Against the grain: Leveraging machine learning to analyze mudbrick structures
Sofia Kouki, Marta Lorenzon, Benjamín Cutillas-Victoria, Achim Lichtenberger, Torben Schreiber, Mkrtich Zardaryan, Anno Hein, Georgios Polymeris · 2026
Read Against the grain: Leveraging machine learning to analyze mudbrick structures by Sofia Kouki, Marta Lorenzon, Benjamín Cutillas-Victoria, Achim Lichtenberger, Torben Schreiber, Mkrtich Zardaryan, Anno Hein, Georgios Polymeris (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0349295.txt.
Description
Sofia Kouki and colleagues tested whether machine-assisted image analysis can describe the grains inside archaeological mudbricks as reliably as a human expert. Mudbrick has been a building material since the first farming villages, and grain size and shape reveal how builders chose and prepared their mud. The standard expert method, counting points under a microscope, is slow and varies between analysts.
Their workflow uses a clustering algorithm on microscope images to extract grain size, sorting, and shape automatically. They applied it to forty-five images from two distant sites: Artaxata in Armenia and Los Villares in Spain. Faster, repeatable measurement would let archaeologists compare building traditions across regions without years of microscope work by specialists.
Catalog record
| CREATOR | Sofia Kouki, Marta Lorenzon, Benjamín Cutillas-Victoria, Achim Lichtenberger, Torben Schreiber, Mkrtich Zardaryan, Anno Hein, Georgios Polymeris |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 54494 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |