CFPR-YOLO: chili flower pose estimation for robotic pollination in unstructured environments
Minqiu Kuang, Yushi Wang, Xiaojian Li, Dawei Liu, Yang Xiang, Feng Liu, Xiangjun Zou, Fangping Xie, Yuxuan Zhang, Xu Li · 2026
Read CFPR-YOLO: chili flower pose estimation for robotic pollination in unstructured environments by Minqiu Kuang, Yushi Wang, Xiaojian Li, Dawei Liu, Yang Xiang, Feng Liu, Xiangjun Zou, Fangping Xie, Yuxuan Zhang, Xu Li (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-13279318.txt.
Description
Minqiu Kuang and colleagues published a 2026 paper on a vision system that finds chili flowers and judges their pose so robots can pollinate them. Flowers are small, crowded, often hidden by leaves, and lit unevenly in greenhouses, which makes machine perception hard. Their framework, CFPR-YOLO, adapts a compact detection network with a stronger feature extractor and modules tuned for tiny and partly hidden targets.
On their own and outside test sets the system reached a precision of 92.6 percent and a detection score of 92.1 percent with only 7.26 million parameters, small enough for field hardware. Running on a Jetson edge computer it processed 39.5 frames per second, fast enough for a moving robot. In indoor trials it supported simulated pollination with a success rate of 90 percent on upward-facing flowers. The authors see it as a building block for precision pollination and fruit-set management.
Catalog record
| CREATOR | Minqiu Kuang, Yushi Wang, Xiaojian Li, Dawei Liu, Yang Xiang, Feng Liu, Xiangjun Zou, Fangping Xie, Yuxuan Zhang, Xu Li |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 81 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |