notesum.ai
Published at November 29Enhancing AI microscopy for foodborne bacterial classification via adversarial domain adaptation across optical and biological variability
eess.IV
cs.CV
cs.LG
Released Date: November 29, 2024
Authors: Siddhartha Bhattacharya1, Aarham Wasit1, Mason Earles2, Nitin Nitin2, Luyao Ma3, Jiyoon Yi4
Aff.: 1Department of Computer Science and Engineering, Michigan State University, East Lansing, MI, USA; 2Department of Biological and Agricultural Engineering, University of California, Davis, CA, USA; 3Department of Food Science and Technology, University of California, Davis, CA, USA; 4Department of Biosystems and Agricultural Engineering, Michigan State University, East Lansing, MI, USA

| Source-only | 1-shot DANN | 3-shot DANN | 5-shot DANN | ||
|---|---|---|---|---|---|
| BF | Source | 94.44% | 91.11% | 94.44% | 93.33% |
| Target | 43.33% | 54.44% | 75% | 73.33% | |
| 20× | Source | 94.44% | 90.0% | 90.0% | 93.33% |
| Target | 34.44% | 54.44% | 82.22% | 88.89% | |
| 20×–5h | Source | 94.44% | 93.33% | 94.44% | 94.44% |
| Target | 40.00% | 71.67% | 83.33% | 83.33% | |