Effects of transfer learning on seedling detection from drone imagery: a layer freezing study with faster R-CNN
Hyungsik Jeong, Joungwon You, Dayoung Kim, Peter Sang-Hoon Lee, Hee Han · 2026
Read Effects of transfer learning on seedling detection from drone imagery: a layer freezing study with faster R-CNN by Hyungsik Jeong, Joungwon You, Dayoung Kim, Peter Sang-Hoon Lee, Hee Han (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-13558572.txt.
Description
Hyungsik Jeong and co-authors test drone photos plus deep learning to find young trees after replanting, in a 2026 study. Forest managers need fast counts of new seedlings, but ground surveys cover little ground. Photos were taken over a two-year-old larch plantation, and seedling spots were checked on foot.
A standard detection model was trained in several ways, mixing stored prior learning with frozen or open layers. The best setup froze early layers, tuned the rest, and started from prior weights. It beat training from scratch on the test score. Extra steps such as image tweaks and repeated splits helped steady the results. The authors say the approach can cut survey cost while giving managers timely data.
Catalog record
| CREATOR | Hyungsik Jeong, Joungwon You, Dayoung Kim, Peter Sang-Hoon Lee, Hee Han |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 59 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |