SavedAdd your journal

Journals / Bayburt Üniversitesi Fen Bilimleri Dergisi / 2026 / Cilt: 9 Sayı: 1

Deep Learning-Based Material Classification Using Thermal and Texture-Based Hybrid Images

Termal ve Doku Bileşenli Hibrit Görüntüler Kullanılarak Derin Öğrenme Tabanlı Malzeme Sınıflandırması

Büşra Tiken(BAYBURT ÜNİVERSİTESİ, UYGULAMALI BİLİMLER FAKÜLTESİ)0009-0004-9426-3895
Mustafa Alptekin Engin(BAYBURT ÜNİVERSİTESİ)0000-0003-3399-9343
Pages
55–65
Publication Date
2026-06-30
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The results indicate that the use of low-ratio thermal information together with texture information can improve material classification performance.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:This study aimed to investigate the performance of deep learning-based material classification using thermal and texture-based hybrid images of different materials.
📊
Key Findings & Data:The materials were imaged at room temperature, after 10 seconds of heating, and after 20 seconds of heating.
💡
Conclusion & Impact:The results indicate that the use of low-ratio thermal information together with texture information can improve material classification performance.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Makine öğrenmesi · Hibrit görüntüleme · Malzeme sınıflandırması

Abstract

This study aimed to investigate the performance of deep learning-based material classification using thermal and texture-based hybrid images of different materials. For this purpose, a controlled experimental setup was established, and images of eight different materials, namely aluminium foil, paper, cardboard, fabric, metal, nylon, chipboard, and wood, were obtained. The materials were imaged at room temperature, after 10 seconds of heating, and after 20 seconds of heating. In addition, each condition was recorded under both illuminated and non-illuminated environments. The obtained images were combined with texture images to include thermal information at ratios of 25%, 50%, 75%, and 100%, and four different datasets were created by dividing them into 128×128-pixel sub-images. In the classification stage, VGG16, VGG19, DenseNet201, InceptionV3, MobileNetV2, and ResNet50V2 architectures were applied using the transfer learning approach, and the accuracy values of the models were compared. The findings showed that the thermal component ratio had a significant effect on classification performance. The highest accuracy value was obtained as 0.95 with the VGG19 architecture on the dataset consisting of hybrid images containing a 25% thermal component. The results indicate that the use of low-ratio thermal information together with texture information can improve material classification performance.

Özet

Bu çalışmada, farklı malzemelere ait termal ve doku bileşenli hibrit görüntüler kullanılarak derin öğrenme tabanlı malzeme sınıflandırma performansının araştırılması amaçlanmıştır. Bu amaç doğrultusunda kontrollü bir deney düzeneği oluşturulmuş ve alüminyum folyo, kâğıt, karton, kumaş, metal, naylon, sunta ve tahta olmak üzere sekiz farklı materyale ait görüntüler elde edilmiştir. Materyaller oda sıcaklığında, 10 saniye ısıtma sonrasında ve 20 saniye ısıtma sonrasında görüntülenmiştir. Ayrıca her durum aydınlatmalı ve aydınlatmasız koşullarda kaydedilmiştir. Elde edilen görüntüler; %25, %50, %75 ve %100 oranlarında termal bilgi içerecek şekilde doku görüntüleriyle birleştirilmiş ve 128×128 piksel boyutlarında alt görüntülere ayrılarak dört farklı veri seti oluşturulmuştur. Sınıflandırma aşamasında VGG16, VGG19, DenseNet201, InceptionV3, MobileNetV2 ve ResNet50V2 mimarileri transfer öğrenme yaklaşımıyla uygulanmış ve modellerin doğruluk değerleri karşılaştırılmıştır. Elde edilen bulgular, termal bileşen oranının sınıflandırma başarımı üzerinde belirgin bir etkiye sahip olduğunu göstermiştir. En yüksek doğruluk değeri, %25 termal bileşen içeren hibrit görüntülerden oluşan veri setinde VGG19 mimarisi ile 0,95 olarak elde edilmiştir. Sonuçlar, düşük oranlı termal bilginin doku bilgisiyle birlikte kullanılmasının malzeme sınıflandırma performansını artırabileceğini göstermektedir.

🔖

Cite This Article

Tiken, B., Engin, M. A. (2026). Deep Learning-Based Material Classification Using Thermal and Texture-Based Hybrid Images. Bayburt Üniversitesi Fen Bilimleri Dergisi, Cilt: 9 Sayı: 1, 55–65. https://doi.org/10.55117/bufbd.1953520

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (28)

  1. R. M. Haralick, K. Shanmugam, and I. Dinstein, “Textural features for image classification,” IEEE Trans. Syst. Man Cybern., vol. SMC-3, no. 6, pp. 610–621, Nov. 1973.
  2. R. Mapari, S. Narkhede, A. Navale, and J. Babrah, “Automatic waste segregator and monitoring system,” Int. J. Adv. Comput. Res., vol. 10, no. 49, pp. 172–181, 2020.
  3. H. Mahami et al., “Material recognition for automated progress monitoring using deep learning methods,” arXiv preprint, arXiv:2006.16344, 2020.
  4. A. Ghosh, M. Ehrlich, L. Davis, and R. Chellappa, “Unsupervised super-resolution of satellite imagery for high fidelity material label transfer,” in IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium, 2019.
  5. Y. Cho, N. Bianchi-Berthouze, N. Marquardt, and S. J. Julier, “Deep thermal imaging,” in Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems, 2018.
  6. J. Govindarajan, “A case for joint development of IR cameras in India,” Int. J. Intell. Electron. Syst., vol. 1, no. 1, pp. 26–30, Nov. 2007.
  7. C. Hildebrandt, C. Raschner, and K. Ammer, “An overview of recent application of medical infrared thermography in sports medicine in Austria,” Sensors (Basel), vol. 10, no. 5, pp. 4700–4715, May. 2010.
  8. R. Usamentiaga, D. F. Garcia, J. Molleda, F. G. Bulnes, and V. G. Orgeira, “Temperature tracking system for sinter material in a rotatory cooler based on infrared thermography,” IEEE Trans. Ind. Appl., vol. 50, no. 5, pp. 3095–3102, Feb. 2014.
  9. R. Vadivambal and D. S. Jayas, “Applications of thermal imaging in agriculture and food industry—A review,” Food Bioproc. Tech., vol. 4, no. 2, pp. 186–199, Feb. 2011.
  10. M. B. Lopez, C. R. del-Blanco, and N. Garcia, “Detecting exercise-induced fatigue using thermal imaging and deep learning,” in 2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA), 2017.
  11. G. D. Boreman, Modulation Transfer Function in Optical and Electro-optical Systems. Bellingham, WA: SPIE Press, 2001.
  12. M. Kontos, R. Wilson, and I. Fentiman, “Digital infrared thermal imaging (DITI) of breast lesions: sensitivity and specificity of detection of primary breast cancers,” Clin. Radiol., vol. 66, no. 6, pp. 536–539, Jun. 2011.
  13. M. Pergeorelis, T. Rust, and C. Kambhamettu, “Open dataset and enhancement method for long-wave thermal diurnal material classification,” in 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2025, pp. 4595–4601.
  14. T.Y. Chen et al., “Artificial intelligence system combining with infrared thermography and visible image for abnormal temperature detection and floor material identification,” IEEE Sens. J., vol. 24, no. 24, pp. 42181–42194, Aug. 2024.
  15. J.-R. Riba, T. Canals, and R. Cantero, “Recovered paperboard samples identification by means of mid-infrared sensors,” IEEE Sens. J., vol. 13, no. 7, pp. 2763–2770, Apr. 2013.
  16. E. Kerr, T. M. McGinnity, and S. Coleman, “Material classification based on thermal properties — A robot and human evaluation,” in 2013 IEEE International Conference on Robotics and Biomimetics (ROBIO), 2013.
  17. L. Wu, S. Liu, and Y. Wu, “Infrared imaging detection of water permeation on field large-scale rock relics,” IEEE Trans. Geosci. Remote Sens., vol. 49, no. 1, pp. 581–590, Jul. 2011.
  18. H. Li, X.-J. Wu, and J. Kittler, “Infrared and visible image fusion using a deep learning framework,” in 2018 24th International Conference on Pattern Recognition (ICPR), 2018.
  19. H. Xu, J. Ma, J. Jiang, X. Guo, and H. Ling, “U2Fusion: A unified unsupervised image fusion network,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 44, no. 1, pp. 502–518, Jul. 2022.
  20. W. Xiao, Y. Zhang, H. Wang, F. Li, and H. Jin, “Heterogeneous knowledge distillation for simultaneous infrared-visible image fusion and super-resolution,” IEEE Trans. Instrum. Meas., vol. 71, pp. 1–15, Feb. 2022.
  21. W. Großmann, H. Horn, and O. Niggemann, “Improving remote material classification ability with thermal imagery,” Sci. Rep., vol. 12, no. 1, p. 17288, Oct. 2022.
  22. Y. Lecun, L. Bottou, Y. Bengio, and P. Haffner, “Gradient-based learning applied to document recognition,” Proc. IEEE Inst. Electr. Electron. Eng., vol. 86, no. 11, pp. 2278–2324, Nov. 1998.
  23. F. Zhuang et al., “A comprehensive survey on transfer learning,” Proc. IEEE Inst. Electr. Electron. Eng., vol. 109, no. 1, pp. 43–76, Jul. 2021.
  24. K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” arXiv [cs.CV], 2014.
  25. G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, “Densely connected convolutional networks,” in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017.
  26. C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna, “Rethinking the inception architecture for computer vision,” in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016.
  27. M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L.-C. Chen, “MobileNetV2: Inverted residuals and linear bottlenecks,” in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018.
  28. K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 770–778.