Optimization of Animal Detection in Thermal Images Using YOLO Architecture

• Autorzy: Popek Łukasz , Perz Rafał , Galiński Grzegorz , Abratański Artur

Identyfikatory

DOI

10.24425/ijet.2023.147707

• Języki publikacji: EN

Abstrakty

EN

The article presents research on animal detection in thermal images using the YOLOv5 architecture. The goal of the study was to obtain a model with high performance in detecting animals in this type of images, and to see how changes in hyperparameters affect learning curves and final results. This manifested itself in testing different values of learning rate, momentum and optimizer types in relation to the model’s learning performance. Two methods of tuning hyperparameters were used in the study: grid search and evolutionary algorithms. The model was trained and tested on an in-house dataset containing images with deer and wild boars. After the experiments, the trained architecture achieved the highest score for Mean Average Precision (mAP) of 83%. These results are promising and indicate that the YOLO model can be used for automatic animal detection in various applications, such as wildlife monitoring, environmental protection or security systems.

Słowa kluczowe

EN

artificial neural networks; YOLOv5; transfer learning; genetic algorithm; thermal imaging

• Wydawca: Polish Academy of Sciences, Committee of Electronics and Telecommunication
• Czasopismo: International Journal of Electronics and Telecommunications
• Rocznik: 2023
• Tom: Vol. 69, No. 4
• Strony: 826--831
• Opis fizyczny: Bibliogr. 18 poz., fot., tab., wykr.

Twórcy

autor

Popek Łukasz

• Warsaw University of Technology, Faculty of Electronics and Information Technology; Sieć badawcza Rafał Perz, Poland

autor

Perz Rafał

• Warsaw University of Technology,Faculty of Power and Aeronautical Engineering; Sieć badawcza Rafał Perz, Poland

autor

Galiński Grzegorz

• Warsaw University of Technology, Faculty of Electronics and Information Technology

autor

Abratański Artur

• Warsaw University of Technology,Faculty of Power and Aeronautical Engineering; Sieć badawcza Rafał Perz, Poland

Bibliografia

• [1] J. Witczuk, S. Pagacz, A. Zmarza, and M. Cypel, “Exploring the feasibility of unmanned aerial vehicles and thermal imaging for ungulate surveys in forests - preliminary results. International journal of remote sensing,” International Journal of Remote Sensing, vol. 39, no. 15-16, pp. 5504-5521, 2018. [Online]. Available: https://doi.org/10.1080/01431161.2017.1390621
• [2] A. Vecvanags, K. Aktas, I. Pavlovs, E. Avots, J. Filipovs, B. A., G. Done, D. Jakovels, and G. Anbarjafari, “Ungulate detection and species classification from camera trap images using reti-nanet and faster r-cnn,” Entropy, vol. 24, no. 3, p. 353, 2022. [Online]. Available: https://doi.org/10.3390/e24030353
• [3] M. Choiński, M. Rogowski, P. Tynecki, D. P. J. Kuijper, M. Churski, and J. W. Bubnicki, “A first step towards automated species recognition from camera trap images of mammals using ai in a european temperate forest,” pp. 299-310, 2021. [Online]. Available: https://doi.org/10.1007/978-3-030-84340-3 24
• [4] M. Ivašić-Kos, M. Krišto, and M. Pobar, “Human detection in thermal imaging using yolo,” 2019, 5th International Conference on Computer and Technology Applications, pp. 19-24. [Online]. Available:https://doi.org/10.1145/3323933.3324076
• [5] M. Krišto, M. Ivasic-Kos, and M. Pobar, “Thermal object detection in difficult weather conditions using yolo,” IEEE Access, vol. PP, no. 3, pp. 125 459-125 476, 2020. [Online]. Available: https://doi.org/10.1109/ACCESS.2020.3007481
• [6] I. R., S. H. Mudumba, and H. R. Adkay, M. Nandi Vardhan, “Human detection in thermal imaging using yolo,” 2020, object Detection Using Thermal Imaging, 17th India Council International Conference (INDICON), pp. 19-24, New Delhi, India. [Online]. Available: https://doi.org/10.1145/3323933.3324076
• [7] A. Ulhaq, P. Adams, T. Cox, L. T. Khan, A., and M. Paul, “Automated detection of animals in low-resolution airborne thermal imagery,” Remote Sensing, vol. PP, no. 3, pp. 125 459-125 476, 2021. [Online]. Available: https://doi.org/10.1109/ACCESS.2020.3007481
• [8] Popek, Ł., Perz, R., and Galiński, G., “Comparison of different methods of animal detection and recognition on thermal camera images,” Electronics, vol. 12, no. 270, pp. 125 459-125 476, 2023. [Online]. Available: https://doi.org/10.3390/electronics12020270
• [9] J. Cilulko, P. Janiszewski, and M. e. a. Bogdaszewski, “Infrared thermal imaging in studies of wild animals,” European Journal of Wildlife Researche, vol. 59, no. 270, pp. 17-23, 2013. [Online]. Available: https://doi.org/10.1007/s10344-012-0688-1
• [10] L. Tan, T. Huangfu, and L. e. a. Wu, “Comparison of retinanet, ssd, and yolo v3 for real-time pill identification,” BMC Med Inform Decis Mak, 2021. [Online]. Available: https://doi.org/10.1186/s12911-021-01691-8
• [11] J. Glen, “Yolov5 by ultralytics (version 7.0) [computer software],” 2014, access: 13.06.2023. [Online]. Available: https://doi.org/10.5281/zenodo.3908559
• [12] Isa, I. S., Rosli, M. S. A, Yusof, U. K., Maruzuki, M. I. F., and Sulaiman, S. N., “Optimizing the hyperparameter tuning of yolov5 for underwater detection,” IEEE Access, vol. 10, pp. 52 818-52 831, 2022. [Online]. Available: https://doi.org/10.1109/ACCESS.2022.3174583
• [13] K. You, M. Long, J. Wang, and M. I. Jordan, “How does learning rate decay help modern neural networks?” arXiv preprint arXiv:1908.01878, 2019. [Online]. Available: https://doi.org/10.48550/arXiv.1908.01878
• [14] B. Lim, S. Zohren, and S. Roberts, “Enhancing time-series momentum strategies using deep neural networks,” The Journal of Financial Data Science, 2019. [Online]. Available: https://doi.org/10.3905/jfds.2019.1.015
• [15] T. M. Breuel, “The effects of hyperparameters on sgd training of neural networks,” arXiv preprint arXiv:1508.02788, 2015. [Online]. Available: https://doi.org/10.48550/arXiv.1508.02788
• [16] I. K. M. Jais, A. R. Ismail, and S. Q. Nisa, “Adam optimization algorithm for wide and deep neural network,” Knowledge Engineering and Data Science, vol. 2, no. 1, pp. 41-46, 2019.
• [17] E. Bisong, “Google colaboratory. in building machine learning and deep learning models on google cloud platform,” 2019.
• [18] Q. Xu, Z. Zhu, H. Ge, Z. Zhang, and X. Zang, “Effective face detector based on yolov5 and superresolution reconstruction.” Computational and mathematical methods in medicine, 2021. [Online]. Available: https://doi.org/10.1155/2021/7748350

Uwagi

1. Opracowanie rekordu ze środków MNiSW, umowa nr SONP/SP/546092/2022 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2024).

2. The authors would like to thank experts for their appropriate and constructive suggestions to improve this article.