LiteMS-YOLO: a lightweight framework for small target detection in complex wheat field environments
Hongliang Ma, Maoxing Song, Mengying Yang, Tong Li, Xuefei Wang, Mochu Wang, Xiaoxue Zhao, Cheng Peng, Huina Huang, Peng Wu, Qing Lu, Zhihui Wu · 2026
Read LiteMS-YOLO: a lightweight framework for small target detection in complex wheat field environments by Hongliang Ma, Maoxing Song, Mengying Yang, Tong Li, Xuefei Wang, Mochu Wang, Xiaoxue Zhao, Cheng Peng, Huina Huang, Peng Wu, Qing Lu, Zhihui Wu (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-13279762.txt.
Description
Hongliang Ma and colleagues published a 2026 paper presenting LiteMS-YOLO, a featherweight detection model for counting wheat spikes in field photographs. Spike counts feed yield estimates, yet the targets are small, densely packed, and photographed against messy backgrounds. Built on a nano-scale detection architecture, the model adds a module that lets spatial and semantic features complement each other. A multi-kernel unit reads features at several scales, while deliberate redundancy cuts shrink its size.
Training used a combined set of 6,378 high-resolution images holding over 44,000 annotated spikes, drawn from a public global dataset plus 100 local field images. The model scored 92.28 percent on the standard detection metric while using only 0.627 million parameters. That is roughly three quarters fewer than the two nano models it was compared with, at competitive accuracy. The authors aim the design at real-time use on modest hardware in the field, where heavier models cannot run.
Catalog record
| CREATOR | Hongliang Ma, Maoxing Song, Mengying Yang, Tong Li, Xuefei Wang, Mochu Wang, Xiaoxue Zhao, Cheng Peng, Huina Huang, Peng Wu, Qing Lu, Zhihui Wu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 49 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |