Detection in Smart Farming Using YOLOv5 and YOLOv8 with Data Augmentation: A Comparative Performance Evaluation
Corresponding author: Pooja Vajpayee
Abstract
Deep learning-based object detection models have emerged as powerful tools for real-time pest monitoring in agriculture. Locust swarms pose a severe threat to crops, necessitating prompt and accurate detection. This study evaluates the performance of two state-of-the-art object detection models, YOLOv5 and YOLOv8, for locust detection in smart farming. After training and testing these models on a custom dataset, their accuracy, precision, recall, mean average precision (mAP), and F1 score are compared. The results demonstrate that YOLOv8 slightly outperforms YOLOv5 in terms of accuracy and recall, while YOLOv5 is faster and requires fewer resources. In addition to baseline testing, data augmentation techniques are applied, leading to improved model performance and accuracy. YOLOv8 achieved an accuracy of 79.80% after data enhancement, highlighting its potential for real-time locust detection. This comparative analysis provides valuable insights into the strengths and weaknesses of each model, which can contribute to smart farming initiatives, aiding in the selection of the most suitable tool for sustainable agriculture and food security, such as pest monitoring and management.
Keywords:
YOLOv5, YOLOv8, data augmentation, locust detection, object detectionDownloads
References
Y. Amit, P. Felzenszwalb, and R. Girshick, "Object Detection," in Computer Vision, Springer, 2021, pp. 875–883.
X. Wu, D. Sahoo, and S. C. H. Hoi, "Recent advances in deep learning for object detection," Neurocomputing, vol. 396, pp. 39–64, Jul. 2020.
Y. Xiao et al., "A review of object detection based on deep learning," Multimedia Tools and Applications, vol. 79, no. 33, pp. 23729–23791, Sep. 2020.
Y. Zhang, X. Li, L. Zhang, and H. Qi, "A Locust Swarm Recognition Algorithm Based on Faster-RCNN and GCN," in 2023 IEEE 6th International Conference on Computer and Communication Engineering Technology (CCET), Beijing, China, Aug. 2023, pp. 97–101.
A. Dharshini, A. Monisha, M. B. Malini, and S. K. Vinoth, "Method to prevent and track Locust’s Intrusion using Object Detection Algorithms," in 2022 International Conference on Communication, Computing and Internet of Things (IC3IoT), Chennai, India, Mar. 2022, pp. 1–6.
D. Zaveri, D. Shah, S. Joshi, T. Ahuja, N. M. Patil, and S. C. Iyer, "A Device for Detection and Deterrence of Locust Threats," in 2023 International Conference on Emerging Trends in Networks and Computer Communications (ETNCC), Windhoek, Namibia, Aug. 2023, pp. 1–6.
Y. Zhang, X. Li, and A. Wu, "A Locust Recognition Method Based on Graph Convolution Neural Network," in 2021 IEEE 4th International Conference on Computer and Communication Engineering Technology (CCET), Beijing, China, Aug. 2021, pp. 58–62.
P. Vajpayee, K. K. Yogi, P. Vajpayee, and K. K. Yogi, "Recognition and Early Stage Detection of Phytophthora in a Crop Farm Using IoT," in Agro-Economic Risks of Phytophthora and an Effective Biocontrol Approach, IntechOpen, 2021.
B. Zoph, E. D. Cubuk, G. Ghiasi, T. Y. Lin, J. Shlens, and Q. V. Le, "Learning Data Augmentation Strategies for Object Detection," in Computer Vision – ECCV 2020, 2020, pp. 566–583.
R. Ahmed and E. H. Abd-Elkawy, "Improved Tomato Disease Detection with YOLOv5 and YOLOv8," Engineering, Technology & Applied Science Research, vol. 14, no. 3, pp. 13922–13928, Jun. 2024.
M. H. F. Afonso, E. H. Teixeira, M. R. Cruz., G. P. Aquino, and E. C. Vilas Boas, "Vehicle and Plate Detection for Intelligent Transport Systems: Performance Evaluation of Models YOLOv5 and YOLOv8," in 2023 IEEE International Conference on Computing (ICOCO), Langkawi, Malaysia, Oct. 2023, pp. 328–333.
E. Soylu and T. Soylu, "A performance comparison of YOLOv8 models for traffic sign detection in the Robotaxi-full scale autonomous vehicle competition," Multimedia Tools and Applications, vol. 83, no. 8, pp. 25005–25035, Mar. 2024.
E. Casas, L. Ramos, C. Romero, and F. Rivas-Echeverría, "A comparative study of YOLOv5 and YOLOv8 for corrosion segmentation tasks in metal surfaces," Array, vol. 22, Jul. 2024, Art. no. 100351.
P. Vajpayee and K. Kr. Yogi, "Enhanced Grasshopper Detection in Smart Crops Using Faster R-CNN Deep Learning Model," in 2024 3rd Edition of IEEE Delhi Section Flagship Conference (DELCON), New Delhi, India, Nov. 2024, pp. 1–7.
A. K. Kanade, M. P. Potdar, A. Kumar, G. Balol, and K. Shivashankar, "Weed detection in cotton farming by YOLOv5 and YOLOv8 object detectors," European Journal of Agronomy, vol. 168, Jul. 2025, Art. no. 127617.
G. Liu, J. Di, Q. Wang, Y. Zhao, and Y. Yang, "An Enhanced and Lightweight YOLOv8-Based Model for Accurate Rice Pest Detection," IEEE Access, vol. 13, pp. 91046–91064, 2025.
M. M. Abo-Zahhad and M. Abo-Zahhad, "Real time intelligent garbage monitoring and efficient collection using Yolov8 and Yolov5 deep learning models for environmental sustainability," Scientific Reports, vol. 15, no. 1, May 2025, Art. no. 16024.
"Locust Dataset > Overview," Roboflow. https://universe.roboflow.com/new-workspace-ocpp0/locust-kvcoz.
P. Chudzik et al., "Mobile Real-Time Grasshopper Detection and Data Aggregation Framework," Scientific Reports, vol. 10, no. 1, Jan. 2020, Art. no. 1150.
A. Mumuni and F. Mumuni, "Data augmentation: A comprehensive survey of modern approaches," Array, vol. 16, Dec. 2022, Art. no. 100258.
P. Vajpayee and K. Yogi, "Locust Images Dataset." Zenodo, Mar. 09, 2025.
P. Vajpayee, "poojavjpy/Locust-Detection." May 01, 2025, [Online]. Available: https://github.com/poojavjpy/Locust-Detection.
Downloads
How to Cite
License
Copyright (c) 2025 Pooja Vajpayee, Kuldeep Kr. Yogi, Amit Kumar
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain the copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) after its publication in ETASR with an acknowledgement of its initial publication in this journal.
