Alleviate Data Scarcity in Remanufacturing: Classifying the Reusability of Parts with Data-Efficient Generative Adversarial Networks (DE-GANs)
Source abstract
Abstract
Remanufacturing, a key element of the circular economy, enables products and parts to have new life cycles through a systematic process. Initially, used products (cores) are visually inspected and categorized according to their manufacturer and variant before being disassembled and cleaned. Subsequently, parts are manually classified as directly reusable, reusable after reconditioning, or recyclable. As demand for remanufactured parts increases, automated classification becomes crucial. However, current Deep Learning (DL) methods, constrained by the scarcity of unique parts, often suffer from insufficient datasets, leading to overfitting. This research explores the effectiveness of Data-Efficient Generative Adversarial Network (DE-GAN) optimization approaches like FastGAN, APA, and InsGen in enhancing dataset diversity. These methods were evaluated against the State-of-the-Art (SOTA) Deep Convolutional Generative Adversarial Network (DCGAN) using metrics such as the Inception Score (IS), Fréchet Inception Distance (FID), and the classification accuracy of ResNet18 models trained with partially synthetic data. FastGAN achieved the lowest FID values among all models and led to a statistically significant improvement in ResNet18 classification accuracy. At a [1:1] real-to-synthetic ratio, the mean accuracy increased from 72% ± 4% (real-data-only) to 87% ± 3% (p < 0.001), and reached 94% ± 3% after hyperparameter optimization. In contrast, synthetic data generated by the SOTA DCGAN did not yield statistically significant improvements.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Industrial Vision Systems and Defect Detection · Image Processing and 3D Reconstruction · Additive Manufacturing Materials and Processes
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Fraunhofer Institute for Manufacturing Engineering and Automation IPA, Universitaetsstrasse 9, 95447 Bayreuth, Germany
Fraunhofer Institute for Manufacturing Engineering and Automation IPA, Nobelstrasse 12, 70569 Stuttgart, Germany
Chair Manufacturing and Remanufacturing Technology, University of Bayreuth, Universitaetsstrasse 30, 95447 Bayreuth, Germany
Funding
Federal Ministry of Education and Research Deutschland
References (89)
Show source-provided references
1. E.M. Foundation (2015). Towards a Circular Economy: Business Rationale for an Accelerated Transition 2015, Ellen MacArthur Foundation.
2. Köhler, D.C.F. (2011). Regenerative Supply Chains: Regenerative Wertschöpfungsketten, Shaker-Verlag.
3. Steinhilper, R. (1999). Produktrecycling: Vielfachnutzen Durch Mehrfachnutzung, Fraunhofer-IRB-Verlag.
4. Achieving remanufacturing inspection using deep learning10.1007/s13243-020-00093-9 ↗
5. Remanufacturing for the automotive aftermarket-strategic factors: Literature review and future research needs10.1016/j.jclepro.2009.03.004 ↗
6. Remanufacturing strategies: A solution for WEEE problem10.1016/j.jclepro.2017.02.004 ↗
7. Saiz, F.A., Alfaro, G., and Barandiaran, I. (2021). An Inspection and Classification System for Automotive Component Remanufacturing Industry Based on Ensemble Learning. Information, 12.10.3390/info12120489 ↗
8. Nwankpa, C., Eze, S., Ijomah, W., Gachagan, A., and Marshal, S. (2019, January 23–25). Deep Learning Based Vision Inspection System for Remanufacturing Application. Proceedings of the International Conference on Remanufacturing, Amsterdam, The Netherlands.
9. Reverse logistics challenges in remanufacturing of automotive mechatronic devices10.1186/2210-4690-3-2 ↗
10. Casper, R. (2021). Automotive Remanufacturing in a Changing Market: Challenges and Opportunities in a Market with a Growing Share of Electric Cars, Linköping University Electronic Press. [1st ed.].10.3384/diss.diva-180947 ↗
11. Lundmark, P., Sundin, E., and Björkman, M. (2025, September 01). Industrial Challenges Within the Remanufacturing System. Available online: https://www.researchgate.net/publication/229049090_Industrial_challenges_within_the_remanufacturing_system.
12. Gao, H., Zhang, Y., Lv, W., Yin, J., Qasim, T., and Wang, D. (2022). A Deep Convolutional Generative Adversarial Networks-Based Method for Defect Detection in Small Sample Industrial Parts Images. Appl. Sci., 12.10.3390/app12136569 ↗
13. Augmenting data with generative adversarial networks: An overview10.3233/IDA-215735 ↗
14. Tanaka, F.H.K.d.S., and Aranha, C. (2025, September 01). Data Augmentation Using GANs. Available online: http://arxiv.org/pdf/1904.09135v1.
15. Antoniou, A., Storkey, A., and Edwards, H. (2025, September 01). Data Augmentation Generative Adversarial Networks. Available online: http://arxiv.org/pdf/1711.04340v3.
16. Goodfellow, I.J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y. (2025, September 01). Generative Adversarial Networks. Available online: http://arxiv.org/pdf/1406.2661v1.
17. Bowles, C., Chen, L., Guerrero, R., Bentley, P., Gunn, R., Hammers, A., Dickie, D.A., Hernández, M.V., Wardlaw, J., and Rueckert, D. (2025, September 01). GAN Augmentation: Augmenting Training Data Using Generative Adversarial Networks. Available online: http://arxiv.org/pdf/1810.10863v1.
18. GANs for generation of synthetic ultrasound images from small datasets10.1515/cdbme-2022-0005 ↗
19. Saiz, F.A., Alfaro, G., Barandiaran, I., and Graña, M. (2021). Generative Adversarial Networks to Improve the Robustness of Visual Defect Segmentation by Semantic Networks in Manufacturing Components. Appl. Sci., 11.10.3390/app11146368 ↗
20. Olaniyi, E., Chen, D., Lu, Y., and Huang, Y. (2025, September 01). Generative Adversarial Networks for Image Augmentation in Agriculture: A Systematic Review. Available online: http://arxiv.org/pdf/2204.04707v2.
21. Zhou, J. (2022, January 20–21). Research on Generative Adversarial Networks and Their Applications in Image Generation. Proceedings of the 2022 IEEE International Conference on Advances in Electrical Engineering and Computer Applications (AEECA), Dalian, China.10.1109/AEECA55500.2022.9918833 ↗
22. Li, Z., Xia, B., Zhang, J., Wang, C., and Li, B. (2025, September 01). A Comprehensive Survey on Data-Efficient GANs in Image Generation. Available online: http://arxiv.org/pdf/2204.08329v2.
23. Yang, M., and Wang, Z. (2025, September 01). Image Synthesis Under Limited Data: A Survey and Taxonomy. Available online: http://arxiv.org/pdf/2307.16879.
24. Katharopoulos, A., and Fleuret, F. (2025, September 01). Not All Samples Are Created Equal: Deep Learning with Importance Sampling. Available online: http://arxiv.org/pdf/1803.00942v3.
25. Zhang, Z., Chen, X., Chen, T., and Wang, Z. (2025, September 01). Efficient Lottery Ticket Finding: Less Data Is More. Available online: http://arxiv.org/pdf/2106.03225v1.
26. DeVries, T., Drozdzal, M., and Taylor, G.W. (2025, September 01). Instance Selection for GANs. Available online: http://arxiv.org/pdf/2007.15255v2.
27. Achicanoy, H., Chaves, D., and Trujillo, M. (2021). StyleGANs and Transfer Learning for Generating Synthetic Images in Industrial Applications. Symmetry, 13.10.3390/sym13081497 ↗
28. Grigoryev, T., Voynov, A., and Babenko, A. (2025, September 01). When, Why, and Which Pretrained GANs Are Useful?. Available online: http://arxiv.org/pdf/2202.08937v2.
29. Kumari, N., Zhang, R., Shechtman, E., and Zhu, J.-Y. (2025, September 01). Ensembling Off-the-Shelf Models for GAN Training. Available online: http://arxiv.org/pdf/2112.09130v3.
30. Mangla, P., Kumari, N., Singh, M., Krishnamurthy, B., and Balasubramanian, V.N. (2025, September 01). Data InStance Prior (DISP) in Generative Adversarial Networks. Available online: http://arxiv.org/pdf/2012.04256v2.
31. Noguchi, A., and Harada, T. (2025, September 01). Image Generation From Small Datasets via Batch Statistics Adaptation. Available online: http://arxiv.org/pdf/1904.01774v4.
32. Ojha, U., Li, Y., Lu, J., Efros, A.A., Lee, Y.J., Shechtman, E., and Zhang, R. (2025, September 01). Few-shot Image Generation via Cross-Domain Correspondence. Available online: http://arxiv.org/pdf/2104.06820v1.
33. Robb, E., Chu, W.-S., Kumar, A., and Huang, J.-B. (2025, September 01). Few-Shot Adaptation of Generative Adversarial Networks. Available online: http://arxiv.org/pdf/2010.11943v1.
34. Toneva, M., Sordoni, A., Des Combes, R.T., Trischler, A., Bengio, Y., and Gordon, G.J. (2025, September 01). An Empirical Study of Example Forgetting During Deep Neural Network Learning. Available online: http://arxiv.org/pdf/1812.05159v3.
35. Wang, Y., Gonzalez-Garcia, A., Berga, D., Herranz, L., Khan, F.S., and van de Weijer, J. (2025, September 01). MineGAN: Effective Knowledge Transfer from GANs to Target Domains with Few Images. Available online: http://arxiv.org/pdf/1912.05270v3.
36. Wang, Y., Gonzalez-Garcia, A., Wu, C., Herranz, L., Khan, F.S., Jui, S., and van de Weijer, J. (2025, September 01). MineGAN++: Mining Generative Models for Efficient Knowledge Transfer to Limited Data Domains. Available online: http://arxiv.org/pdf/2104.13742v1.
37. Yang, C., Shen, Y., Zhang, Z., Xu, Y., Zhu, J., Wu, Z., and Zhou, B. (2025, September 01). One-Shot Generative Domain Adaptation. Available online: http://arxiv.org/pdf/2111.09876v1.
38. Zhao, M., Cong, Y., and Carin, L. (2025, September 01). On Leveraging Pretrained GANs for Generation with Limited Data. Available online: http://arxiv.org/pdf/2002.11810v3.
39. Karras, T., Aittala, M., Hellsten, J., Laine, S., Lehtinen, J., and Aila, T. (2025, September 01). Training Generative Adversarial Networks with Limited Data. Available online: http://arxiv.org/pdf/2006.06676v2.
40. Jiang, L., Dai, B., Wu, W., and Loy, C.C. (2025, September 01). Deceive D Adaptive Pseudo Augmentation for GAN Training with Limited Data. Available online: https://arxiv.org/pdf/2111.06849.
41. On Data Augmentation for GAN Training10.1109/TIP.2021.3049346 ↗
42. Zhao, S., Liu, Z., Lin, J., Zhu, J.-Y., and Han, S. (2025, September 01). Differentiable Augmentation for Data-Efficient GAN Training. Available online: http://arxiv.org/pdf/2006.10738v4.
43. Zhao, Z., Zhang, Z., Chen, T., Singh, S., and Zhang, H. (2025, September 01). Image Augmentations for GAN Training. Available online: http://arxiv.org/pdf/2006.02595v1.
44. Dai, M., Hang, H., and Guo, X. (2025, September 01). Adaptive Feature Interpolation for Low-Shot Image Generation. Available online: http://arxiv.org/pdf/2112.02450v3.
45. Jeong, J., and Shin, J. (2025, September 01). Training GANs with Stronger Augmentations via Contrastive Discriminator. Available online: http://arxiv.org/pdf/2103.09742.
46. Hou, L., Cao, Q., Shen, H., Pan, S., Li, X., and Cheng, X. (2025, September 01). Augmentation-Aware Self-Supervision for Data-Efficient GAN Training. Available online: http://arxiv.org/pdf/2205.15677v1.
47. Masked Generative Adversarial Networks are Data-Efficient Generation Learners
48. A Systematic Survey of Regularization and Normalization in GANs
49. Tseng, H.-Y., Jiang, L., Liu, C., Yang, M.-H., and Yang, W. (2025, September 01). Regularizing Generative Adversarial Networks Under Limited Data. Available online: http://arxiv.org/pdf/2104.03310v1.
50. Fang, T., Sun, R., and Schwing, A. (2025, September 01). DigGAN: Discriminator gradIent Gap Regularization for GAN Training with Limited Data. Available online: http://arxiv.org/pdf/2211.14694.
51. ProtoGAN: Towards high diversity and fidelity image synthesis under limited data10.1016/j.ins.2023.03.042 ↗
52. Kim, J., Choi, Y., and Uh, Y. (2025, September 01). Feature Statistics Mixing Regularization for Generative Adversarial Networks. Available online: http://arxiv.org/pdf/2112.04120v2.
53. Pengwei, W. (2025, September 01). Latent Space is Feature Space: Regularization Term for GANs Training on Limited Dataset. Available online: https://arxiv.org/pdf/2210.16251.
54. Yang, C., Shen, Y., Xu, Y., and Zhou, B. (2025, September 01). Data-Efficient Instance Generation from Instance Discrimination. Available online: https://arxiv.org/pdf/2106.04566.
55. Kong, C., Kim, J., Han, D., and Kwak, N. (2025, September 01). Few-Shot Image Generation with Mixup-Based Distance Learning. Available online: http://arxiv.org/pdf/2111.11672v2.
56. Chu, C., Minami, K., and Fukumizu, K. (2020). Smoothness and Stability in GANs. arXiv.
57. Karras, T., Laine, S., and Aila, T. (2025, September 01). A Style-Based Generator Architecture for Generative Adversarial Networks. Available online: http://arxiv.org/pdf/1812.04948v3.
58. Karras, T., Laine, S., Aittala, M., Hellsten, J., Lehtinen, J., and Aila, T. (2025, September 01). Analyzing and Improving the Image Quality of StyleGAN. Available online: http://arxiv.org/pdf/1912.04958v2.
59. Karras, T., Aila, T., Laine, S., and Lehtinen, J. (2025, September 01). Progressive Growing of GANs for Improved Quality, Stability, and Variation. Available online: http://arxiv.org/pdf/1710.10196v3.
60. Liu, B., Zhu, Y., Song, K., and Elgammal, A. (2025, September 01). Towards Faster and Stabilized GAN Training for High-Fidelity Few-Shot Image Synthesis. Available online: http://arxiv.org/pdf/2101.04775v1.
61. Li, T., Li, Z., Luo, A., Rockwell, H., Farimani, A., and Lee, T.S. (2022, January 25–29). Prototype Memory and Attention Mechanisms for Few Shot Image Generation. Proceedings of the International Conference on Learning Representations 2022, Online.
62. Yang, M., Wang, Z., Chi, Z., and Zhang, Y. (2025, September 01). FreGAN: Exploiting Frequency Components for Training GANs Under Limited Data. Available online: https://arxiv.org/pdf/2210.05461.
63. Cui, K., Huang, J., Luo, Z., Zhang, G., Zhan, F., and Lu, S. (2025, September 01). GenCo: Generative Co-Training for Generative Adversarial Networks with Limited Data. Available online: http://arxiv.org/pdf/2110.01254v2.
64. Yang, C., Shen, Y., Xu, Y., Zhao, D., Dai, B., and Zhou, B. (2025, September 01). Improving GANs with A Dynamic Discriminator. Available online: https://arxiv.org/pdf/2209.09897.
65. Chen, T., Cheng, Y., Gan, Z., Liu, J., and Wang, Z. (2025, September 01). Data-Efficient GAN Training Beyond (Just) Augmentations: A Lottery Ticket Perspective. Available online: https://arxiv.org/pdf/2103.00397.
66. AutoInfo GAN: Toward a better image synthesis GAN framework for high-fidelity few-shot datasets via NAS and contrastive learning10.1016/j.knosys.2023.110757 ↗
67. Saxena, D., Cao, J., Xu, J., and Kulshrestha, T. (2023, January 17–24). Re-GAN: Data-Efficient GANs Training via Architectural Reconfiguration. Proceedings of the Conference on Computer Vision and Pattern Recognition (CVPR) 2023, Vancouver, BC, Canada.10.1109/CVPR52729.2023.01557 ↗
68. With synthetic data towards part recognition generalized beyond the training instances10.1063/5.0189847 ↗
69. A Framework for Anomaly Classification and Segmentation in Remanufacturing using Autoencoders and Simulated Data10.1016/j.procir.2024.08.383 ↗
70. Wu, C., Zhou, K., Kaiser, J.-P., Mitschke, N., Klein, J.-F., Pfrommer, J., Beyerer, J., Lanza, G., Heizmann, M., and Furmans, K. (2025, September 01). MotorFactory: A Blender Add-On for Large Dataset Generation of Small Electric Motors. Available online: https://arxiv.org/pdf/2301.05028.
71. Recognition of car parts in automotive supply chains by combining synthetically generated training data with classical and deep learning based image processing10.1016/j.procir.2020.03.142 ↗
72. AI-Driven Crack Detection for Remanufacturing Cylinder Heads Using Deep Learning and Engineering-Informed Data Augmentation10.3390/automation5040033 ↗
73. An efficient method to determine sample size in oversampling based on classification complexity for imbalanced data10.1016/j.eswa.2021.115442 ↗
74. Radford, A., Metz, L., and Chintala, S. (2025, September 01). Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks. Available online: https://arxiv.org/pdf/1511.06434.
75. The Fréchet distance between multivariate normal distributions10.1016/0047-259X(82)90077-X ↗
76. Shirazi, M., Schmitz, M., Janssen, S., Thies, A., Safronov, G., Rizk, A., Mayr, P., and Engelhardt, P. (2021, January 13–16). Verifying the Applicability of Synthetic Image Generation for Object Detection in Industrial Quality Inspection. Proceedings of the IEEE International Conference on Machine Learning and Applications (ICMLA) 2021, Pasadena, CA, USA.10.1109/ICMLA52953.2021.00221 ↗
77. Salimans, T., Goodfellow, I., Zaremba, W., Cheung, V., Radford, A., and Chen, X. (2025, September 01). Improved Techniques for Training GANs. Available online: https://arxiv.org/pdf/1606.03498.
78. Maturana, C.N.V., Orozco, A.L.S., and Villalba, L.J.G. (2023). Exploration of Metrics and Datasets to Assess the Fidelity of Images Generated by Generative Adversarial Networks. Appl. Sci., 13.10.3390/app131910637 ↗
79. He, K., Zhang, X., Ren, S., and Sun, J. (2016, January 27–30). Deep Residual Learning for Image Recognition. Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.10.1109/CVPR.2016.90 ↗
80. Image Generation of Tomato Leaf Disease Identification Based on Small-ACGAN
81. Hutter, F., Kotthoff, L., and Vanschoren, J. (2019). Hyperparameter Optimization. Automated Machine Learning: Methods, Systems, Challenges, Springer.10.1007/978-3-030-05318-5 ↗
82. Masters, D., and Luschi, C. (2025, September 01). Revisiting Small Batch Training for Deep Neural Networks. Available online: http://arxiv.org/pdf/1804.07612.
83. Kossale, Y., Airaj, M., and Darouichi, A. (2022, January 6–7). Mode Collapse in Generative Adversarial Networks: An Overview. Proceedings of the International Conference on Optimization and Applications (ICOA), Genoa, Italy.10.1109/ICOA55659.2022.9934291 ↗
84. Karnewar, A., and Wang, O. (2025, September 01). MSG-GAN: Multi-Scale Gradients for Generative Adversarial Networks. Available online: http://arxiv.org/pdf/1903.06048v4.
85. Barratt, S., and Sharma, R. (2025, September 01). A Note on the Inception Score. Available online: https://arxiv.org/pdf/1801.01973.
86. Jordon, J., Szpruch, L., Houssiau, F., Bottarelli, M., Cherubin, G., Maple, C., Cohen, S.N., and Weller, A. (2025, September 01). Synthetic Data—What, Why and How?. Available online: http://arxiv.org/pdf/2205.03257v1.
87. Alaa, A.M., van Breugel, B., Saveliev, E., and van der Schaar, M. (2025, September 01). How Faithful Is Your Synthetic Data? Sample-Level Metrics for Evaluating and Auditing Generative Models. Available online: http://arxiv.org/pdf/2102.08921.
88. Parker, D., Riley, K., Robinson, S., Symington, H., Teweson, J., Jannson, K., Ramkumar, S., and Peck, D. (2025, May 28). Remanufacturing Market Study. Available online: https://www.remanufacturing.eu/assets/pdfs/remanufacturing-market-study.pdf.
89. Dutschke, W. (2002). Fertigungsmesstechnik, Vieweg+Teubner Verlag. [4th ed.].10.1007/978-3-322-96810-4 ↗