{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,17]],"date-time":"2026-02-17T12:23:54Z","timestamp":1771331034594,"version":"3.50.1"},"reference-count":40,"publisher":"MDPI AG","issue":"1","license":[{"start":{"date-parts":[[2021,12,23]],"date-time":"2021-12-23T00:00:00Z","timestamp":1640217600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Future Internet"],"abstract":"<jats:p>Deep learning provides new ways for defect detection in automatic optical inspections (AOI). However, the existing deep learning methods require thousands of images of defects to be used for training the algorithms. It limits the usability of these approaches in manufacturing, due to lack of images of defects before the actual manufacturing starts. In contrast, we propose to train a defect detection unsupervised deep learning model, using a much smaller number of images without defects. We propose an unsupervised deep learning model, based on transfer learning, that extracts typical semantic patterns from defect-free samples (one-class training). The model is built upon a pre-trained VGG16 model. It is further trained on custom datasets with different sizes of possible defects (printed circuit boards and soldered joints) using only small number of normal samples. We have found that the defect detection can be performed very well on a smooth background; however, in cases where the defect manifests as a change of texture, the detection can be less accurate. The proposed study uses deep learning self-supervised approach to identify if the sample under analysis contains any deviations (with types not defined in advance) from normal design. The method would improve the robustness of the AOI process to detect defects.<\/jats:p>","DOI":"10.3390\/fi14010008","type":"journal-article","created":{"date-parts":[[2021,12,23]],"date-time":"2021-12-23T10:16:18Z","timestamp":1640254578000},"page":"8","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":11,"title":["The Impact of a Number of Samples on Unsupervised Feature Extraction, Based on Deep Learning for Detection Defects in Printed Circuit Boards"],"prefix":"10.3390","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-9670-4034","authenticated-orcid":false,"given":"Ihar","family":"Volkau","sequence":"first","affiliation":[{"name":"School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore 639798, Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Abdul","family":"Mujeeb","sequence":"additional","affiliation":[{"name":"School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore 639798, Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Wenting","family":"Dai","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Nanyang Technological University, Singapore 639798, Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Marius","family":"Erdt","sequence":"additional","affiliation":[{"name":"Fraunhofer Research Center, Nanyang Technological University, Singapore 639798, Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Alexei","family":"Sourin","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Nanyang Technological University, Singapore 639798, Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2021,12,23]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"287","DOI":"10.1006\/cviu.1996.0020","article-title":"Automatic PCB Inspection Algorithms: A Survey","volume":"63","author":"Moganti","year":"1996","journal-title":"Comput. Vis. Image Underst."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Anitha, D.B., and Rao, M. (2017, January 22\u201324). A survey on defect detection in bare PCB and assembled PCB using image processing techniques. Proceedings of the 2017 International Conference on Wireless Communications, Signal Processing and Networking (WiSPNET), Chennai, India.","DOI":"10.1109\/WiSPNET.2017.8299715"},{"key":"ref_3","first-page":"1095","article-title":"Automatic optical inspection for PCB manufacturing: A survey","volume":"5","author":"Taha","year":"2014","journal-title":"Int. J. Sci. Eng. Res."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"391","DOI":"10.1177\/001872088302500404","article-title":"Human and Machine Performance in an Inspection Task","volume":"25","author":"Drury","year":"1983","journal-title":"Hum. Factors J. Hum. Factors Ergon. Soc."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.compind.2014.10.006","article-title":"Automated visual inspection in the semiconductor industry: A survey","volume":"66","author":"Huang","year":"2015","journal-title":"Comput. Ind."},{"key":"ref_6","first-page":"1415","article-title":"Improved bare PCB defect detection approach based on deep feature learning","volume":"2018","author":"Zhang","year":"2018","journal-title":"J. Eng."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"103","DOI":"10.1016\/0166-3615(95)00063-1","article-title":"Automated inspection of printed circuit boards through machine vision","volume":"28","author":"Wu","year":"1996","journal-title":"Comput. Ind."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Eun, H.Y., Seung, H.P., Cheong-Sool, P., and Jun-Geol, B. (2018). Feature-learning-based printed circuit board inspection via speeded-up robust features and random forest. Appl. Sci., 8.","DOI":"10.3390\/app8060932"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"415","DOI":"10.1007\/s10044-012-0305-7","article-title":"Defect detection using feature point matching for non-repetitive patterned images","volume":"17","author":"Kim","year":"2012","journal-title":"Pattern Anal. Appl."},{"key":"ref_10","first-page":"141","article-title":"Define, Measure, Analyze, Improve, and Control (DMAIC) Process","volume":"Volume 1","year":"2015","journal-title":"The SAGE Encyclopedia of Quality and the Service Economy"},{"key":"ref_11","first-page":"13702","article-title":"Metal surface inspection for defect detection and classification using Gabor Filter","volume":"3","author":"Tikhe","year":"2014","journal-title":"Int. J. Innov. Res. Sci. Eng. Technol."},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Timm, F., and Barth, E. (2011, January 23\u201327). Non-parametric texture defect detection using Weibull features. Proceedings of the IS&T\/SPIE Electronic Imaging, San Francisco, CA, USA.","DOI":"10.1117\/12.872463"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"600","DOI":"10.1109\/TIP.2003.819861","article-title":"Image quality assessment: From error visibility to structural similarity","volume":"13","author":"Wang","year":"2004","journal-title":"IEEE Trans. Image Process."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Carro, R.C., Larios, J.-M.A., Huerta, E.B., Caporal, R.M., and Cruz, F.R. (2015, January 20\u201323). Face Recognition Using SURF. Proceedings of the Intelligent Computing Theories and Methodologies, Fuzhou, China.","DOI":"10.1007\/978-3-319-22180-9_31"},{"key":"ref_15","first-page":"161","article-title":"Steel surface defects detection and classification using SIFT and voting strategy","volume":"6","author":"Suvdaa","year":"2012","journal-title":"Int. J. Softw. Eng. Appl."},{"key":"ref_16","first-page":"1","article-title":"An algorithm to group defects on printed circuit board for automated visual inspection","volume":"9","author":"Khalid","year":"2008","journal-title":"Int. J. Simul. Syst. Sci. Technol."},{"key":"ref_17","first-page":"87","article-title":"Neural network paradigm for classification of defects","volume":"39","author":"Heriansyah","year":"2003","journal-title":"J. Teknol."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Wu, X., Cao, K., and Gu, X. (2017, January 13\u201314). A Surface Defect Detection Based on Convolutional Neural Network. Proceedings of the International Conference on Computer Vision Systems, Lecture Notes in Computer Science, Shanghai, China.","DOI":"10.1007\/978-3-319-68345-4_17"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Masci, J., Meier, U., Ciresan, D., Schmidhuber, J., and Fricout, G. (2012, January 10\u201315). Steel defect classification with Max-Pooling Convolutional Neural Networks. Proceedings of the 2012 International Joint Conference on Neural Networks (IJCNN), Brisbane, QLD, Australia.","DOI":"10.1109\/IJCNN.2012.6252468"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"3465","DOI":"10.1007\/s00170-017-0882-0","article-title":"A fast and robust convolutional neural network-based defect detection model in product quality control","volume":"94","author":"Wang","year":"2018","journal-title":"Int. J. Adv. Manuf. Technol."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"66","DOI":"10.1016\/j.engappai.2019.01.008","article-title":"Railway track fastener defect detection based on image processing and deep learning techniques: A comparative study","volume":"80","author":"Wei","year":"2019","journal-title":"Eng. Appl. Artif. Intell."},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Chaudhary, V., Dave, I.R., and Upla, K.P. (2017, January 22\u201324). Automatic visual inspection of printed circuit board for defect detection and classification. Proceedings of the 2017 International Conference on Wireless Communications, Signal Processing and Networking (WiSPNET), Chennai, India.","DOI":"10.1109\/WiSPNET.2017.8299858"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"255","DOI":"10.1515\/teme-2014-1006","article-title":"Automated fault detection using deep belief networks for the quality inspection of electromotors","volume":"81","author":"Sun","year":"2014","journal-title":"Tech. Mess."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"341","DOI":"10.1016\/j.procs.2017.09.040","article-title":"The Application of One-Class Classifier Based on CNN in Image Defect Detection","volume":"114","author":"Zhang","year":"2017","journal-title":"Procedia Comput. Sci."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Mei, S., Wang, Y., and Wen, G. (2018). Automatic Fabric Defect Detection with a Multi-Scale Convolutional Denoising Autoencoder Network Model. Sensors, 18.","DOI":"10.3390\/s18041064"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"247","DOI":"10.1177\/0040517519862880","article-title":"Unsupervised fabric defect detection based on a deep convolutional generative adversarial network","volume":"90","author":"Hu","year":"2020","journal-title":"Text. Res. J."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Masci, J., Meier, U., Cire\u015fan, D., and Schmidhuber, J. (2011, January 14\u201317). Stacked convolutional auto-encoders for hierarchical feature extraction. Proceedings of the ICANN\u201911: Proceedings of the 21th International Conference on Artificial Neural Networks, Espoo, Finland.","DOI":"10.1007\/978-3-642-21735-7_7"},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Mujeeb, A., Dai, W., Erdt, M., and Sourin, A. (2018, January 3\u20135). Unsupervised Surface Defect Detection Using Deep Autoencoders and Data Augmentation. Proceedings of the 2018 International Conference on Cyberworlds (CW), Singapore.","DOI":"10.1109\/CW.2018.00076"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"119","DOI":"10.1007\/s11633-017-1053-3","article-title":"A survey on deep learning-based fine-grained object classification and semantic segmentation","volume":"14","author":"Zhao","year":"2017","journal-title":"Int. J. Autom. Comput."},{"key":"ref_30","unstructured":"Koch, G., Zemel, R., and Salakhutd, R. (2015, January 6\u201311). Siamese Neural Networks for One-Shot Image Recognition. Proceedings of the ICML Deep Learning Workshop, Lille, France."},{"key":"ref_31","unstructured":"Vinyals, O., Blundell, C., Lillicrap, T., Kavukcuoglu, K., and Wierstr, D. (2016, January 9\u201312). Matching Networks for One Shot Learning. Proceedings of the NIPS\u201430th International Conference on Neural Information Processing Systems, Barcelona, Spain."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Lowe, D. (1999, January 20\u201325). Object Recognition from Local Scale-Invariant Features. Proceedings of the Seventh IEEE International Conference on Computer Vision, Corfu, Greece.","DOI":"10.1109\/ICCV.1999.790410"},{"key":"ref_33","unstructured":"Simonyan, K., and Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"He, K.M., Zhang, X.Y., Ren, S.Q., and Sun, J. (2016, January 27\u201330). Deep Residual Learning for Image Recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA.","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref_35","unstructured":"Gidaris, S., Singh, P., and Komodakis, N. (May, January 30). Unsupervised representation learning by predicting image rotations. Proceedings of the ICLR 2018, Vancouver, BC, Canada."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Volkau, I., Mujeeb, A., Wenting, D., Marius, E., and Alexei, S. (2019, January 2\u20134). Detection Defect in Printed Circuit Boards using Unsupervised Feature Extraction Upon Transfer Learning. Proceedings of the 2019 International Conference on Cyberworlds (CW), Kyoto, Japan.","DOI":"10.1109\/CW.2019.00025"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"442","DOI":"10.1016\/0005-2795(75)90109-9","article-title":"Comparison of the predicted and observed secondary structure of T4 phage lysozyme","volume":"405","author":"Matthews","year":"1975","journal-title":"Biochim. Biophys. Acta BBA Protein Struct."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"353","DOI":"10.1207\/s15327906mbr1003_8","article-title":"A Monte Carlo Study of the Stability of Canonical Correlations, Canonical Weights and Canonical Variate-Variable Correlations","volume":"10","author":"Barcikowski","year":"1975","journal-title":"Multivar. Behav. Res."},{"key":"ref_39","unstructured":"Geirhos, R., Rubisch, P., Michaelis, C., Bethge, M., Wichmann, F.A., and Brendel, W. (2018). ImageNet-Trained CNNs Are Biased towards Texture; Increasing Shape Bias Improves Accuracy and Robustness. arXiv."},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Baierle, I.C., Benitez, G.B., Nara, E.O.B., Schaefer, J.L., and Sellitto, M.A. (2020). Influence of Open Innovation Variables on the Competitive Edge of Small and Medium Enterprises. J. Open Innov. Technol. Mark. Complex., 6.","DOI":"10.3390\/joitmc6040179"}],"container-title":["Future Internet"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1999-5903\/14\/1\/8\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T07:52:23Z","timestamp":1760169143000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1999-5903\/14\/1\/8"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,12,23]]},"references-count":40,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2022,1]]}},"alternative-id":["fi14010008"],"URL":"https:\/\/doi.org\/10.3390\/fi14010008","relation":{},"ISSN":["1999-5903"],"issn-type":[{"value":"1999-5903","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,12,23]]}}}