A reproducible U-net workflow for SEM/BSE bright-particle quantification
Ibrahim M Ibrahim, Geoffrey Will, Stuart Bell · 2026
Read A reproducible U-net workflow for SEM/BSE bright-particle quantification by Ibrahim M Ibrahim, Geoffrey Will, Stuart Bell (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/pmc-13571523.txt.
Description
Heat-aged steels grow bright specks in microscope images, chromium-rich particles that quietly rot strength and rust resistance, but counting them by hand is slow and uneven. Ibrahim Ibrahim, Geoffrey Will, and Stuart Bell describe a reproducible U-net workflow in 2026 for turning those specks into numbers.
The pipeline preps the scanning electron images and builds reference masks that respect each heat condition. It then trains a small segmentation network on overlapping patches, and converts its masks into standard size and count measures. Tested on a duplex steel, scores measure agreement with the reference masks, a limit the authors state plainly: proving the particles' identity still takes other evidence. The value offered is consistency across labs.
Catalog record
| CREATOR | Ibrahim M Ibrahim, Geoffrey Will, Stuart Bell |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 64 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |