Improved XCT image automatic segmentation for quantitative characterization of the meso-morphological features in the damaged braided composite fabric

Kehong Zheng,Xiaoqi Cao,Zheyang Jiang,Hao Chen,Bingjing Qiu,Wenpan Lu,Chenglie Wu,Zhongxiang Pan,Zhenyu Wu

DOI: https://doi.org/10.1016/j.compscitech.2023.110395

IF: 9.1

2023-12-27

Composites Science and Technology

Abstract:Accurately identifying various meso-morphological features and sub-phases within the damaged braided composite fabric is crucial for assessing the mechanical properties of composites under complex loading conditions and different processing parameters. However, obtaining 3D reconstructions that can be quantitatively analyzed is still a challenge, primarily due to the low contrast of the meso-morphological features (yarn tow and matrix crack networks). In this work, a new data enhancement algorithm is proposed to generate a realistic-looking artificial training dataset for enriching the information of the meso-morphological features. Then, the effect of various training networks and training parameters (the size of real and hybrid training dataset, number of epochs) on segmentation performance were examined. Finally, the meso-morphological features were statistically analyzed to accurately evaluate the mechanical properties of the composite material. The work presented here provides an effective tool that enables the quantitative analysis of the dynamic crack evolution process under the axial load of the composite materials.

materials science, composites

What problem does this paper attempt to address?

This paper attempts to address the problem of accurately identifying and quantitatively characterizing various mesoscopic morphological features (such as yarn bundles and matrix crack networks) in damaged woven composites. Specifically, the authors focus on the impact of these features on the mechanical properties of composites under complex loading conditions and different processing parameters. However, due to the low contrast of mesoscopic morphological features, obtaining 3D reconstructed images that can be quantitatively analyzed remains a challenge. To tackle this challenge, the authors propose a new data augmentation algorithm that generates an artificial training dataset rich in mesoscopic morphological features and trains it using the Swin-Unet algorithm. Through this method, the authors aim to improve the accuracy of image segmentation, thereby enabling a more precise assessment of the mechanical properties of composites.

Improved XCT image automatic segmentation for quantitative characterization of the meso-morphological features in the damaged braided composite fabric

Progressive failure analysis of 2D woven composites at the meso-micro scale

Improved 3D image segmentation for X-ray tomographic data of biaxial warp-knitted composites

Trans-scale analysis of 3D braided composites with voids based on micro-CT imaging and unsupervised machine learning

An Improved Dataset Augmentation Approach for Deep Learning-Based XCT Images Segmentation in Layered Composite Fabric

3D meso-scale fracture modelling and validation of concrete based on in-situ X-ray Computed Tomography images using damage plasticity model

Damage Mechanism Characterisation of Plain Weave Ceramic Matrix Composites under In-Plane Shear Using In-Situ X-ray Micro-Ct and Deep-Learning-based Image Segmentation

Automated reconstruction and conformal discretization of 3D woven composite CT scans with local fiber volume fraction control

3 D In-situ XCT Image Based Mesoscale Fracture Modelling and Validation of Concrete Using Voxel Hexahedron Meshing and Damage Plasticity Model

Segmentation of computed tomography images and high-precision reconstruction of rubber composite structure based on deep learning

CT Image Segmentation Method of Composite Material Based on Improved Watershed Algorithm and U-Net Neural Network Model

Experimental investigation on shear damage evolution of 3D woven composites using in situ computed tomography and multi-scale digital image correlation

Predicting tensile behaviour of plain weave CMCs using a nonlinear data-driven constitutive model for fibre tow composites

Meso-scale modeling and damage analysis of carbon/epoxy woven fabric composite under in-plane tension and compression loadings

Characterisation of damage evolution in plain weave SiC/SiC composites using in situ X-ray micro-computed tomography

A Highly Efficient Self-Consistent Clustering Analysis Method with Field Refinement Capability for the Mesoscale Damage Behavior of 3D Woven Composites

Hierarchical Multi-Scale Analysis of the Effect of Varying Fiber Bundle Geometric Properties on the Mechanical Properties of 3D Braided Composites

Tensile Damage Evolution and Mechanical Behaviour of SiCf/SiC Mini-Composites Through 4D In-Situ Micro-Ct and Data-Driven Modelling

Visualization and interpretation of the impact failure behavior of textile composites using a highly efficient Meso-FE model

Micro-CT Characterization on the Meso-Structure of Three-Dimensional Full Five-Directional Braided Composite

Multiscale image-based modeling of progressive damage in 3D5D braided composites with yarn-reduction