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

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 human-free segmentation pipeline. In this paper a modular interpretation of U-Net (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.

[1]  Serge Beucher,et al.  Watershed, Hierarchical Segmentation and Waterfall Algorithm , 1994, ISMM.

[2]  S. Beucher.  Use of watersheds in contour detection , 1979 .

[3]  Qingjie Liu,et al.  Road Extraction by Deep Residual U-Net , 2017, IEEE Geoscience and Remote Sensing Letters.

[4]  Rolf Adams,et al.  Seeded Region Growing , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Nima Tajbakhsh,et al.  UNet++: A Nested U-Net Architecture for Medical Image Segmentation , 2018, DLMIA/ML-CDS@MICCAI.

[6]  Yuzhe Yang,et al.  Rethinking the Value of Labels for Improving Class-Imbalanced Learning , 2020, NeurIPS.

[7]  Jérôme Darbon,et al.  Fast nonlocal filtering applied to electron cryomicroscopy , 2008, 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[8]  H. Robbins.  A Stochastic Approximation Method , 1951 .

[9]  Yoshua Bengio,et al.  Convolutional networks for images, speech, and time series , 1998 .

[10]  Jean-Michel Morel,et al.  Non-Local Means Denoising , 2011, Image Process. Line.

[11]  Trevor Darrell,et al.  Fully Convolutional Networks for Semantic Segmentation , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  Shin Ando,et al.  Deep Over-sampling Framework for Classifying Imbalanced Data , 2017, ECML/PKDD.

[13]  Andreas Mortensen,et al.  20 Hz X-ray tomography during an in situ tensile test , 2016, International Journal of Fracture.

[14]  Ling Shao,et al.  Striking the Right Balance With Uncertainty , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[15]  A. Phillion,et al.  Fast synchrotron X-ray tomographic quantification of dendrite evolution during the solidification of Mg-Sn alloys , 2016 .

[16]  Peter W. Voorhees,et al.  Optimizing convolutional neural networks to perform semantic segmentation on large materials imaging datasets: X-ray tomography and serial sectioning , 2020, Materials Characterization.

[17]  P. Withers,et al.  Quantitative X-ray tomography , 2014 .

[18]  François Goulette,et al.  On Power Jaccard Losses for Semantic Segmentation , 2021, VISIGRAPP.

[19]  Weitang Liu,et al.  Energy-based Out-of-distribution Detection , 2020, NeurIPS.

[20]  Kevin W Eliceiri,et al.  NIH Image to ImageJ: 25 years of image analysis , 2012, Nature Methods.

[21]  Roberto Cipolla,et al.  SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[22]  S. Wilkins,et al.  Simultaneous phase and amplitude extraction from a single defocused image of a homogeneous object , 2002, Journal of microscopy.

[23]  Francesco De Carlo,et al.  Automated correlative segmentation of large Transmission X-ray Microscopy (TXM) tomograms using deep learning , 2018, Materials Characterization.

[24]  J. Duncan,et al.  AdaBelief Optimizer: Adapting Stepsizes by the Belief in Observed Gradients , 2020, NeurIPS.

[25]  Giulia Pacchioni.  An upgrade to a bright future , 2019 .

[26]  W. Kalender.  X-ray computed tomography , 2006, Physics in medicine and biology.

[27]  Emmanuel Brun,et al.  PyHST2: an hybrid distributed code for high speed tomographic reconstruction with iterative reconstruction and a priori knowledge capabilities , 2013, ArXiv.

[28]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[29]  F ROSENBLATT,et al.  The perceptron: a probabilistic model for information storage and organization in the brain. , 1958, Psychological review.

[30]  G. Requena,et al.  Semantic segmentation of synchrotron tomography of multiphase Al-Si alloys using a convolutional neural network with a pixel-wise weighted loss function , 2019, Scientific Reports.

[31]  J. Hanley,et al.  The meaning and use of the area under a receiver operating characteristic (ROC) curve. , 1982, Radiology.

[32]  Matthias Weber,et al.  Machine Learning Techniques for the Segmentation of Tomographic Image Data of Functional Materials , 2019, Front. Mater..

[33]  Loïc Le Folgoc,et al.  Attention U-Net: Learning Where to Look for the Pancreas , 2018, ArXiv.

[34]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[35]  Johannes E. Schindelin,et al.  Fiji: an open-source platform for biological-image analysis , 2012, Nature Methods.

[36]  Colin Wei,et al.  Learning Imbalanced Datasets with Label-Distribution-Aware Margin Loss , 2019, NeurIPS.

[37]  Thomas Brox,et al.  3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation , 2016, MICCAI.

[38]  Shu-Ching Chen,et al.  Dynamic Sampling in Convolutional Neural Networks for Imbalanced Data Classification , 2018, 2018 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR).

[39]  Simon Dixon,et al.  Wave-U-Net: A Multi-Scale Neural Network for End-to-End Audio Source Separation , 2018, ISMIR.