Image Processing Based Deep Learning Approach for Classifying Hybrid Filler Ratio in Glass Fiber Reinforced Polymer

Year 2026, Volume: 9 Issue: 2, 980 - 988, 15.03.2026

Hüseyin Köse , İsmail Bayar

https://doi.org/10.34248/bsengineering.1891391

https://izlik.org/JA36DF24AB

Abstract

In this study, an image processing-based deep learning approach is developed to classify the filler ratios in glass fiber reinforced polymer composite with hybrid MgO-CuO nanoparticles in a non-contact and non-destructive manner, based solely on surface color and texture information. The originality of the study lies in its reliance on the direct learning of hybrid nanoparticle dopant ratios from optical surface properties, unlike image-based methods in the literature that mainly focus on damage detection, phase separation, or mechanical property estimation. In this context, eight different composite classes with different MgO and CuO weight ratios were produced and the samples were imaged at high resolution under homogeneous LED illumination at a fixed camera-sample distance. The obtained images were evaluated in a multi-class classification problem using the transfer learning-based EfficientNet-B0 architecture without any data enhancement. Model performance was analyzed with accuracy, sensitivity, recall, and F1-score metrics. Test results show that the proposed model achieved 97% overall accuracy and a macro-mean F1-score of 0.97; The study demonstrated that single-dopant and high-contrast hybrid systems were classified with high accuracy rate. Limited class overlap was observed in some hybrid classes with low CuO content, which is attributed to the optical similarity of the dopants. The findings reveal that deep learning approaches based on surface images offer a powerful, low-cost, and industrially viable alternative for dope ratio verification and rapid quality control applications in composite manufacturing processes.

Keywords

GFRP , Composite , Image processing , Deep learning , Hybrid filler classification

Ethical Statement

Ethics committee approval was not required for this study because of there was no study on animals or humans.

References

• Badran, A., Marshall, D., Legault, Z., Makovetsky, R., Provencher, B., Piché, N., & Marsh, M. (2020). Automated segmentation of computed tomography images of fiber-reinforced composites by deep learning. Journal of Materials Science, 55(34), 16273-16289. https://doi.org/10.1007/s10853-020-05148-7
• Bobovich, B. B. (2019). Glass-Fiber Reinforced Plastics — Construction Materials of the Sixth Technological Paradigm? Glass and Ceramics, 76(1), 38-41. https://doi.org/10.1007/s10717-019-00128-z
• Chrispin Laila, A., Narayanan, M., Bhadrakumar Sindhu, D., & AlbyRoy, A. (2022). Mechanical properties of polymer matrix/glass fiber composites containing metal/hybrid nanoparticles-an overview. High Performance Polymers, 34(8), 859-870. https://doi.org/10.1177/09540083221094964
• Fan, S., Zhang, J., Wang, B., Chen, J., Yang, W., Liu, W., & Li, Y. (2023). A deep learning method for fast predicting curing process-induced deformation of aeronautical composite structures. Composites Science and Technology, 232, 109844. https://doi.org/https://doi.org/10.1016/j.compscitech.2022.109844
• Hart, G. L. W., Mueller, T., Toher, C., & Curtarolo, S. (2021). Machine learning for alloys. Nature Reviews Materials, 6(8), 730-755. https://doi.org/10.1038/s41578-021-00340-w
• Hu, Q., Wei, X., Guo, H., Xu, H., Li, C., He, W., & Pei, B. (2023). Study on intelligent and visualization method of ultrasonic testing of composite materials based on deep learning. Applied Acoustics, 207, 109363. https://doi.org/https://doi.org/10.1016/j.apacoust.2023.109363
• Karnati, S. R., Agbo, P., & Zhang, L. (2020). Applications of silica nanoparticles in glass/carbon fiber-reinforced epoxy nanocomposite. Composites Communications, 17, 32-41. https://doi.org/https://doi.org/10.1016/j.coco.2019.11.003
• Kose, H., Bayar, I., & Ergün, R. K. (2024). Experimental optimization of CuO and MgO hybrid nanoparticle reinforcement ratios to enhance fatigue life of GFRP composites. Polymer Composites, 45(12), 11125-11137. https://doi.org/https://doi.org/10.1002/pc.28536
• Li, C., & Zheng, K. (2023). Methods, progresses, and opportunities of materials informatics. InfoMat, 5(8), e12425. https://doi.org/https://doi.org/10.1002/inf2.12425
• Ma, H.-l., Jia, Z., Lau, K.-t., Leng, J., & Hui, D. (2016). Impact properties of glass fiber/epoxy composites at cryogenic environment. Composites Part B: Engineering, 92, 210-217. https://doi.org/https://doi.org/10.1016/j.compositesb.2016.02.013
• Maghami, A., Salehi, M., & Khoshdarregi, M. (2021). Automated vision-based inspection of drilled CFRP composites using multi-light imaging and deep learning. CIRP Journal of Manufacturing Science and Technology, 35, 441-453. https://doi.org/https://doi.org/10.1016/j.cirpj.2021.07.015
• Naveen, J., Babu, M. S., & Sarathi, R. (2021). Impact of MgO nanofiller-addition on electrical and mechanical properties of glass fiber reinforced epoxy nanocomposites. Journal of Polymer Research, 28(10), 377. https://doi.org/10.1007/s10965-021-02746-0
• Sathishkumar, T., Satheeshkumar, S., & Naveen, J. (2014). Glass fiber-reinforced polymer composites – a review. Journal of Reinforced Plastics and Composites, 33(13), 1258-1275. https://doi.org/10.1177/0731684414530790
• Shim, Y.-B., Lee, I. Y., & Park, Y.-B. (2024). Predicting the material behavior of recycled composites: Experimental analysis and deep learning hybrid approach. Composites Science and Technology, 249, 110464. https://doi.org/https://doi.org/10.1016/j.compscitech.2024.110464