A comparative study of Q-Seg, quantum-inspired techniques, and U-Net for crack image segmentation
-
Akshaya Srinivasan
,
Alexander Geng
Abstract
Exploring the potential of quantum hardware for enhancing classical and real-world applications is an ongoing challenge. This study evaluates the performance of quantum and quantum-inspired methods compared to classical models for crack segmentation. Using annotated grayscale image patches of concrete samples, we benchmark a classical mean Gaussian mixture technique, a quantum-inspired fermion-based method, Q-Seg – a quantum-annealing-based method, and a U-Net deep learning architecture. Our results indicate that quantum-inspired and quantum methods offer a promising alternative to image segmentation, particularly for complex crack patterns, and could be applied in near-future applications.
Zusammenfassung
Die Erforschung des Potenzials von Quanten-Hardware zur Verbesserung von klassischen und realen Anwendungen ist eine ständige Herausforderung. In dieser Studie wird die Leistung von Quanten- und quanteninspirierten Methoden im Vergleich zu klassischen Modellen zur Risssegmentierung bewertet. Anhand von kommentierten Graustufenbildfeldern von Betonproben vergleichen wir eine klassische mittlere Gaußsche Mischtechnik, eine quanteninspirierte fermionenbasierte Methode, Q-Seg – eine quantenannealingbasierte Methode und eine U-Net Deep Learning Architektur. Unsere Ergebnisse deuten darauf hin, dass quanteninspirierte und Quantenmethoden eine vielversprechende Alternative zur Bildsegmentierung bieten, insbesondere bei komplexen Rissmustern, und dass sie in naher Zukunft für Anwendungen eingesetzt werden könnten.
Funding source: Bundesministerium für Wirtschaft und Klimaschutz
Award Identifier / Grant number: EniQmA project – 01MQ22007A
About the authors
Akshaya Srinivasan is a physicist with an M.Sc. in Physics, specializing in Quantum Computing and Machine Learning. Since 2024, she has been a PhD student in the Image Processing department at the Fraunhofer Institute for Industrial Mathematics (ITWM) in Kaiserslautern, Germany. Her research focuses on Quantum Image Processing and Quantum Machine Learning.
Dr. Alexander Geng is a mathematician with a PhD in quantum image processing. He has been working as a researcher in the Image Processing department at the Fraunhofer Institute of Industrial Mathematics ITWM in Kaiserslautern, Germany, since 2024. His research focuses on quantum image processing, quantum machine learning, and the integration of classical image processing with quantum computing.
Dr. Antonio Macaluso is a Senior Researcher at the German Research Center for Artificial Intelligence (DFKI), Saarbruecken, Germany. His research interests span quantum algorithms for Artificial Intelligence, including supervised and reinforcement learning, and multi-agent systems. Dr. Macaluso received his Ph.D. in Computer Science and Engineering from the University of Bologna, Italy in 2021.
Dr. Maximilian Kiefer-Emmanouilidis is a physicist who earned his PhD in 2022, focusing on disordered quantum many-body systems. Afterwards, he became a Senior Researcher at the German Research Center for Artificial Intelligence (DFKI), Kaiserslautern, Germany and RPTU Kaiserslautern-Landau, Kaiserslautern, Germany. His research spans quantum generative models in quantum machine learning and the study of key physical effects in quantum and quantum-inspired neural networks.
Dr. Ali Moghiseh holds a PhD in Mathematics from the Geomathematics Group, Department of Mathematics, TU Kaiserslautern. Since joining the Image Processing Department at Fraunhofer ITWM in 2007, he has developed expertise in machine learning with a focus on industrial image processing. Since 2020, he has been leading a group in the field of quantum image processing and has extensive experience in applying quantum computing to image processing in the Noisy Intermediate Scale Quantum (NISQ) era.
Acknowledgments
We’d like to thank the Quantum Initiative Rhineland-Palatinate (QUIP) for their support. This work was also partially funded by the Research Initiative ‘Quantum Computing for Artificial Intelligence’ (QC-AI) and the Federal Ministry for Economic Affairs and Climate Action through the EniQmA project (funding number 01MQ22007A).
-
Research ethics: Not applicable.
-
Informed consent: Not applicable.
-
Author contributions: All authors have accepted responsibility for the entire content of this manuscript and approved its submission.
-
Use of Large Language Models, AI and Machine Learning Tools: AI Tools such as chatGPT 4o, DeepL and Grammarly have been used in the paper writing for tasks such as grammar correction, spell checking, simplification of sentences, translation tasks from German to English. The authors made sure to exclude personal data (Names, emails and other sensible data) from prompts. The authors have checked that there are no hallucinations in the final paper and are aware of their responsibilities when handling generative AI tools.
-
Conflict of interest: The authors state no conflict of interest.
-
Research funding: Research Initiative ‘Quantum Computing for Artificial Intelligence’ (QC-AI) and Federal Ministry for Economic Affairs and Climate Action through the EniQmA project (funding number 01MQ22007A).
-
Data availability: The raw data can be obtained on request from the corresponding author.
References
[1] J. Preskill, “Quantum computing in the NISQ era and beyond,” Quantum, vol. 2, p. 79, 2018, https://doi.org/10.22331/q-2018-08-06-79.Search in Google Scholar
[2] A. Geng, A. Moghiseh, C. Redenbach, and K. Schladitz, “Hybrid quantum transfer learning for crack image classification on NISQ hardware,” arXiv preprint arXiv:2307.16723, 2023.Search in Google Scholar
[3] A. Geng, “Hybrid quantum transfer learning for crack image classification on NISQ hardware,” Fraunhofer Fordatis, 2023. Available at: https://fordatis.fraunhofer.de/handle/fordatis/371.2.Search in Google Scholar
[4] T. Presles, C. Enderli, G. Burel, and E. H. Baghious, “Synthetic aperture radar image segmentation with quantum annealing,” IET Radar, Sonar Navig., vol. 18, no. 5, pp. 812–824, 2024. https://doi.org/10.1049/rsn2.12523.Search in Google Scholar
[5] S. M. Venkatesh, A. Macaluso, M. Nuske, M. Klusch, and A. Dengel, “Q-Seg: quantum annealing-based unsupervised image segmentation,” IEEE Comput. Graph. Appl., vol. 44, no. 5, pp. 27–39, 2024. https://doi.org/10.1109/mcg.2024.3455012.Search in Google Scholar
[6] J. Bowles, S. Ahmed, and M. Schuld, “Better than classical? The subtle art of benchmarking quantum machine learning models,” arXiv:2403.07059, 2024.Search in Google Scholar
[7] A. P. Dempster, N. M. Laird, and D. B. Rubin, “Maximum likelihood from incomplete data via the em algorithm,” J. Roy. Stat. Soc. B, vol. 39, no. 1, pp. 1–22, 1977. https://doi.org/10.1111/j.2517-6161.1977.tb01600.x.Search in Google Scholar
[8] G. J. McLachlan and D. Peel, Finite Mixture Models, New York City, United States, John Wiley & Sons, 2000.10.1002/0471721182Search in Google Scholar
[9] N. Otsu, et al.., “A threshold selection method from gray-level histograms,” Automatica, vol. 11, nos. 285–296, pp. 23–27, 1975.Search in Google Scholar
[10] N. Palaiodimopoulos, V. F. Rey, M. Tschöpe, C. Jörg, P. Lukowicz, and M. Kiefer-Emmanouilidis, “Quantum inspired image augmentation applicable to waveguides and optical image transfer via anderson localization,” in ICASSP 2024 – 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024, pp. 116–120.10.1109/ICASSP48485.2024.10447416Search in Google Scholar
[11] B. Singh, S. Majumdar, and S. Indu, “A systematic comparative analysis of quantum mechanics-based image processing and denoising algorithms,” Quantum Stud. Math. Found., vol. 11, no. 3, pp. 427–458, 2024. https://doi.org/10.1007/s40509-024-00330-x.Search in Google Scholar
[12] M. Moore and A. Narayanan, “Quantum-inspired computing,” Exeter, UK, Dept. Comput. Sci., Univ. Exeter, 1995.Search in Google Scholar
[13] L. Huynh, J. Hong, A. Mian, H. Suzuki, Y. Wu, and S. Camtepe, “Quantum-inspired machine learning: a survey,” arXiv preprint arXiv:2308.11269, 2023.Search in Google Scholar
[14] I. Peschel, “On the reduced density matrix for a chain of free electrons,” J. Stat. Mech. Theory Exp., vol. 2004, no. 6, p. P06004, 2004. https://doi.org/10.1088/1742-5468/2004/06/p06004.Search in Google Scholar
[15] S. Dutta, A. Basarab, B. Georgeot, and D. Kouamé, “Quantum mechanics-based signal and image representation: application to denoising,” IEEE Open J. Signal Process., vol. 2, pp. 190–206, 2021, https://doi.org/10.1109/ojsp.2021.3067507.Search in Google Scholar
[16] E. Abrahams, 50 Years of Anderson Localization, vol. 24, Singapore, World Scientific, 2010.10.1142/9789814299084Search in Google Scholar
[17] P. W. Anderson, “Absence of diffusion in certain random lattices,” Phys. Rev., vol. 109, no. 5, pp. 1492–1505, 1958. https://doi.org/10.1103/physrev.109.1492.Search in Google Scholar
[18] N. Palaiodimopoulos, et al.., “DisQu: investigating the impact of disorder in quantum generative models,” arXiv preprint arXiv:2409.16180, 2024.Search in Google Scholar
[19] A. Mafi and J. Ballato, “Review of a decade of research on disordered anderson localizing optical fibers,” Front. Phys., vol. 9, 2021, https://doi.org/10.3389/fphy.2021.736774.Search in Google Scholar
[20] A. Das and B. K. Chakrabarti, “Colloquium: quantum annealing and analog quantum computation,” Rev. Mod. Phys., vol. 80, no. 3, pp. 1061–1081, 2008. https://doi.org/10.1103/revmodphys.80.1061.Search in Google Scholar
[21] E. Farhi, J. Goldstone, S. Gutmann, and M. Sipser, “Quantum computation by adiabatic evolution,” arXiv:quant-ph/0001106, 2000.Search in Google Scholar
[22] F. Glover, G. Kochenberger, R. Hennig, and Y. Du, “Quantum bridge analytics I: a tutorial on formulating and using qubo models,” Ann. Oper. Res., vol. 314, no. 1, pp. 141–183, 2022. https://doi.org/10.1007/s10479-022-04634-2.Search in Google Scholar
[23] S. M. Venkatesh, A. Macaluso, and M. Klusch, “Gcs-q: quantum graph coalition structure generation,” in International Conference on Computational Science, Springer, 2023, pp. 138–152.10.1007/978-3-031-36030-5_11Search in Google Scholar
[24] O. Ronneberger, P. Fischer, and T. Brox, “U-Net: convolutional networks for biomedical image segmentation,” in Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5–9, 2015, Proceedings, Part III 18, Springer, 2015, pp. 234–241.10.1007/978-3-319-24574-4_28Search in Google Scholar
[25] H. Iqbal, “Harisiqbal88/plotneuralnet v1.0.0,” Zenodo, 2018. https://doi.org/10.5281/zenodo.2526395.Search in Google Scholar
[26] A. Geng and A. Moghiseh, “HairWidthCracks dataset,” Fraunhofer Fordatis, 2024. Available at: https://fordatis.fraunhofer.de/handle/fordatis/429.Search in Google Scholar
[27] B. Cheng, R. Girshick, P. Dollár, A. C. Berg, and A. Kirillov, “Boundary iou: improving object-centric image segmentation evaluation,” arXiv preprint arXiv:2103.16562, 2021.10.1109/CVPR46437.2021.01508Search in Google Scholar
[28] A. A. Taha and A. Hanbury, “Metrics for evaluating 3d medical image segmentation: analysis, selection, and tool,” BMC Med. Imag., vol. 15, no. 1, pp. 1–28, 2015. https://doi.org/10.1186/s12880-015-0068-x.Search in Google Scholar PubMed PubMed Central
[29] V. Yeghiazaryan and I. Voiculescu, “Family of boundary overlap metrics for the evaluation of medical image segmentation,” J. Med. Imag., vol. 5, no. 1, p. 015006, 2018. https://doi.org/10.1117/1.jmi.5.1.015006.Search in Google Scholar PubMed PubMed Central
[30] R. Okuta, Y. Unno, D. Nishino, S. Hido, and C. Loomis, “Cupy: a numpy-compatible library for nvidia gpu calculations,” in Proceedings of Workshop on Machine Learning Systems (LearningSys) in the Thirty-First Annual Conference on Neural Information Processing Systems, NIPS, 2017.Search in Google Scholar
[31] S. Barbosa, M. Kiefer-Emmanouilidis, F. Lang, J. Koch, and A. Widera, “Characterizing localization effects in an ultracold disordered fermi gas by diffusion analysis,” Phys. Rev. Res., vol. 6, no. 3, p. 033039, 2024. https://doi.org/10.1103/physrevresearch.6.033039.Search in Google Scholar
[32] W. S. Bakr, J. I. Gillen, A. Peng, S. Fölling, and M. Greiner, “A quantum gas microscope for detecting single atoms in a hubbard-regime optical lattice,” Nature, vol. 462, pp. 74–77, 2009, https://doi.org/10.1038/nature08482.Search in Google Scholar PubMed
© 2025 Walter de Gruyter GmbH, Berlin/Boston
Articles in the same Issue
- Frontmatter
- Editorial
- Image Processing Forum – Forum Bildverarbeitung 2024
- Research Articles
- Optimizing speed and quality in rendering of sensor-realistic images
- Characterisation metrics for performance comparison of area scan and event-based sensors
- A comparative study on multi-task uncertainty quantification in semantic segmentation and monocular depth estimation
- Comparing fast semantic segmentation CNNs for FPGAs with standard methods
- A comparative study of Q-Seg, quantum-inspired techniques, and U-Net for crack image segmentation
- Bayesian optimization of single-pulse laser drilling using advanced image processing
- Robuste Ampeldetektion und Haltelinienfreigabe durch HD-Karten Assoziation
Articles in the same Issue
- Frontmatter
- Editorial
- Image Processing Forum – Forum Bildverarbeitung 2024
- Research Articles
- Optimizing speed and quality in rendering of sensor-realistic images
- Characterisation metrics for performance comparison of area scan and event-based sensors
- A comparative study on multi-task uncertainty quantification in semantic segmentation and monocular depth estimation
- Comparing fast semantic segmentation CNNs for FPGAs with standard methods
- A comparative study of Q-Seg, quantum-inspired techniques, and U-Net for crack image segmentation
- Bayesian optimization of single-pulse laser drilling using advanced image processing
- Robuste Ampeldetektion und Haltelinienfreigabe durch HD-Karten Assoziation
