Automated Identification of Wood Veneer Surface Defects Using Faster Region-Based Convolutional Neural Network with Data Augmentation and Transfer Learning
Source abstract
Abstract
In the lumber and wood processing industry, most visual quality inspections are still done by trained human operators. Visual inspection is a tedious and repetitive task that involves a high likelihood of human error. Currently, new automated solutions with high-resolution cameras and visual inspection algorithms are being tested, but they are not always fast and accurate enough for real-time industrial applications. This paper proposes an automatic visual inspection system for the location and classification of defects on the wood surface. We adopted a faster region-based convolutional neural network (faster R-CNN) for the identification of defects on wood veneer surfaces. Faster R-CNN has been successfully used in medical image processing and object tracking before, but it has not yet been applied for wood panel surface quality assurance. To improve the results, we used pre-trained AlexNet, VGG16, BNInception, and ResNet152 neural network models for transfer learning. The results of the experiments using a synthetically augmented dataset are presented. The best average accuracy of 80.6% was obtained using the pretrained ResNet152 neural network model. By combining all the defect classes, a 96.1% accuracy of finding wood panel surface defects was achieved.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Industrial Vision Systems and Defect Detection · Surface Roughness and Optical Measurements · Remote Sensing and LiDAR Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
UAB “Vakaru Medienos Grupe”, LT-92100 Klaipėda, Lithuania
UAB “Power of Eye”, LT-51392 Kaunas, Lithuania
Institute of Mathematics, Silesian University of Technology, 44-100 Gliwice, Poland
References (66)
Show source-provided references
1. (2019, January 15). Forest Products Statistics. Available online: http://www.fao.org/forestry/statistics/80938/en/.
2. Gu, I.Y.H., and Andersson, H. (2008). Automatic Classification of Wood Defects Using Support Vector Machines. Computer Vision and Graphics, ICCVG 2008, Springer.10.1007/978-3-642-02345-3_35 ↗
3. Automated vision inspection of timber surface defect: A review
4. Soulié, F.F., and Gallinari, P. (1998). Wood Surface Inspection System Based on Generic Visual Features. Industrial Applications of Neural Networks, World Scientific.10.1142/3510 ↗
5. Cetiner, I., Ali Var, A., and Cetiner, H. (2016). Classification of Knot Defect Types Using Wavelets and KNN. Electron. Electr. Eng., 22.10.5755/j01.eie.22.6.17227 ↗
6. Wang, Y., Wang, H., and Mo, L. (2010, January 13–15). Research on recognition of wood texture based on integrated neural network classifier. Proceedings of the International Conference on Intelligent Control and Information Processing, ICICIP 2010, Part 2, Dalian, China.10.1109/ICICIP.2010.5564210 ↗
7. Study on wood board defect detection based on artificial neural network10.2174/1874444301507010290 ↗
8. An artificial neural network for real-time hardwood lumber grading10.1016/j.compag.2016.11.018 ↗
9. Deep learning for use in lumber classification tasks10.1007/s00226-019-01086-z ↗
10. Loke, K.S. (2018, January 28–30). Texture recognition using a novel input layer for deep convolutional neural network. Proceedings of the IEEE 3rd International Conference on Communication and Information Systems, ICCIS, Singapore.10.1109/ICOMIS.2018.8645055 ↗
11. Karayiannis, Y.A., Stojanovic, R., Mitropoulos, P., Koulamas, C., Stouraitis, T., Koubias, S., and Papadopoulos, G. (1999, January 5–8). Defect Detection and Classification on Web Textile Fabric Using Multiresolution Decomposition and Neural Networks. Proceedings of the 6th IEEE International Conference Electronics, Circuits Systems, Pafos, Cyprus.
12. Carew, T., Ghita, O., and Whelan, P.F. (2001, January 17–21). A Vision System for Detecting Paint Faults on Painted Slates. Proceedings of the ICASE International Conference on Control, Automation and Systems, Jeju Island, Korea.
13. Online Fabric Defect Inspection Using Smart Visual Sensors10.3390/s130404659 ↗
14. Defect detection in textured materials using Gabor filters10.1109/28.993164 ↗
15. Slub Extraction in Woven Fabric Images Using Gabor Filters10.1177/0040517507090495 ↗
16. Chacon, M.I., and Alonso, G.R. (2006). Wood Defects Classification Using a SOM/FFP Approach with Minimum Dimension Feature Vector. Advances in Neural Networks, Springer.10.1007/11760191_161 ↗
17. Wood defect classification based on image analysis and support vector machines10.1007/s00226-009-0287-9 ↗
18. Mahram, A., Shayesteh, M.G., and Jafarpour, S. (2012, January 3–4). Classification of wood surface defects with hybrid usage of statistical and textural features. Proceedings of the 35th International Conference on Telecommunications and Signal Processing (TSP), Prague, Czech Republic.10.1109/TSP.2012.6256397 ↗
19. Study on the identification of the wood surface defects based on texture features10.1016/j.ijleo.2015.05.101 ↗
20. Hittawe, M.M., Muddamsetty, S.M., Sidibé, D., and Mériaudeau, F. (2015, January 27–30). Multiple features extraction for timber defects detection and classification using SVM. Proceedings of the IEEE International Conference on Image Processing (ICIP), Quebec, QC, Canada.10.1109/ICIP.2015.7350834 ↗
21. Zhao, P., and Wang, C.-K. (2019). Hardwood Species Classification with Hyperspectral Microscopic Images. J. Spectrosc., 2019.10.1155/2019/2039453 ↗
22. What Makes for Effective Detection Proposals?10.1109/TPAMI.2015.2465908 ↗
23. Girshick, R., Donahue, J., Darrell, T., and Malik, J. (2014, January 24–27). Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA.10.1109/CVPR.2014.81 ↗
24. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks10.1109/TPAMI.2016.2577031 ↗
25. Nasrullah, N., Sang, J., Alam, M.S., Mateen, M., Cai, B., and Hu, H. (2019). Automated lung nodule detection and classification using deep learning combined with multiple strategies. Sensors, 19.10.3390/s19173722 ↗
26. Dominguez-Sanchez, A., Cazorla, M., and Orts-Escolano, S. (2018). A new dataset and performance evaluation of a region-based CNN for urban object detection. Electronics, 7.10.3390/electronics7110301 ↗
27. Tao, X., Zhang, D., Ma, W., Liu, X., and Xu, D. (2018). Automatic Metallic Surface Defect Detection and Recognition with Convolutional Neural Networks. Appl. Sci., 8.10.3390/app8091575 ↗
28. Neural network design and feature selection using principal component analysis and taguchi method for identifying wood veneer defects
29. A generic deep-learning-based approach for automated surface inspection10.1109/TCYB.2017.2668395 ↗
30. Donahue, J., Jia, Y., Vinyals, O., Hoffman, J., Zhang, N., Tzeng, E., and Darrell, T. (2014, January 21–26). DeCAF: A deep convolutional activation feature for generic visual recognition. Proceedings of the 31st International Conference on International Conference on Machine Learning, ICML’14, Beijing, China.
31. Rudakov, N., Eerola, T., Lensu, L., Kälviäinen, H., and Haario, H. (2019). Detection of mechanical damages in sawn timber using convolutional neural networks. Ger. Conf. Pattern Recognit., 115–126.10.1007/978-3-030-12939-2_9 ↗
32. Huang, G., Liu, Z., van der Maaten, L., and Weinberger, K.Q. (2017, January 21–26). Densely Connected Convolutional Networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA.10.1109/CVPR.2017.243 ↗
33. Minaee, S., and Abdolrashidi, A. (2019). Deep-emotion: Facial expression recognition using attentional convolutional network. arXiv.
34. MTBI Identification From Diffusion MR Images Using Bag of Adversarial Visual Features10.1109/TMI.2019.2905917 ↗
35. Redmon, J., Divvala, S., Girshick, R., and Farhadi, A. (July, January 26). You Only Look Once: Unified, Real-Time Object Detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.
36. Overlaid Arrow Detection for Labeling Regions of Interest in Biomedical Images10.1109/MIS.2016.24 ↗
37. Santosh, K.C., Wendling, L., Antani, S.K., and Thoma, G.R. (2014, January 24–28). Scalable Arrow Detection in Biomedical Images. Proceedings of the 22nd International Conference on Pattern Recognition, Stockholm, Sweden.10.1109/ICPR.2014.561 ↗
38. A Simple and Efficient Arrowhead Detection Technique in Biomedical Images10.1142/S0218001416570020 ↗
39. Fuzzy partition of two-dimensional histogram and its application to thresholding10.1016/S0031-3203(98)00080-6 ↗
40. Lin, T., Dollár, P., Girshick, R., He, K., Hariharan, B., and Belongie, S. (2017, January 21–26). Feature Pyramid Networks for Object Detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA.10.1109/CVPR.2017.106 ↗
41. SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation10.1109/TPAMI.2016.2644615 ↗
42. Minaee, S., and Wang, Y. (2016, January 25–28). Screen content image segmentation using sparse decomposition and total variation minimization. Proceedings of the IEEE International Conference on Image Processing (ICIP), Phoenix, AZ, USA.10.1109/ICIP.2016.7533087 ↗
43. Qi, G.-J., and Luo, J. (2019). Small Data Challenges in Big Data Era: A Survey of Recent Progress on Unsupervised and Semi-Supervised Method. arXiv.
44. Wang, J., and Perez, L. (2017). Convolutional Neural Networks Visual Recognition. arXiv.
45. Taylor, L., and Nitschke, G. (2017). Improving Deep Learning using Generic Data Augmentation. arXiv.10.1109/SSCI.2018.8628742 ↗
46. Boukli Hacene, G., Gripon, V., Farrugia, N., Arzel, M., and Jezequel, M. (2018). Transfer Incremental Learning Using Data Augmentation. Appl. Sci., 8.10.3390/app8122512 ↗
47. Girshick, R. (2015). Fast R-CNN. arXiv.10.1109/ICCV.2015.169 ↗
48. A Survey on Transfer Learning10.1109/TKDE.2009.191 ↗
49. ImageNet classification with deep convolutional neural networks10.1145/3065386 ↗
50. Simonyan, K., and Zisserman, A. (2015, January 7–9). Very deep convolutional networks for large-scale image recognition. Proceedings of the 3rd International Conference on Learning Representations (ICLR 2015), San Diego, CA, USA.
51. He, K., Zhang, X., Ren, S., and Sun, J. (2016, January 27–30). Deep Residual Learning for Image Recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.10.1109/CVPR.2016.90 ↗
52. Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., and Rabinovich, A. (2015, January 7–12). Going Deeper with Convolutions. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA.10.1109/CVPR.2015.7298594 ↗
53. Lin, M., and Chen, Q. (2014, January 14–16). Network in Network. Proceedings of the International Conference on Learning Representations, Banff, AB, Canada.
54. Ioffe, S., and Szegedy, C. (2015, January 6–11). Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. Proceedings of the 32nd International Conference on International Conference on Machine Learning (ICML’15), Lille, France.
55. Qayyum, R., Kamal, K., Zafar, T., and Mathavan, S. (2016, January 7–8). Wood defects classification using GLCM based features and PSO trained neural network. Proceedings of the 22nd International Conference on Automation and Computing (ICAC), Colchester, UK.10.1109/IConAC.2016.7604931 ↗
56. Packianather, M.S., Drake, P.R., and Pham, D.T. (October, January 28). Feature selection method for neural network for the classification of wood veneer defects. Proceedings of the World Automation Congress, Hawaii, HI, USA.
57. Zhao, D. (2009, January 11–12). Automated Recognition of Wood Damages Using Artificial Neural Network. Proceedings of the International Conference on Measuring Technology and Mechatronics Automation, Zhangjiajie, China.10.1109/ICMTMA.2009.40 ↗
58. Qi, D., Zhang, P., Jin, X., and Zhang, X. (2010, January 10–12). Applying Hopfield neural network to defect edge detection of wood image. Proceedings of the 6th International Conference on Natural Computation, Yantai, China.10.1109/ICNC.2010.5582850 ↗
59. Chen, H., Zhao, H., Han, D., Liu, W., Chen, P., and Liu, K. (2019). Structure-Aware-based Crack Defect Detection for Multicrystalline Solar Cells. Measurement.10.1016/j.measurement.2019.107170 ↗
60. Jia, L., Chen, C., Xu, S., and Shen, J. (2019). Fabric defect inspection based on lattice segmentation and template statistics. Inf. Sci.10.1016/j.ins.2019.10.032 ↗
61. Wang, C., Liu, Y., and Wang, P. (2019). Extraction and Detection of Surface Defects in Particleboards by Tracking Moving Targets. Algorithms, 12.10.3390/a12010006 ↗
62. Azizah, L.M., Umayah, S.F., Riyadi, S., Damarjati, C., and Utama, N.A. (2017, January 24–26). Deep learning implementation using convolutional neural network in mangosteen surface defect detection. Proceedings of the 7th IEEE International Conference on Control System, Computing and Engineering (ICCSCE), Penang, Malaysia.10.1109/ICCSCE.2017.8284412 ↗
63. Frohlich, H.B., Fantin, A.V., de Oliveira, B.C.F., Willemann, D.P., Iervolino, L.A., Benedet, M.E., and Goncalves, A.A. (2018, January 8–13). Defect classification in shearography images using convolutional neural networks. Proceedings of the International Joint Conference on Neural Networks (IJCNN), Rio de Janeiro, Brazil.10.1109/IJCNN.2018.8489133 ↗
64. Konrad, T., Lohmann, L., and Abel, D. (2019, January 1–4). Surface Defect Detection for Automated Inspection Systems using Convolutional Neural Networks. Proceedings of the 27th Mediterranean Conference on Control and Automation (MED), Akko, Israel.10.1109/MED.2019.8798497 ↗
65. A Fine-Tuned Convolution Neural Network Based Approach For Phenotype Classification Of Zebrafish Embryo10.1016/j.procs.2018.08.051 ↗
66. Deep convolution neural network for accurate diagnosis of glaucoma using digital fundus images10.1016/j.ins.2018.01.051 ↗