Segmenting Low-Contrast XCTs of Concretes: An Unsupervised Approach

Abstract

This work tests a self-annotation-based unsupervised methodology for training a convolutional neural network (CNN) model for semantic segmentation of X-ray computed tomography (XCT) scans of concretes. Concrete poses a unique challenge for XCT imaging due to similar X-ray attenuation coefficients of aggregates and mortar, resulting in low-contrast between the two phases in the ensuing images. While CNN-based models are a proven technique for semantic segmentation in such challenging cases, they typically require labeled training data, which is often unavailable for new datasets or are costly to obtain. To counter that limitation, a self-annotation technique is used here which leverages superpixel algorithms to identify perceptually similar local regions in an image and relates them to the global context in the image by utilizing the receptive field of a CNN-based model. This enables the model to learn a global-local relationship in the images and enables identification of semantically similar structures. We therefore present the performance of the unsupervised training methodology on our XCT datasets and discuss potential avenues for further improvements.

Figures (24)

• Figure 1: Example of the concrete XCT specimen in 3D.
• Figure 2: Example of low-contrast regions in an XCT image of concrete. (a)–(c) illustrate aggregates with only part of the aggregate-mortar interface being clearly discernible (indicated by - -). (d) highlights ambiguity in determining the aggregate-mortar interface on a part of the interface.
• Figure 3: Inter-sample variation of voxel intensity depicted using histograms. For each sample, the peak to the left corresponds to the porosity/air phase, while the peak to the right represents the aggregate and mortar phases in conjunction.
• Figure 5: Examples of concentric ring artifact (a), and bluring (b) seen near the centre of the cylinder and halo-effect (c), where regions around the sample show a bright band similar to a halo where no solid material is present.
• Figure 8: The U-Net model architecture. Green rectangles correspond to feature tensors. Input (grey rectangle) is the grayscale input image. Numbers above or below each rectangle indicate the number of feature channels (or layers or depth) of each tensor, which is also schematically represent the width of the rectangle, whereas the height of each rectangle illustrates the change in spatial dimension (height, width) along the network. Output tensor (in yellow) in this case has $c=3$, i.e. $3$ feature channels, one for each output class.