Pixel level anomaly detection in second harmonic generation microscopy for non destructive defect inspection of AlGaN/GaN heterostructures
Chen-Fang Kang, Chih-Chang Hung, Che-Hsuan Huang, Ting-Chuan Li, Wu-Yih Uen, Ching-Hsueh Chiu · 2026
Read Pixel level anomaly detection in second harmonic generation microscopy for non destructive defect inspection of AlGaN/GaN heterostructures by Chen-Fang Kang, Chih-Chang Hung, Che-Hsuan Huang, Ting-Chuan Li, Wu-Yih Uen, Ching-Hsueh Chiu (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/PMC13260511.txt.
Description
Chen-Fang Kang and co-authors present a 2026 study of a non-destructive checking method for semiconductor structures made of AlGaN and GaN, materials used in high-power electronics. Their framework combines second harmonic generation microscopy, an optical method sensitive to broken crystal symmetry, with pixel-level anomaly detection. Unlike object detectors that draw boxes around defects, their method models the normal lattice pattern with statistics, without hand-labeled examples, then flags pixels that deviate.
The approach was compared with standard detectors such as YOLOv8 and Mask R-CNN. It showed far better spatial overlap and sensitivity, resolving more than five thousand micron-scale defect-affected regions within a single field of view. Many of those regions were largely out of reach for box- or mask-based methods. Optical defect features were also supported by correlation with light-emission based defect measures.
Catalog record
| CREATOR | Chen-Fang Kang, Chih-Chang Hung, Che-Hsuan Huang, Ting-Chuan Li, Wu-Yih Uen, Ching-Hsueh Chiu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 56487 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |