DEVELOPMENT OF A 3D AND LINE-SCANNING CAMERA FOR AUTOMATIC WOOD DEFECT MARKING
DOI:
https://doi.org/10.68302/std2026.vol2.13Keywords:
Computer Vision, Deep Machine Learning, defect detection, FPGA, HPC, labelling, quality control, wood processingAbstract
The industrial sector is focused on efficiency of an industrial vision through automation as one of the most popular solutions as in recent decades. The main bottleneck of the process is image processing due to high speed and precision. Graphic processing units (GPUs) and Field-Programmable Gate Array (FPGAs) have been proposed for real-time detection, where tasks are now performed repeatedly by machines assisting or replacing humans, reducing the occurrence of errors. New frameworks are trying to be in pace with Industry 5.0 advancements and to rely heavily on real-time sensor systems, with integration to Smart Factory which are essential for assessing dynamics of industrial data. In the paper, the research project focuses on developing an application of innovative combined 3D and line-scan camera system capable of simultaneously capturing visual and 3D information and identifying wood properties such as fiber direction and resin pockets. The camera employs laser triangulation and controlled LED illumination for precise wood structure analysis, while high-speed Field-Programmable Gate Array (FPGA) with High-performance computing (HPC) technologies enables real-time image analysis in-camera. Additionally, the research will focus on presenting fully functional prototype with discussion on modular system architecture scalability and parallel development of FPGA and AI components in simulation, moving closer to industrial implementation in subsequent stages. As the critical step for framework elaboration, the development of a functional application in manufacturing environment of FPGA firmware, laser triangulation, line-scan imaging, controlled illumination, trigger boards, and AI modules and evaluation of technological options that would ensure effective recognition of wood defects, as well as creation of a basis for further technological development are performed.
Supporting Agencies
Development of a 3D and Line-Scanning Camera for Automatic Wood Defect Marking. Research contract Nr: 2.2.1.3.i.0/1/24/A/CFLA/007 signed between Competence Centre of Mechanical Engineering and Central Finance and Contracting Agency. The research leading to these results has been supported by the European Regional Development Fund project "Competence Centre of Mechanical Engineering". Research No. D.1.5. MASOC KC support for digital product development”.Downloads
References
[1] E. Young, “Use of linescan cameras and a DSP processing system for high-speed wood inspection,” in Machine Vision Applications, Architectures, and Systems Integration IV, B. G. Batchelor, S. S. Solomon, and F. M. Waltz, Eds., Oct. 1995, pp. 259–264. doi: 10.1117/12.223987.
[2] M. Barjaktarovic, S. Petricevic, and J. Radunovic, “High performance coated board inspection system based on commercial components,” J. Instrum., vol. 2, no. 07, pp. T07001–T07001, Jul. 2007, doi: 10.1088/1748-0221/2/07/T07001.
[3] H. A. A. Mohammed and A. A. Gribanov, “Research and Development of Automatic Sawn Timber Defect Removal Line,” in 2022 International Conference on Industrial Engineering, Applications and Manufacturing (ICIEAM), IEEE, May 2022, pp. 515–519. doi: 10.1109/ICIEAM54945.2022.9787211.
[4] M. Ericsson, D. Johansson, and D. Stjern, “AI-Based Quality Control of Wood Surfaces with Autonomous Material Handling,” Appl. Sci., vol. 11, no. 21, p. 9965, Oct. 2021, doi: 10.3390/app11219965.
[5] A. Urbonas, V. Raudonis, R. Maskeliunas, and R. Damaševičius, “Automated identification of wood veneer surface defects using faster region-based convolutional neural network with data augmentation and transfer learning,” Appl. Sci., vol. 9, no. 22, 2019, doi: 10.3390/app9224898.
[6] L. P. Yi, M. F. Akbar, M. N. A. Wahab, B. A. Rosdi, M. A. Fauthan, and N. H. M. M. Shrifan, “The Prospect of Artificial Intelligence-Based Wood Surface Inspection: A Review,” IEEE Access, vol. 12, pp. 84706–84725, 2024, doi: 10.1109/ACCESS.2024.3412928.
[7] V. Nandini, R. Deepak Vishal, C. Arun Prakash, and S. Aishwarya, “A Review on Applications of Machine Vision Systems in Industries,” Indian J. Sci. Technol., vol. 9, no. 48, Dec. 2016, doi: 10.17485/ijst/2016/v9i48/108433.
[8] C. Fan, Z. Zhuang, Y. Liu, Y. Yang, H. Zhou, and X. Wang, “Bilateral Defect Cutting Strategy for Sawn Timber Based on Artificial Intelligence Defect Detection Model,” Sensors, vol. 24, no. 20, p. 6697, Oct. 2024, doi: 10.3390/s24206697.
[9] S. Landscheidt, M. Kans, and M. Winroth, “Opportunities for Robotic Automation in Wood Product Industries: The Supplier and System Integrators’ Perspective,” Procedia Manuf., vol. 11, pp. 233–240, 2017, doi: 10.1016/j.promfg.2017.07.231.
[10] R. Salim and J. Johansson, “Automation decisions in investment projects: A study in the Swedish wood products industry,” Procedia Manuf., vol. 25, pp. 255–262, 2018, doi: 10.1016/j.promfg.2018.06.081.
[11] C. Praschl and G. Zwettler, “Three-step Approach for Localization, Instance Segmentation and Multi-facet Classification of Individual Logs in Wooden Piles,” in Proceedings of the 11th International Conference on Pattern Recognition Applications and Methods, SCITEPRESS - Science and Technology Publications, 2022, pp. 683–688. doi: 10.5220/0010892100003122.
[12] A. Rinnhofer et al., “A multisensor system for texture-based high-speed hardwood lumber inspection,” E. R. Dougherty, J. T. Astola, and K. O. Egiazarian, Eds., Mar. 2005, p. 34. doi: 10.1117/12.588199.
[13] W. Polzleitner and G. Schwingshakl, “Real-time color-based texture analysis for sophisticated defect detection on wooden surfaces,” D. P. Casasent, E. L. Hall, and J. Roning, Eds., Oct. 2004, pp. 54–69. doi: 10.1117/12.580135.
[14] Y. Zhu, Z. Xu, Y. Lin, D. Chen, Z. Ai, and H. Zhang, “A Multi-Source Data Fusion Network for Wood Surface Broken Defect Segmentation,” Sensors, vol. 24, no. 5, p. 1635, Mar. 2024, doi: 10.3390/s24051635.
[15] Y. Zhu, Z. Xu, Y. Lin, D. Chen, K. Zheng, and Y. Yuan, “Surface defect detection of sawn timbers based on efficient multilevel feature integration,” Meas. Sci. Technol., vol. 35, no. 4, 2024, doi: 10.1088/1361-6501/ad15de.
[16] W.-H. Lim, M. B. Bonab, and K. H. Chua, “An Optimized Lightweight Model for Real-Time Wood Defects Detection based on YOLOv4-Tiny,” in 2022 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS), IEEE, Jun. 2022, pp. 186–191. doi: 10.1109/I2CACIS54679.2022.9815274.
[17] P. Kodytek, A. Bodzas, and P. Bilik, “A large-scale image dataset of wood surface defects for automated vision-based quality control processes,” F1000Research, vol. 10, p. 581, Jul. 2021, doi: 10.12688/f1000research.52903.1.
[18] Z. Yi, L. Luo, Q. Lu, M. Chen, W. Zhu, and Y. Zhang, “An efficient and accurate surface defect detection method for quality supervision of wood panels,” Meas. Sci. Technol., vol. 35, no. 5, p. 055209, May 2024, doi: 10.1088/1361-6501/ad26c9.
[19] S. Du, Z. Dong, Y. Li, and T. Ikenaga, “Straight-Line Detection Within 1 Millisecond Per Frame for Ultrahigh-Speed Industrial Automation,” IEEE Trans. Ind. Informatics, vol. 19, no. 4, pp. 5965–5975, Apr. 2023, doi: 10.1109/TII.2022.3170585.
[20] B. Jambreković, F. Veselčić, I. Ištok, T. Sinković, V. Živković, and T. Sedlar, “A Comparative Analysis of Oak Wood Defect Detection Using Two Deep Learning (DL)-Based Software,” Appl. Syst. Innov., vol. 7, no. 2, p. 30, Apr. 2024, doi: 10.3390/asi7020030.
[21] M. Asad, W. Azeem, H. Jiang, H. Tayyab Mustafa, J. Yang, and W. Liu, “2M3DF: Advancing 3D Industrial Defect Detection With Multi-Perspective Multimodal Fusion Network,” IEEE Trans. Circuits Syst. Video Technol., vol. 35, no. 7, pp. 6803–6815, Jul. 2025, doi: 10.1109/TCSVT.2025.3536475.
[22] Z. Liu, C. Peng, T. Work, J. N. Candau, A. Desrochers, and D. Kneeshaw, “Application of machine-learning methods in forest ecology: Recent progress and future challenges,” Environ. Rev., vol. 26, no. 4, pp. 339–350, 2018, doi: 10.1139/er-2018-0034.
[23] J. K. Park, B. K. Kwon, J. H. Park, and D. J. Kang, “Machine learning-based imaging system for surface defect inspection,” Int. J. Precis. Eng. Manuf. - Green Technol., vol. 3, no. 3, pp. 303–310, 2016, doi: 10.1007/s40684-016-0039-x.
[24] M. M. Taye, “Understanding of Machine Learning with Deep Learning: Architectures, Workflow, Applications and Future Directions,” Computers, vol. 12, no. 5, 2023, doi: 10.3390/computers12050091.
[25] C. Janiesch, P. Zschech, and K. Heinrich, “Machine learning and deep learning,” Electron. Mark., vol. 31, no. 3, pp. 685–695, Sep. 2021, doi: 10.1007/s12525-021-00475-2.
[26] B. Neyses and A. Scharf, “Using machine learning to predict the density profiles of surface-densified wood based on cross-sectional images,” Eur. J. Wood Wood Prod., vol. 80, no. 5, pp. 1121–1133, Oct. 2022, doi: 10.1007/s00107-022-01826-2.
[27] H. Ergun, “Wood identification based on macroscopic images using deep and transfer learning approaches,” PeerJ, vol. 12, p. e17021, Feb. 2024, doi: 10.7717/peerj.17021.
[28] M. Mohsin, O. S. Balogun, K. Haataja, and P. Toivanen, “Real-time defect detection and classification on wood surfaces using deep learning,” Electron. Imaging, vol. 34, no. 10, pp. 382-1-382–6, Jan. 2022, doi: 10.2352/EI.2022.34.10.IPAS-382.
[29] D. Ferreira, F. Moutinho, J. P. Matos-Carvalho, M. Guedes, and P. Deusdado, “Generic FPGA Pre-Processing Image Library for Industrial Vision Systems,” Sensors, vol. 24, no. 18, p. 6101, Sep. 2024, doi: 10.3390/s24186101.
[30] T. Chisholm, R. Lins, and S. Givigi, “FPGA-Based Design for Real-Time Crack Detection Based on Particle Filter,” IEEE Trans. Ind. Informatics, vol. 16, no. 9, pp. 5703–5711, 2020, doi: 10.1109/TII.2019.2950255.
[31] P. Cizek and J. Faigl, “Real-Time FPGA-Based Detection of Speeded-Up Robust Features Using Separable Convolution,” IEEE Trans. Ind. Informatics, vol. 14, no. 3, pp. 1155–1163, 2018, doi: 10.1109/TII.2017.2764485.
[32] A. Saday and I. A. Ozkan, “Robotic Welding Path Identification Using FPGA-Based Image Processing,” in 2022 14th International Conference on Computational Intelligence and Communication Networks (CICN), IEEE, Dec. 2022, pp. 115–118. doi: 10.1109/CICN56167.2022.10008273.
[33] J. De Vylder, S. Donné, D. Van Haerenborgh, and B. Goossens, “Real-time Machine Vision with GPU-acceleration using Quasar,” Electron. Imaging, vol. 28, no. 14, pp. 1–2, Feb. 2016, doi: 10.2352/ISSN.2470-1173.2016.14.IPMVA-375.
[34] D. Boscaini, F. Poiesi, S. Messelodi, A. Younes, and D. A. Grande, “Localisation of Defects in Volumetric Computed Tomography Scans of Valuable Wood Logs,” 2021, pp. 692–704. doi: 10.1007/978-3-030-68799-1_50.
[35] M. Pistellato et al., “Quantization-Aware NN Layers with High-throughput FPGA Implementation for Edge AI,” Sensors, vol. 23, no. 10, p. 4667, May 2023, doi: 10.3390/s23104667.
[36] N. Thompson, K. Greenewald, K. Lee, and G. F. Manso, “The Computational Limits of Deep Learning,” 2023, doi: 10.21428/bf6fb269.1f033948.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Vladimirs Šatrevičs, Gundars Kuļikovskis, Jēkabs Baumanis, Viktorija Skvarciany
This work is licensed under a Creative Commons Attribution 4.0 International License.
