A comprehensive review of automatic defect detection in wooden surface inspection

Abstract

This review focused on automated detection and classification of wood surface defects, such as knots, cracks, and resin pockets. Machine learning and deep learning methods are increasingly replacing traditional inspection techniques, such as manual checks and imaging tools like X-rays and ultrasound. However, these conventional approaches suffer from significant limitations in accuracy and efficiency. We reviewed both single-stage models, such as you only look once (YOLO) and its variants, and two-stage models like faster region-based convolutional neural networks (R-CNN), and examined their strengths, limitations, and relevance in real industrial applications. We further investigated emerging zero-shot learning approaches that use vision-language models and NLP techniques to detect previously unseen wood defects. Zero-shot models do not rely on annotated training data. These methods offer scalable and flexible solutions, especially in scenarios where collecting labelled samples for all defect types is impractical. The paper also discussed publicly available labelled datasets used to train and test these models, and discussed standard performance metrics such as precision, recall, and mean average precision. This review aimed to support further research and practical improvements in automated wood surface quality assessment by analyzing defect types, detection methods, and evaluation techniques.

Keywords

wood defect detection / deep learning / machine learning / YOLO model / single-stage model / two-stage model

References

[1] C. Ünsalan, Wood surface inspection using structural and conditional statistical features, arXiv: 2407.03630. https://doi.org/10.48550/arXiv.2407.03630

[2] M. Mohsin, O. S. Balogun, K. Haataja, P. Toivanen, Real-time defect detection and classification on wood surfaces using deep learning, Electronic Imaging, 34 (2022), IPAS-382. https://doi.org/10.2352/EI.2022.34.10.IPAS-382

[3] H. An, Z. Liang, M. Qin, Y. Huang, F. Xiong, G. Zeng, Wood defect detection based on the CWB-YOLOv8 algorithm, J. Wood Sci., 70 (2024), 26. https://doi.org/10.1186/s10086-024-02139-z

[4] H. C. Teo, U. R. Hashim, S. Ahmad, L. Salahuddin, H. C. Ngo, K. Kanchymalay, A review of the automated timber defect identification approach, International Journal of Electrical and Computer Engineering, 13 (2023), 2156-2166. https://doi.org/10.11591/ijece.v13i2.pp2156-2166

[5] Z. Yi, L. Luo, Q. Lu, M. Chen, W. Zhu, Y. Zhang, An efficient and accurate surface defect detection method for quality supervision of wood panels, Meas. Sci. Technol., 35 (2024), 055209. https://doi.org/10.1088/1361-6501/ad26c9

[6] B. Syla, S. Saleh, W. Hardt, Automated identification of wood surface defects based on deep learning, Embedded Selforganising Systems, 10 (2023), 55-61. https://doi.org/10.14464/ess.v10i7.627

[7] J. Redmon, S. Divvala, R. Girshick, A. Farhadi, You only look once: unified, real-time object detection, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, 779-788. https://doi.org/10.1109/CVPR.2016.91

[8] S. Ren, K. He, R. Girshick, J. Sun, Faster R-CNN: towards real-time object detection with region proposal networks, IEEE Trans. Pattern Anal., 39 (2017), 1137-1149. https://doi.org/10.1109/TPAMI.2016.2577031

[9] X. Su, Y. Zhang, H. Luo, X. Liu, L. Huang, Mistake notebook learning: selective batch-wise context optimization for in-context learning, arXiv: 2512.11485. https://doi.org/10.48550/arXiv.2512.11485

[10] Q. Zhang, L. Liu, Z. Yang, J. Yin, Z. Jing, Wlsd-yolo: a model for detecting surface defects in wood lumber, IEEE Access, 12 (2024), 65088-65098. https://doi.org/10.1109/ACCESS.2024.3395623

[11] Y. Guo, W. Cao, Wood surface defect detection using improved deep learning algorithm: FRCE-YOLO, Proceedings of 6th International Conference on Electronic Engineering and Informatics (EEI), 2024, 324-328. https://doi.org/10.1109/EEI63073.2024.10696054

[12] R. Wang, Y. Chen, F. Liang, B. Wang, X. Mou, G. Zhang, BPN-YOLO: a novel method for wood defect detection based on YOLOv7, Forests, 15 (2024), 1096. https://doi.org/10.3390/f15071096

[13] H. Xi, R. Wang, F. Liang, Y. Chen, G. Zhang, B. Wang, SiM-YOLO: a wood surface defect detection method based on the improved YOLOv8, Coatings, 14 (2024), 1001. https://doi.org/10.3390/coatings14081001

[14] P. Kodytek, A. Bodzás, P. Bilik, A large-scale image dataset of wood surface defects for automated vision-based quality control processes, F1000Res., 10 (2022), 581. https://doi.org/10.12688/f1000research.52903.2

[15] R. Li, Z. Xu, F. Yang, B. Yang, Defect detection for melamine-impregnated paper decorative particleboard surface based on deep learning, Wood Mater. Sci. Eng., in press. https://doi.org/10.1080/17480272.2024.2428963

[16] K. Simonyan, A. Zisserman, Very deep convolutional networks for large-scale image recognition, arXiv: 1409.1556. https://doi.org/10.48550/arXiv.1409.1556

[17] S. Koman, S. Feher, J. Abraham, R. Taschner, Effect of knots on the bending strength and the modulus of elasticity of wood, Wood Res., 58 (2013), 617-626.

[18] Y. Fang, X. Guo, K. Chen, Z. Zhou, Q. Ye, Accurate and automated detection of surface knots on sawn timbers using YOLO-V5 model, BioResources, 16 (2021), 5390-5406. https://doi.org/10.15376/biores.16.3.5390-5406

[19] J. van Blokland, A. Olsson, J. Oscarsson, G. Daniel, S. Adamopoulos, Crack formation, strain distribution and fracture surfaces around knots in thermally modified timber loaded in static bending, Wood Sci. Technol., 54 (2020), 1001-1028. https://doi.org/10.1007/s00226-020-01190-5

[20] J. Lyu, H. Qu, M. Chen, Influence of wood knots of chinese weeping cypress on selected physical properties, Forests, 14 (2023), 1148. https://doi.org/10.3390/f14061148

[21] P. Guindos, Comparison of different failure approaches in knotty wood, Drewno, 57 (2014), 51-68. https://doi.org/10.12841/wood.1644-3985.065.03

[22] R. Jockwer, E. Serrano, P. J. Gustafsson, R. Steiger, Impact of knots on the fracture propagating along grain in timber beams, Int. Wood Prod. J., 8 (2017), 39-44. https://doi.org/10.1080/20426445.2016.1275093

[23] A. Senalik, G. Schueneman, R. Ross, Ultrasonic-based nondestructive evaluation methods for wood: a primer and historical review, FPL-GTR-235, 2014. https://doi.org/10.2737/FPL-GTR-235

[24] N. Pahnabi, T. Schumacher, A. Sinha, Imaging of structural timber based on in situ radar and ultrasonic wave measurements: a review of the state-of-the-art, Sensors, 24 (2024), 2901. https://doi.org/10.3390/s24092901

[25] M. Zielińska, M. Rucka, Assessment of wooden beams from historical buildings using ultrasonic transmission tomography, Int. J. Archit. Herit., 17 (2023), 249-261. https://doi.org/10.1080/15583058.2022.2086505

[26] Y. Chen, C. Sun, Z. Ren, B. Na, Review of the current state of application of wood defect recognition technology, BioResources, 18 (2023), 2288-2302. https://doi.org/10.15376/biores.18.1.Chen

[27] X. Zou, C. Wu, H. Liu, Z. Yu, X. Kuang, An accurate object detection of wood defects using an improved faster R-CNN model, Wood Mater. Sci. Eng., 20 (2025), 413-419. https://doi.org/10.1080/17480272.2024.2352605

[28] A. Urbonas, V. Raudonis, R. Maskeliūnas, R. Damaševičius, Automated identification of wood veneer surface defects using faster region-based convolutional neural network with data augmentation and transfer learning, Appl. Sci., 9 (2019), 4898. https://doi.org/10.3390/app9224898

[29] R. Ehtisham, W. Qayyum, C. V. Camp, V. Plevris, J. Mir, Q. Z. Khan, et al., Classification of defects in wooden structures using pre-trained models of convolutional neural network, Case Stud. Constr. Mat., 19 (2023), e02530. https://doi.org/10.1016/j.cscm.2023.e02530

[30] H. Shah, N. Myller, C. Sedano, Forecasting daily customer flow in restaurants: a multifactor machine learning approach, Appl. Comput. Intell., 5 (2025), 168-190. https://doi.org/10.3934/aci.2025011

[31] H. Shah, P. Fränti, Combining statistical, structural, and linguistic features for keyword extraction from web pages, Appl. Comput. Intell., 2 (2022), 115-132. https://doi.org/10.3934/aci.2022007

[32] H. Shah, R. Mariescu-Istodor, P. Fränti, Webrank: language-independent extraction of keywords from webpages, Proceedings of IEEE International Conference on Progress in Informatics and Computing (PIC), 2021, 184-192. https://doi.org/10.1109/PIC53636.2021.9687047

[33] R. Wang, F. Liang, B. Wang, X. Mou, ODCA-YOLO: an omni-dynamic convolution coordinate attention-based YOLO for wood defect detection, Forests, 14 (2023), 1885. https://doi.org/10.3390/f14091885

[34] Y. Cui, S. Lu, S. Liu, Real-time detection of wood defects based on spp-improved yolo algorithm, Multimed. Tools Appl., 82 (2023), 21031-21044. https://doi.org/10.1007/s11042-023-14588-7

[35] L. Yi, M. Akbar, M. Wahab, B. Rosdi, M. Fauthan, N. Shrifan, The prospect of artificial intelligence-based wood surface inspection: a review, IEEE Access, 12 (2024), 84706-84725. https://doi.org/10.1109/ACCESS.2024.3412928

[36] S. Hwang, J. Sugiyama, Computer vision-based wood identification and its expansion and contribution potentials in wood science: a review, Plant Methods, 17 (2021), 47. https://doi.org/10.1186/s13007-021-00746-1

[37] J. Silva, R. Bordalo, J. Pissarra, P. de Palacios, Computer vision-based wood identification: a review, Forests, 13 (2022), 2041. https://doi.org/10.3390/f13122041

[38] A. Howard, M. Sandler, G. Chu, L. Chen, B. Chen, M. Tan, et al., Searching for MobileNetV3, Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2019, 1314-1324. https://doi.org/10.1109/ICCV.2019.00140

[39] M. Ali, U. Hashim, K. Kanchymalay, A. Wibawa, L. Salahuddin, R. Rahiddin, A review of recent deep learning applications in wood surface defect identification, IAES International Journal of Artificial Intelligence, 14 (2025), 1696. https://doi.org/10.11591/ijai.v14.i3.pp1696-1707

[40] K. He, X. Zhang, S. Ren, J. Sun, Deep residual learning for image recognition, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, 770-778. https://doi.org/10.1109/CVPR.2016.90

[41] C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, et al., Going deeper with convolutions, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015, 1-9. https://doi.org/10.1109/CVPR.2015.7298594

[42] A. Krizhevsky, I. Sutskever, G. E. Hinton, Imagenet classification with deep convolutional neural networks, Commun. ACM, 60 (2017), 84-90. https://doi.org/10.1145/3065386

[43] L. Zhang, Y. Bian, P. Jiang, F. Zhang, A transfer residual neural network based on ResNet-50 for detection of steel surface defects, Appl. Sci., 13 (2023), 5260. https://doi.org/10.3390/app13095260

[44] CRAN, Available CRAN packages by date of publication, The comprehensive R archive network, 2025. Available from: https://cran.r-project.org/web/packages/available_packages_by_date.html.

[45] Posit software, RPubs, RStudio, 2025. Available from: https://rpubs.com/.

[46] M. Deitke, Scaling embodied artificial intelligence: massive 3d simulations to train agents, Bachelors Thesis, University of Washington, 2023.

[47] EPA, Ap-42: compilation of air pollutant emission factors from stationary sources, third edition, U. S. Environmental Protection Agency, 2024. Available from: https://www.epa.gov/air-emissions-factors-and-quantification/ap-42-compilation-air-emissions-factors-stationary-sources.

[48] ISEF, Regeneron international science and engineering fair, Society for Science, 2023. Available from: https://www.societyforscience.org/annual-reports/2023-annual-report/regeneron-isef/.