Automatic Detection and Classification of Natural Weld Defects Using Alternating Magneto-Optical Imaging and ResNet50
Yanfeng Li, Pengyu Gao, Yongbiao Luo, Xianghan Luo, Chunmei Xu, Jiecheng Chen, Yanxi Zhang, Genxiang Lin, Wei Xu · 2024
Read Automatic Detection and Classification of Natural Weld Defects Using Alternating Magneto-Optical Imaging and ResNet50 by Yanfeng Li, Pengyu Gao, Yongbiao Luo, Xianghan Luo, Chunmei Xu, Jiecheng Chen, Yanxi Zhang, Genxiang Lin, Wei Xu (2024) 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-11644933.txt.
Description
Li and eight co-authors present a study from twenty twenty-four on finding flaws in welds without cutting parts open. They use light-based imaging that works through a changing magnet field. The method takes pictures of cracks, pits, holes, and spots where the weld did not melt through. They clean the pictures with filters, then train smart computer models to sort each picture by flaw type.
The best model, called ResNet50, sorted the test pictures right ninety-nine times out of one hundred. It beat two older models and did much better on gas holes, a hard type to spot. The work gives shops a fast, safe way to check welds with no radiation risk.
Catalog record
| CREATOR | Yanfeng Li, Pengyu Gao, Yongbiao Luo, Xianghan Luo, Chunmei Xu, Jiecheng Chen, Yanxi Zhang, Genxiang Lin, Wei Xu |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 59 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |