YOLOv8s-Longan: a lightweight detection method for the longan fruit-picking UAV
Jun Li, Kaixuan Wu, Meiqi Zhang, Hengxu Chen, Hengyi Lin, Yuju Mai, Linlin Shi · 2025
Read YOLOv8s-Longan: a lightweight detection method for the longan fruit-picking UAV by Jun Li, Kaixuan Wu, Meiqi Zhang, Hengxu Chen, Hengyi Lin, Yuju Mai, Linlin Shi (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-11794187.txt.
Description
Jun Li and co-authors design a lightweight vision model that helps picking drones spot longan fruit. Picking drones fly fast and carry weak computers, so the detector must be both quick and accurate. The team improves the YOLOv8s model in several ways. A new attention module blends average and max pooling to sharpen focus while keeping the network small. A redesigned structure fuses features at several scales, and a new loss function guides the placement of detection boxes.
In tests, the model finds densely packed and partly hidden longan clusters against complex backgrounds, scoring 84.3 percent on a standard accuracy measure. Against other YOLOv8 versions, accuracy rises by 3.9 points while the parameter count falls by about a fifth. The study appeared in 2024. The authors conclude the design meets the speed and accuracy needs of fruit-picking drones.
Catalog record
| CREATOR | Jun Li, Kaixuan Wu, Meiqi Zhang, Hengxu Chen, Hengyi Lin, Yuju Mai, Linlin Shi |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 59 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |