PCB Defect Images Super-Resolution Reconstruction Based on Improved SRGAN
Abstract
:1. Introduction
2. Related Theories
2.1. Generating Adversarial Network
2.2. SRGAN
3. Algorithm Improvement Proposed Algorithm
3.1. Generator Improvements
3.2. Discriminator Improvements
3.3. Loss Function
4. Simulation Experiments and Results Analysis
4.1. Experimental Environment and Data Set
4.2. Evaluation Indicators
4.3. Experimental Results
5. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Liao, X.; Li, Q.; Qiu, Q.; Xu, H.; Liu, Z. Deep learning based PCB defect detection method. Reliab. Environ. Test. Electron. Prod. 2022, 40, 30–34. [Google Scholar]
- Wu, H.; Shi, Y.; Yang, W. Super-resolution adversarial network for circuit board defect detection. J. Harbin Univ. Commer. 2021, 37, 428–435. [Google Scholar]
- Chen, X.; Su, T.; Qi, W. Improved Faster RCNN-based defect detection algorithm for printed circuit boards. J. Jianghan Univ. 2022, 50, 87–96. [Google Scholar]
- Ruijuan, M. Research on Image Super-Resolution Reconstruction Based on Low-Rank Sparse Decomposition and Dictionary Learning Jinan; Shandong Normal University: Jinan, China, 2017. [Google Scholar]
- Yanqiu, T.; Hong, P.; Yaping, Z. A review of image super-resolution reconstruction research. J. Electron. 2020, 48, 1407–1420. [Google Scholar]
- Anwar, S.; Khan, S.; Barnes, N. A Deep Journey into Super-resolution: A Survey. ACM Comput. Surv. 2020, 53, 31–34. [Google Scholar] [CrossRef]
- Zhang, L.; Wu, X. An edge-guided image interpolation algorithm via directional filtering and data fusion. IEEE Trans. Image Process. 2006, 15, 2226–2238. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Jing, S.; Qiangqiang, Y.; Jiwei, L. Super-resolution reconstruction of license plate images with joint luminance-gradient constraints. Chin. J. Image Graph. 2018, 23, 802–813. [Google Scholar]
- Dong, C.; Loy, C.C.; He, K.M. Image super-resolution using deep convolutional network. IEEE Trans. Pattern Anal. Mach. Intell. 2016, 38, 295–307. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Dong, C.; Loy, C.C.; Tang, X.O. Accelerating the super resolution convolutional neural network. In Proceedings of the 14th European Conference on Computer Vision, Amsterdam, The Netherlands, 11–14 October 2016; Springer: Berlin/Heidelberg, Germany, 2016; pp. 391–407. [Google Scholar]
- Tai, Y.; Yang, J.; Liu, X. Image super-resolution via deep recursive residual network. In Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA, 21–26 July 2017; IEEE: Piscataway, NJ, USA, 2017; pp. 2790–2798. [Google Scholar]
- Lu, Z.; Li, J.; Liu, H.; Huang, C.; Zhang, L.; Zeng, T. Transformer for Single Image Super-Resolution. In Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, New Orleans, LA, USA, 19–20 June 2022; IEEE: Piscataway, NJ, USA, 2022. [Google Scholar]
- Ariav, I.; Cohen, I. Fully Cross-Attention Transformer for Guided Depth Super-Resolution. Sensors 2023, 23, 2723. [Google Scholar] [CrossRef] [PubMed]
- Siwei, H. Research on Super-Resolution Image Reconstruction Algorithm Based on Deep Learning; Taiyuan University of Technology: Taiyuan, China, 2018. [Google Scholar]
- Goodfellow, I.J.; Pouget-Abadie, J.; Mirza, M. Generative adversarial nets. In Proceedings of the 27th International Conference on Neural Information Processing Systems, Montreal, QC, Canada, 8–13 December 2014; MIT: Cambridge, MA, USA, 2014; pp. 2672–2680. [Google Scholar]
- Ledig, C.; Theis, L.; Huszár, F. Photo-realistic single image super-resolution using a generative adversarial net work. In Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA, 21–26 July 2017; IEEE: Piscataway, NJ, USA, 2017; pp. 105–114. [Google Scholar]
- Vaswani, A.; Shazeer, N.; Parmar, N. Attention is all you need. Adv. Neural Inf. Process. Syst. 2017, 30, 64–67. [Google Scholar]
- Ding, R.W.; Dall, H.; Li, G.P. TDD-Net: A tiny defect detection net work for printed circuit boards. CAAI Trans. Intell. Technol. 2019, 4, 110–116. [Google Scholar] [CrossRef]
Programming Environment | Server Hardware Configuration | Server Software Environment |
---|---|---|
Versions: Python3.8.13 | CPU: Intel(R) Xeon(R) CPU E5-2650 v4 @ 2.20 GHz | Ubuntu 20.04.1 |
Deep Learning Framework: Pytorch | GPU: NVIDIA Tesla V100 RAM: 32 G |
Method | PSNR | SSIM | MOS |
---|---|---|---|
Bicubic [7] | 26.15 | 0.69 | 3.15 |
SRCNN [9] | 26.25 | 0.71 | 2.55 |
FSRCNN [10] | 28.34 | 0.79 | 1.95 |
VDSR [11] | 29.16 | 0.80 | 1.98 |
SRGAN [14] | 31.45 | 0.86 | 4.41 |
SRVIT | 32.27 | 0.89 | 4.63 |
Method | Discriminator (Params/Model Size) | Generator (Params/Model Size) |
---|---|---|
SRGAN [14] | 5.22 M/19.9 MB | 0.73 M/2.9 MB |
SRVIT | 3.21 M/12.4 MB | 0.71 M/2.8 MB |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Liu, Z.; He, P.; Wang, F. PCB Defect Images Super-Resolution Reconstruction Based on Improved SRGAN. Appl. Sci. 2023, 13, 6786. https://doi.org/10.3390/app13116786
Liu Z, He P, Wang F. PCB Defect Images Super-Resolution Reconstruction Based on Improved SRGAN. Applied Sciences. 2023; 13(11):6786. https://doi.org/10.3390/app13116786
Chicago/Turabian StyleLiu, Zhihang, Pengfei He, and Feifei Wang. 2023. "PCB Defect Images Super-Resolution Reconstruction Based on Improved SRGAN" Applied Sciences 13, no. 11: 6786. https://doi.org/10.3390/app13116786