Conference Logo ProgramMaster Logo
Conference Tools for 2026 TMS Annual Meeting & Exhibition
Login
Register as a New User
Help
Submit An Abstract
Propose A Symposium
Presenter/Author Tools
Organizer/Editor Tools

About this Abstract

Meeting 2026 TMS Annual Meeting & Exhibition
Symposium Novel Strategies for Rapid Acquisition and Processing of Large Datasets From Advanced Characterization Techniques
Presentation Title Enabling Data Starved Microstructural Segmentation With Foundation Models as First-Pass Segmentors in Low-Contrast Al-Si Solidification
Author(s) Nicholas Amano, Aramanda Kiran, Ashwin Shahani, Elizabeth A Holm
On-Site Speaker (Planned) Nicholas Amano
Abstract Scope Microstructural image segmentation is a critical tool in materials science, enabling quantitative analysis of phase behavior and solidification evolution, keys in understanding material properties. Image foundation models enable the extraction of valuable information from common object images but are often limited in their capture of microstructural information. While segmentations produced by models like meta’s SAM 2 are often insufficient for comprehensive microstructural analysis, a limited number of accurate frames can be leveraged for further model training. The training frames that capture representative features of the underlying structures enable the training of supervised models from a small subset of reasonably segmented examples. In this study, we demonstrate that rough, unsupervised segmentations generated by SAM 2 can be used to bootstrap a strong supervised segmentation model. We apply this approach to detecting Al-Si solidification fronts in extremely low-contrast optical imaging conditions, showing significant improvements in segmentation performance and robustness.
Proceedings Inclusion? Planned:
Keywords Characterization, Computational Materials Science & Engineering, Solidification

OTHER PAPERS PLANNED FOR THIS SYMPOSIUM

A Topology Optimized Specimen that Provides the Yield Surface in a Single Test
Advancements in Titanium Microstructure Characterization: Integrating AI for Automated Complex Quantification
Advances in Analytical S/TEM-Based Workflows: Emerging Techniques for Smarter Data Acquisition and Processing
Advances in Instrumentation for Spatially Resolved Acoustic Spectroscopy (SRAS)
Advancing Segregation Characterization in Steel and Alloys: A Novel BEX-EDS-SEM Approach
Affine Transformations to Correlate Experimental and Simulated EDS Spectra for Multi-Element Systems
Automated 2D Microscopy Workflows for Multimodal Characterization of Structure, Crystallography and Composition
Comprehensive Microstructural Characterization of Aging-Treated Aluminum Alloy AA2024 Using Integrated SEM, STEM, EDS, and EBSD Workflows
Conditional Diffusion Models for Microscopy Modality Transfer of Electron Backscattering Microscopy Diffraction Misorientations Maps from Optical Microscopy
Discovering Hidden Fingerprints in Multimodal Process-Structure-Property Data via Joint Embedding
Emphasizing the Importance of Data Exchange in Constructing a Digital Twin for Metals-AM
Enabling Data Starved Microstructural Segmentation With Foundation Models as First-Pass Segmentors in Low-Contrast Al-Si Solidification
Examining Growth Twinning in Ni-Based Films via a High-Throughput Methodology
Frequency-Domain Thermoreflectance Automation for High Throughput Microstructural and Thermal Characterization
High-Throughput Crystallography by Quantitative Large-Area Polarized-Light Microscopy
High-Throughput Exploration of Large Material Design Spaces Using Small Samples and Bayesian Strategies
High-Throughput Processing and Accelerated Characterization of Cu–Ti Alloy
High-Throughput Time-Resolved X-Ray Computed Tomography to Characterize Flaw Evolution in LPBF Parts During Creep
High Throughput Texture Analysis of Quartz via Automated Polarized Reflective Light Microscopy
Increasing the Throughput of Ultra-High Temperature Ceramic Tensile Testing
Lightweight Services and Federated Storage for Data-Intensive Structural Materials Research
Linking Energetic Material Sensitivity and Microstructure Variability Across Length Scales
Machine Learning Assisted Structure-Property Relationships by Nanoindentation
Mapping Microstructure to Field Properties in Materials Under Dynamical Loading
Nanocrystalline Films: Imaging, Orientation Mapping, Machine Learning and Data Analytics
New Strategy of Surface Defect Detection in Metallic Coatings Using a Machine Learning-Based Model
Same-Day Product Quality Assessment of CaCl₂-Assisted Direct Reduction of Chromite via µCT and Deep-Learning Segmentation Against a Reference Dataset
Segmentation Methods for Tracking Dislocation Dynamics

Questions about ProgramMaster? Contact programming@programmaster.org | TMS Privacy Policy | Accessibility Statement