Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
162 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

A modular U-Net for automated segmentation of X-ray tomography images in composite materials (2107.07468v2)

Published 15 Jul 2021 in eess.IV and cs.CV

Abstract: X-ray Computed Tomography (XCT) techniques have evolved to a point that high-resolution data can be acquired so fast that classic segmentation methods are prohibitively cumbersome, demanding automated data pipelines capable of dealing with non-trivial 3D images. Deep learning has demonstrated success in many image processing tasks, including material science applications, showing a promising alternative for a humanfree segmentation pipeline. In this paper a modular interpretation of UNet (Modular U-Net) is proposed and trained to segment 3D tomography images of a three-phased glass fiber-reinforced Polyamide 66. We compare 2D and 3D versions of our model, finding that the former is slightly better than the latter. We observe that human-comparable results can be achievied even with only 10 annotated layers and using a shallow U-Net yields better results than a deeper one. As a consequence, Neural Network (NN) show indeed a promising venue to automate XCT data processing pipelines needing no human, adhoc intervention.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (45)
  1. Withers, P. J. et al. X-ray computed tomography. Nature Reviews Methods Primers 1, 1–21 (2021).
  2. Quantitative x-ray tomography. International Materials Reviews 59, 1–43 (2014).
  3. Pacchioni, G. An upgrade to a bright future. Nature Reviews Physics 1, 100–101 (2019).
  4. Shuai, S. et al. Fast synchrotron x-ray tomographic quantification of dendrite evolution during the solidification of mgsn alloys. Acta Materialia 118, 260–269 (2016).
  5. 20 hz x-ray tomography during an in situ tensile test. International Journal of Fracture 200, 3–12 (2016).
  6. Shashank Kaira, C. et al. Automated correlative segmentation of large Transmission X-ray Microscopy (TXM) tomograms using deep learning. Materials Characterization 142, 203–210 (2018).
  7. Strohmann, T. et al. Semantic segmentation of synchrotron tomography of multiphase Al-Si alloys using a convolutional neural network with a pixel-wise weighted loss function. Scientific Reports 9, 1–9 (2019).
  8. Beucher, S. Watershed, Hierarchical Segmentation and Waterfall Algorithm. In Morphology and Its Applications to Image Processing, vol. 2, 69–76 (1994).
  9. Use of Watersheds in Contour Detection.In International Workshop on Image Processing : Real-time Edge and Motion Detection/Estimation, Rennes, France. (1979).
  10. F.Rosenblatt. The perceptron: a probabilistic model for information storage and organization in the brain. Psychological review 65 6, 386–408 (1958).
  11. Convolutional networks for images, speech, and time-series (1997).
  12. Abadi, M. et al. TensorFlow: Large-scale machine learning on heterogeneous systems (2015). URL https://www.tensorflow.org/. Software available from tensorflow.org.
  13. Fully Convolutional Networks for Semantic Segmentation. arXiv:1411.4038 [cs] (2015). URL http://arxiv.org/abs/1411.4038. ArXiv: 1411.4038.
  14. Optimizing convolutional neural networks to perform semantic segmentation on large materials imaging datasets: X-ray tomography and serial sectioning. Materials Characterization 160, 110119 (2020).
  15. SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence 39, 2481–2495 (2017). URL 10.1109/TPAMI.2016.2644615. Conference Name: IEEE Transactions on Pattern Analysis and Machine Intelligence arxiv: 1511.00561.
  16. U-Net: Convolutional Networks for Biomedical Image Segmentation. arXiv:1505.04597 [cs] (2015). 1505.04597.
  17. 3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation. In Medical Image Computing and Computer-Assisted Intervention (MICCAI) (2016).
  18. U22{}^{2}start_FLOATSUPERSCRIPT 2 end_FLOATSUPERSCRIPT-Net: Going Deeper with Nested U-Structure for Salient Object Detection. Pattern Recognition 106, 107404 (2020). 2005.09007.
  19. UNet++: A Nested U-Net Architecture for Medical Image Segmentation. arXiv:1807.10165 [cs, eess, stat] (2018). URL http://arxiv.org/abs/1807.10165. ArXiv: 1807.10165.
  20. Oktay, O. et al. Attention U-Net: Learning Where to Look for the Pancreas. arXiv:1804.03999 [cs] (2018). 1804.03999.
  21. Wave-U-Net: A Multi-Scale Neural Network for End-to-End Audio Source Separation. arXiv:1806.03185 [cs, eess, stat] (2018). 1806.03185.
  22. Road Extraction by Deep Residual U-Net. IEEE Geoscience and Remote Sensing Letters 15, 749–753 (2018).
  23. Furat, O. et al. Machine Learning Techniques for the Segmentation of Tomographic Image Data of Functional Materials. Frontiers in Materials 6 (2019).
  24. NIH Image to ImageJ: 25 years of image analysis. Nature Methods 9, 671–675 (2012).
  25. Schindelin, J. et al. Fiji: An open-source platform for biological-image analysis. Nature Methods 9, 676–682 (2012).
  26. Glass fiber-reinforced polyamide 66 3D X-ray computed tomography dataset for deep learning segmentation (2021). URL https://doi.org/10.5281/zenodo.4587827.
  27. The pyhst2 hybrid distributed code for high speed tomographic reconstruction with iterative reconstruction and a priori knowledge capabilities. Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms 324, 41–48 (2014). 1st International Conference on Tomography of Materials and Structures.
  28. Simultaneous phase and amplitude extraction from a single defocused image of a homogeneous object. Journal of Microscopy 206, 33–40 (2002).
  29. Seeded region growing. IEEE Transactions on Pattern Analysis and Machine Intelligence 16, 641–647 (1994).
  30. Glass fiber-reinforced polyamide 66 3D X-ray computed tomography segmentation segmentation U-Nets (2021). URL https://doi.org/10.5281/zenodo.4601560.
  31. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 143, 29–36 (1982).
  32. Pouyanfar, S. et al. Dynamic Sampling in Convolutional Neural Networks for Imbalanced Data Classification. In 2018 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR), 112–117 (2018).
  33. Deep Over-sampling Framework for Classifying Imbalanced Data. In Ceci, M., Hollmén, J., Todorovski, L., Vens, C. & Džeroski, S. (eds.) Machine Learning and Knowledge Discovery in Databases, Lecture Notes in Computer Science, 770–785 (Springer International Publishing, Cham, 2017).
  34. Learning Imbalanced Datasets with Label-Distribution-Aware Margin Loss. arXiv:1906.07413 [cs, stat] (2019). 1906.07413.
  35. Striking the Right Balance with Uncertainty. arXiv:1901.07590 [cs] (2019). 1901.07590.
  36. Rethinking the Value of Labels for Improving Class-Imbalanced Learning. arXiv:2006.07529 [cs, stat] (2020). 2006.07529.
  37. Energy-based Out-of-distribution Detection. arXiv:2010.03759 [cs] (2020). 2010.03759.
  38. Non-Local Means Denoising. Image Processing On Line 1, 208–212 (2011).
  39. Fast nonlocal filtering applied to electron cryomicroscopy. In 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, 1331–1334 (2008).
  40. European Organization For Nuclear Research & OpenAIRE. Zenodo (2013). URL https://www.zenodo.org/.
  41. Duque-Arias, D. et al. On power Jaccard losses for semantic segmentation. In VISAPP 2021 : 16th International Conference on Computer Vision Theory and Applications (Vienne (on line), Austria, 2021).
  42. Zhuang, J. et al. AdaBelief Optimizer: Adapting Stepsizes by the Belief in Observed Gradients. arXiv:2010.07468 [cs, stat] (2020). 2010.07468.
  43. Adam: A Method for Stochastic Optimization. arXiv:1412.6980 [cs] (2017). 1412.6980.
  44. Robbins, H. A stochastic approximation method. Annals of Mathematical Statistics 22, 400–407 (2007).
  45. Chollet, F. et al. Keras. https://keras.io (2015).
Citations (12)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now