Intelligent Defect Detection for Aircraft Composite Materials: Integrating Infrared Thermography and YOLO-Based Deep Learning

Kuo-Chien  Liao; Jirayu  Lau; Yu-Chiang  Hsu

doi:10.6180/jase.202603_29(4).0011

Intelligent Defect Detection for Aircraft Composite Materials: Integrating Infrared Thermography and YOLO-Based Deep Learning

Kuo-Chien Liao¹This email address is being protected from spambots. You need JavaScript enabled to view it., Jirayu Lau², and Yu-Chiang Hsu²

¹Graduate Institute of Aeronautics, Chaoyang University of Technology, Taiwan

²Department of Aeronautical Engineering, Chaoyang University of Technology, Taiwan

Received: May 16, 2025
Accepted: July 13, 2025
Publication Date: August 1, 2025

Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.

Download Citation: ||https://doi.org/10.6180/jase.202603_29(4).0011

This paper presents a defect detection system for aircraft composite materials that integrates active infrared thermography with a lightweight YOLO11n deep learning model. The proposed approach utilize the heat conduction behavior described by Fourier’s Law to identify subsurface defects such as delamination and disbonds by visualizing heat flow anomalies under controlled thermal excitation. To enhance mobility and practical deployment, the system combines portable thermal cameras, UAV-mounted thermal imaging, and real-time detection pipelines implemented through a custom iOS application. Results show that the mobile detection pipeline achieves real-time performance with an average speed of approximately 24 FPS, while UAV inspections are limited by battery life and network transmission delays. This study demonstrates the feasibility of combining thermography with mobile deep learning to support flexible, efficient inspections during routine base maintenance and highlights key areas for future optimization.

Keywords: Aircraft Composite Materials; Infrared Thermography; YOLO11n; Deep Learning; Non-Destructive Testing (NDT); UAV Inspection; Real-Time Defect Detection

[1] A. Baker, D. Kelly, and S. Dutton. Composite Materials for Aircraft Structures. 3rd ed. AIAA Education Series. AIAA, 2016.
[2] X. P. V. Maldague. Theory and Practice of Infrared Technology for Nondestructive Testing. 2nd ed. Wiley, 2019.
[3] ASTM International. ASTM E2582–07: Standard Practice for Infrared Flash Thermography of Composite Panels. 2012.
[4] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi. “You Only Look Once: Unified, Real-Time Object Detection”. In: Proceedings of the IEEE Conference on Com puter Vision and Pattern Recognition (CVPR). 2016, 779–788.
[5] Ultralytics. YOLO Documentation. Accessed: 2025. 2025. URL: https: //docs.ultralytics.com.
[6] N. P. Avdelidis, D. P. Almond, A. Dobmann, and C. C. B. Pang, (2003) “Infrared Thermography and Thermal Non-Destructive Testing" Insight– Non Destructive Testing and Condition Monitoring 45(9): 616–621.
[7] Z. Hidayat, N. P. Avdelidis, and H. Fernandes, (2025) “Brief Review of Vibrothermography and Optical Thermography for Defect Quantification in CFRP Material" Sensors 25(6): 1847.
[8] K.-C. Liao, J.-L. Liou, M. Hidayat, H.-T. Wen, and H.-Y. Wu, (2024) “Detection and Analysis of Aircraft Composite Material Structures Using UAV" Inventions 9(3): 47. DOI: 10.3390/inventions9030047.
[9] A. Hameed, (2021) “Effect of Illumination and Heat Flow on Thermographic Defect Detection" Journal of Composite Structures 280: 114882.
[10] FLIR Systems. FLIR ONE Pro User Manual. Accessed: 2025. URL: https://www.flir.com/products/flir-one pro/.
[11] DJI. DJI Mavic 2 Enterprise Series Specifications. Accessed: 2025. URL: https: //www.dji.com/mavic-2-enterprise.
[12] R. Wilson and L. Chen, (2023) “Considerations for thermal imaging in deep learning-based defect detection" Ma chine Vision and Applications 34(5): 1234–1256.
[13] Federal Aviation Administration (FAA). Advisory Circular AC 43.13–1B: Acceptable Methods, Techniques, and Practices—Aircraft Inspection and Repair. Accessed: 2025. URL: https://www.faa.gov/documentLibrary/ media/Advisory_Circular/AC_43.13-1B.pdf.
[14] C. J. Hellier. Handbook of Nondestructive Evaluation. 2nd ed. McGraw-Hill, 2012.
[15] Roboflow. Roboflow Documentation. Accessed: 2025. URL: https: //roboflow.com/.
[16] Apple Developer. Core ML Framework Documentation. Accessed: 2025. URL: https: //developer.apple.com/documentation/coreml.
[17] Apple Developer. AVFoundation Framework Documentation. Accessed: 2025. URL: https: //developer.apple.com/av-foundation/.
[18] FLIRSystems. Flir mobilesdk documentation. Accessed: 10 October 2024. 2024. URL: https: //developer.flir.com/mobile-sdk/.
[19] J. Gillespie, (2021) “Aerial Thermographic Inspection Using UAVs" Journal of Nondestructive Testing 38(2): 45–52.
[20] Raspberry Pi Foundation. Raspberry Pi 4 Model B Product Page. Accessed: 2025. URL: https://www.raspberrypi.com/products/raspberry-pi-4-model b/.
[21] Canonical Ltd. Ubuntu 20.04.6 LTS (Focal Fossa). Accessed: 15 October 2024. 2024. URL: https: //releases.ubuntu.com/20.04/.
[22] Arut. Nginx-based media streaming server. Accessed: 15 October 2024. 2024. URL: https: //github.com/arut/nginx-rtmp-module.
[23] S. Ren, K. He, R. Girshick, and J. Sun. “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks”. In: Advances in Neural Informa tion Processing Systems (NeurIPS). 2015.
[24] T.-Y. Lin, P. Goyal, R. Girshick, K. He, and P. Dollár. “Focal Loss for Dense Object Detection”. In: Proceed ings of the IEEE International Conference on Computer Vision (ICCV). 2017.
[25] Z. Liu, Y. Lin, Y. Cao, H. Hu, Y. Wei, Z. Zhang, S. Lin, and B. Guo. “Swin Transformer: Hierarchical Vision Transformer using Shifted Windows”. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). 2021.
[26] NVIDIA. CUDA Toolkit Documentation. Accessed: 2025. URL: https: //docs.nvidia.com/cuda/.
[27] NVIDIA. cuDNN Documentation. Accessed: 2025. URL: https: //docs.nvidia.com/deeplearning/cudnn/.
[28] T. Akiba, S. Sano, T. Yanase, T. Ohta, and M. Koyama. “Optuna: A Next-generation Hyperparameter Optimization Framework”. In: Proceedings of the 25th ACMSIGKDD International Conference on Knowledge Discovery and Data Mining. 2019, 2623–2631.
[29] Apple Inc. iPhone 12 Pro Max Technical Specifications. Accessed: 2025. URL: https://www.apple.com/ iphone-12-pro/specs/.
[30] DJI. Dji Mavic 2 Enterprise Specifications. Accessed: 15 October 2024. 2024. URL: https://www.dji.com/ mavic-2-enterprise.