Predicting image processing-normalized DL-EPRT performance from Rudimentary Cihal method-based counterpart
Conventional DL-EPRT was extrapolated into accounting for inherent inconsistencies amongst grain boundary (GB) dimensions whilst correlating two distinct methods of ir/ia normalization, namely the approximate Cihal method (CM) and the relatively involved image processing (IP). Inasmuch as facilitati...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-12-01
|
| Series: | Nuclear Materials and Energy |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2352179124001558 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Conventional DL-EPRT was extrapolated into accounting for inherent inconsistencies amongst grain boundary (GB) dimensions whilst correlating two distinct methods of ir/ia normalization, namely the approximate Cihal method (CM) and the relatively involved image processing (IP). Inasmuch as facilitating the forecast of IP- from CM-based normalizations, the full spectrum of plausible predictive models was subjected to cross validations that revealed those emanating from GB length-based endeavors to be quadratic with R2 and mean error (E) of 0.9215 and 0.5769, respectively. Meanwhile, GB area normalizations succumbed to a linear polynomial yielding R2 and E values of 0.8795 and 0.3664, respectively. |
|---|---|
| ISSN: | 2352-1791 |