ISSN 0474-8662. Information Extraction and Processing. 2022. Issue 50 (126)
Home Back to issue
Application of YOLOX deep learning model for automated object detection on thermograms
Skladchykov I. O.
Igor Sikorsky Kyiv Polytechnic Institute, Kyiv
Momot A. S.
Igor Sikorsky Kyiv Polytechnic Institute, Kyiv
Galagan R. M.
Igor Sikorsky Kyiv Polytechnic Institute, Kyiv
Bohdan H. A.
Igor Sikorsky Kyiv Polytechnic Institute, Kyiv
Trotsiuk K. М.
Igor Sikorsky Kyiv Polytechnic Institute, Kyiv

https://doi.org/10.15407/vidbir2022.50.069

Keywords: thermal monitoring, deep learning, object detection.

Cite as: Skladchykov I. O., Momot A. S., Galagan R. M., Bohdan H. A., Trotsiuk K. М. Application of YOLOX deep learning model for automated object detection on thermograms. Information Extraction and Processing. 2022, 50(126), 69-77. DOI:https://doi.org/10.15407/vidbir2022.50.069

Abstract

A method of automating the data analysis of thermal imaging systems in the field of safety control is proposed. It has been established that today video surveillance technologies have a number of disadvantages that can be eliminated by using thermal imaging cameras. Analysis of infrared images can be automated in order to reduce percentage of false positives and increase the effi-ciency of thermal imaging video surveillance systems. A high level of interference, unclear object contours and low image resolution are real problems in automating the object detecting process on thermographic images. The traditional and promising methods of thermograms analysis and approaches that can lead to creating the automated thermal video surveillance systems are discussed. It is proposed to use deep learning, which in recent years has proven itself as an effective way of image analysis. The study is based on review of existing works, as methods of automating object detection process on thermograms. It is proposed to use YOLOX as a deep learning model. This model has one of the best quality indicators and speed processing input parameters on standard datasets. FLIR’s Thermal Starter annotated set of thermal images is used to train the model. The value of mAP at the level 55% is obtained according the results of model training for recognizing 4 classes of objects on thermograms. Different advantages and disadvantages of this development are analyzed. Ways of further improvement of the neural network method of automation of thermal imaging safety control systems have been determined.

References

    1. Wong, L.; Wai, K. An effective surveillance system using thermal camera. In IEEE 2009 Interna-tional Conf. on Signal Acquisition and Processing, Kuala Lumpur, Maylaysia, 3-5 April 2009, pp. 13-17. https://doi.org/10.1109/ICSAP.2009.12

    2. Alokhina, O.V.; Ivchenko, D.V.; Pits, N.A. Thermal remote sensing data analysis in monitoring of natural objects. Information Extraction and Processing, 2020, 48(124), 61-71. https://doi.org/10.15407/vidbir2020.48.061

    3. Muraviov, O.V.; Petryk, V.F.; Lysenko, I.I.; Bohdan, H.A.; Nakonechna, A.V. Automatization of thermographic diagnostic method of human body pathologies. Таuridа Scientific Herald. Series: Technical Sciences, 2022, 1, 47-53. (in Ukrainian)

    4. Kosarevych, R.Ya.; Alokhina, O.V.; Rusyn, B.P.; Lutsyk, O.A.; Pits, N.A.; Ivchenko, D.V. Analysis of remote sensing images by methods of convolutional neural networks and marked random point fields. Information Extraction and Processing, 2021, 49(125), 45-51. https://doi.org/10.15407/vidbir2021.49.045

    5. Zhivkovic, A.; Muravyov, A. Modern technologies of non-contact temperature measurement. Nauka i studia, 2020. (in Russian) 6. Xia, C.; Ren, M. Infrared thermography based diagnostics on power equipment: State of the art. High Voltage, 2020, 6, 387-407. https://doi.org/10.1049/hve2.12023

    7. Galagan, R.; Momot, A. Analysis of application of neural networks to improve the reliability of active thermal NDT. KPI Science News, 2019, 1, 7-14. https://doi.org/10.20535/kpi-sn.2019.1.157374

    8. Skladchikov, I.; Momot, A. Using the MLX90640 sensor as part of a smart thermographic camera. In Efficiency and automation of engineering solutions in instrument construction, 2020, 16, 240-242. (in Ukrainian)

    9. Momot, A.; Skladchykov, I. Deep learning automated data analysis of security infrared cameras. Slovak International Scientific Journal, 2021, 52, 13-16.

    10. Trask, A. Grokking Deep Learning. Manning Publications, 2019.

    11. Mingxing, T.; Ruoming, P. EfficientDet: Scalable and Efficient Object Detection. IEEE Xplore, 2021, 10781-10790.

    12. FLIR Thermal Images Dataset. 2021. https://www.kaggle.com/datasets/albertofv/flir-thermal-images-dataset-reduced (accessed 2022-07-08)

    13. Zheng, G.; Songtao, L. Exceeding yolo series in 2021. arXiv preprint arXiv, 2021, # 2107.08430.

    14. Zheng, G.; Songtao, L. Exceeding yolo series in 2021. 2021a. https://github.com/Megvii-BaseDetection/YOLOX. (accessed 2022-07-08)

    15. Storozhyk, D.V.; Muraviov, O.V.; Protasov, A.G.; Bazhenov, V.G.; Bohdan, G.A. Multispectral image combining as a method of information content increasing at binary segmentation. KPI Science News, 2020, 82-87. (in Ukrainian) https://doi.org/10.20535/kpi-sn.2020.2.197955

    16. Cao, J.; Wu, W.; Wang, R.; Kwong, S. No-reference image quality assessment by using convolutional neural networks via object detection. International Journal of Machine Learning and Cybernetics, 2022, 1-12.