Related Experiment Video
Updated: May 23, 2026

Image Recognition and Parameter Analysis of Concrete Vibration State Based on Support Vector Machine
Published on: January 5, 2024
Against the grain: Leveraging machine learning to analyze mudbrick structures
Sofia Kouki1, Marta Lorenzon2,3, Benjamín Cutillas-Victoria4
1Interdisciplinary Center for Archaeology and the Evolution of Human Behavior, University of Algarve, Faro, Portugal.
Automated image analysis quantifies mudbrick fabric, revealing significant differences in grain size and shape between archaeological sites. This computational approach offers a reproducible method for characterizing ancient building materials.
Area of Science:
- Archaeological science
- Materials science
- Computational archaeology
Background:
- Mudbricks are ancient building materials with complex compositional variability.
- Traditional petrographic point counting for mudbrick fabric characterization is time-consuming and subjective.
- Quantitative grain morphometry offers insights into raw material selection and construction techniques.
Purpose of the Study:
- To evaluate automated image analysis for quantitative grain morphometry in archaeological mudbricks.
- To develop a computational workflow for fabric characterization.
- To compare morphometric differences between two distinct archaeological sites.
Main Methods:
- Developed a K-means clustering workflow for automated grain size, sorting, and shape analysis.
- Analyzed 45 cross-polarized light images from Artaxata (Armenia) and Los Villares de la Encarnación (Spain).
- Employed non-parametric statistics, multivariate ordination, and supervised classification for comparative analysis.
Main Results:
- Statistically significant differences in grain size, sorting, and shape were identified between the two sites.
- Multivariate analysis showed clear separation of samples based on morphometric data.
- Automated morphometry achieved high classification accuracy (>84%) and corresponded well with existing petrographic fabrics.
Conclusions:
- Automated grain morphometry effectively captures assemblage-level technological variability in mudbricks.
- The proposed computational workflow provides a foundation for quantitative and comparative studies in earthen architecture.
- This method supports the integration of computational tools into archaeological thin-section petrography for enhanced reproducibility.
More Related Videos
09:00Visualization of Failure and the Associated Grain-Scale Mechanical Behavior of Granular Soils under Shear using Synchrotron X-Ray Micro-Tomography
Published on: September 29, 2019
07:15Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
Related Concept Videos
Reinforced Brick Masonry
To fortify brick walls...
Microcracking in Concrete
Brick Masonry
For thicker walls, multiple wythes are bonded together using...
Brick Cutting Techniques
Cut bricks are categorized by size. Bricks cut to half their original length are called half-bats, while those cut to three-fourths their length are known as three-fourth bats.
Special types of cut...
Mass Analyzers: Overview
Quarrying of Stone
One common method involves using a diamond belt saw to cut large blocks from the quarry face. These blocks can be about 50 feet long and 12 feet high. After the initial vertical cut, drilling is performed at the base of the block.