arXiv subjects

Qinyi Tian

Publications and source records attributed to Qinyi Tian.

3 recordsLinked to original sources

arXiv2025

StrengthLawExtractor: A Fiji plugin for 3D morphological feature extraction from X-ray micro-CT data

Non-destructive methods are essential for linking the microstructural geometry of porous materials to their mechanical behavior, as destructive testing is often infeasible due to limited material availability or irreproducible conditions. Micro-computed tomography (micro-CT) provides high resolution three dimensional reconstructions of porous microstructures, enabling direct quantification of geometric descriptors. Recent advances in morphometric theory have demonstrated that four independent morphometric measures (porosity, surface area, mean curvature, and Euler characteristic) are required to capture the relationship between microstructure and strength, thereby forming the basis of generalized strength laws. To facilitate practical application of this framework, a Fiji plugin was developed to extract the four morphometric measures (porosity, surface area, mean curvature, Euler characteristic) from micro-CT datasets automatically. The plugin integrates within the Fiji platform to provide reproducible, accessible, and user friendly analysis. The application of the tool demonstrates that the extracted descriptors can be readily incorporated into constitutive models and machine learning workflows, enabling the forward prediction of stress-strain behavior as well as the inverse design of microstructures. This approach supports non-destructive evaluation, accelerates materials selection, and advances the integration of imaging with predictive modeling in porous media research.

cs.CE↗

arXiv2025

InSituRes: A Physics-Informed Same-Grid Model for Enhanced Dynamic X-ray Micro-CT Reconstructions

X-ray micro-computed tomography (micro-CT) provides non-destructive three-dimensional (3D) imaging of porous material microstructures. In situ experiments, including mechanical loading and reactive transport, increasingly require dynamic four-dimensional (4D) imaging with volumes repeatedly acquired during experiments. However, rapid acquisition typically requires fewer projections, shorter exposures, or reduced fields of view, producing reconstructions with noise, blur, and artifacts that obscure pores, microcracks, and interfaces. To address this challenge, this study introduces InSituRes, a physics-informed same-grid volumetric enhancement framework for fast dynamic X-ray micro-CT imaging of temporally evolving materials. InSituRes maps fast-acquisition volumes to higher-quality long-acquisition reconstructions using paired scans of the same specimens. The model integrates 3D convolutional feature extraction with slice-wise transformer attention to capture local and broader in-plane context. A learnable forward degradation model approximates rapid acquisition effects, including spatial blurring, intensity scaling differences, and signal-dependent noise. During training, reconstructed volumes should match high-quality reference scans and reproduce observed fast acquisition data after propagation through the forward model, imposing a physics-guided consistency constraint. Experiments on unseen micro-CT datasets demonstrate improved reconstruction fidelity and enhanced visibility of fine microstructural features relative to conventional interpolation and learning-based enhancement approaches. The framework supports quantitative interpretation of fast 4D X-ray micro-CT scans of evolving materials.

cs.CV↗

arXiv2025

Learning Latent Hardening (LLH): Enhancing Deep Learning with Domain Knowledge for Material Inverse Problems

Advancements in deep learning and machine learning have improved the ability to model complex, nonlinear relationships, such as those encountered in complex material inverse problems. However, the effectiveness of these methods often depends on large datasets, which are not always available. In this study, the incorporation of domain-specific knowledge of the mechanical behavior of material microstructures is investigated to evaluate the impact on the predictive performance of the models in data-scarce scenarios. To overcome data limitations, a two-step framework, Learning Latent Hardening (LLH), is proposed. In the first step of LLH, a Deep Neural Network is employed to reconstruct full stress-strain curves from randomly selected portions of the stress-strain curves to capture the latent mechanical response of a material based on key microstructural features. In the second step of LLH, the results of the reconstructed stress-strain curves are leveraged to predict key microstructural features of porous materials. The performance of six deep learning and/or machine learning models trained with and without domain knowledge are compared: Convolutional Neural Networks, Deep Neural Networks, Extreme Gradient Boosting, K-Nearest Neighbors, Long Short-Term Memory, and Random Forest. The results from the models with domain-specific information consistently achieved higher $R^2$ values compared to models without prior knowledge. Models without domain knowledge missed critical patterns linking stress-strain behavior to microstructural changes, whereas domain-informed models better identified essential stress-strain features predictive of microstructure. These findings highlight the importance of integrating domain-specific knowledge with deep learning to achieve accurate outcomes in materials science.

cs.LG↗