MDE-DETR: multi-domain enhanced feature fusion algorithm for bayberry detection and counting in complex orchards
Cheng Zhou, Yuyu Zhang, Wei Fu, Lili Yao, Chengliang Yin · 2025
Read MDE-DETR: multi-domain enhanced feature fusion algorithm for bayberry detection and counting in complex orchards by Cheng Zhou, Yuyu Zhang, Wei Fu, Lili Yao, Chengliang Yin (2025) 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-12695788.txt.
Description
Cheng Zhou, Yuyu Zhang, Wei Fu, Lili Yao, and Chengliang Yin build a vision system that finds and counts bayberries in orchard photos. Counting fruit early helps growers predict yield, but bayberries are small, packed close together, and often hidden behind leaves, which defeats simple detectors.
Their system, called MDE-DETR, sharpens how features are pulled from an image, blends detail across scales, and adjusts where it samples so hidden fruit still register. On their own orchard dataset it beat the base model it grew from by clear margins while using about a quarter fewer parameters and less memory. Tests on drone photos and tomato plant images suggest the gains carry over to other small, crowded targets.
Catalog record
| CREATOR | Cheng Zhou, Yuyu Zhang, Wei Fu, Lili Yao, Chengliang Yin |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 60 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |