Abstract
This paper describes an approach for automatic classification of multi-class printing defects based on semantic segmentation models. Classification of current printing defects strongly depends on visual inspection of skilled workers. Therefore, we developed an application that captures the expert’s perception and knowledge directly into the teaching image data, and classify the data automatically using semantic segmentation. We compared U-Net, SegNet, and PSPNet by benchmarking to find the best model for our situation where the number of input images for every defect type is set in the range of 10–120 by applying data augmentation. As the result, we found SegNet is the best model for our tiny and faint images. Finally, we added another grayscale channel to the input layer of SegNet to improve sensitivity to obscurity and show the effect.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Ministry of Economy, Trade and Industry of Japan.: White Paper on Manufacturing Industries (Monodzukuri) (2019). (English version) https://www.meti.go.jp/english/press/2019/0611_001.html 2019/1/22
Chugoku Industrial Innovation Center.: An investigation into possibility to promoting the automation of inspection process in the manufacturing company (2016)
Gollisch, T., Meister, M.: Eye smarter than scientists believed: neural computations in circuits of the retina. Neuron 65(2), 150–164 (2010)
Teppei, T.: POODL–Image recognition cloud plat form for printing factory, https://www.slideshare.net/TeppeiTamaki/poodl-a-image-recognition-cloud-platform-for-every-printing-factory. Accessed 22 Jan 2019
Shinichi, H., Takeshi, U., Toshinori, M., Nobuyuki, I.: Image recognition AI to promote the automation of visual inspections. Fujitsu 69(4), 42–48 (2018)
Kingma, D.P., Welling, M.: Auto-encoding variational bayes. arXiv:1312.6114 (2013)
Llorca, D.F., Arroyo, R., Sotelo, M.A.: Vehicle logo recognition in traffic images using HOG features and SVM. In: 16th International IEEE Conference on Intelligent Transportation Systems (ITSC 2013), pp. 2229–2234. IEEE, Hague (2013)
Krizhevsky, A., Sutskever, I., & Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: Advances in neural information processing systems, pp. 1097–1105. Neural Information Processing Systems Conference (NIPS), Nevada (2012)
Cha, Y.J., Choi, W., Büyüköztürk, O.: Deep learning-based crack damage detection using convolutional neural networks. Comput. Aided Civ. Infrastruct. Eng. 32(5), 361–378 (2017)
Imoto, K., Nakai, T., Ike, T., Haruki, K., Sato, Y.: A CNN-based transfer learning method for defect classification in semiconductor manufacturing. IEEE Trans. Semicond. Manuf. 32(4), 455–459 (2019)
Girshick, R., Donahue, J., Darrell, T., & Malik, J.: Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 580–587. IEEE, Ohio (2014)
Long, J., Shelhamer, E., Darrell, T.: Fully convolutional networks for semantic segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition pp. 3431–3440. IEEE, Massachusetts (2015)
Olaf, R., Philipp, F., Thomas, B.: U-Net: convolutional networks for biomedical image segmentation. In: Medical Image Computing and Computer-Assisted Intervention (MICCAI), LNCS, vol. 9351, pp. 234–241. Springer, Munich (2015)
Badrinarayanan, V., Kendall, A., Cipolla, R.: SegNet: a deep convolutional encoder-decoder architecture for image segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 39(12), 2481–2495 (2017)
Zhao, H., Shi, J., Qi, X., Wang, X., & Jia, J.: Pyramid scene parsing network. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 2881–2890. IEEE, Honolulu (2017)
Image Polygonal Annotation with Python. https://github.com/wkentaro/labelme. Accessed 01 Jan 2019
Lin, T.Y., Goyal, P., Girshick, R., He, K., Dollár, P.: Focal loss for dense object detection. In: Proceedings of the IEEE international conference on computer vision, pp. 2980–2988. IEEE, Honolulu (2017)
Dice, L.R.: Measures of the amount of ecologic association between species. Ecology 26(3), 297–302 (1945)
Acknowledgments
This research is supported by Cross-ministerial Strategic Innovation Promotion Program (SIP), “Big-data and AI-enabled Cyberspace Technologies” (Funding Agency: NEDO). We appreciate the support. The authors also appreciate all reviewers’ constructive comments.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Tsuji, T., Arima, S. (2021). Automatic Multi-class Classification of Tiny and Faint Printing Defects Based on Semantic Segmentation. In: Zimmermann, A., Howlett, R., Jain, L. (eds) Human Centred Intelligent Systems. Smart Innovation, Systems and Technologies, vol 189. Springer, Singapore. https://doi.org/10.1007/978-981-15-5784-2_9
Download citation
DOI: https://doi.org/10.1007/978-981-15-5784-2_9
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-15-5783-5
Online ISBN: 978-981-15-5784-2
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)