Towards Resilient Agriculture: A Novel UAV-Based Lightweight Deep Learning Framework for Wheat Head Detection

Luo, Na; Yang, Yao; Yang, Xiwei; Yang, Di; Tang, Jiao; Duan, Siyuan; Huang, Hou; Zhu, He

doi:10.3390/math13233844

Open AccessArticle

Towards Resilient Agriculture: A Novel UAV-Based Lightweight Deep Learning Framework for Wheat Head Detection

by

Na Luo

¹,

Yao Yang

¹,

Xiwei Yang

¹,

Di Yang

¹,

Jiao Tang

¹,

Siyuan Duan

¹,

Hou Huang

¹ and

He Zhu

^2,*

¹

School of Economics and Management, Puer University, Puer 665000, China

²

Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, Chengdu 610213, China

^*

Author to whom correspondence should be addressed.

Mathematics 2025, 13(23), 3844; https://doi.org/10.3390/math13233844

Submission received: 13 October 2025 / Revised: 17 November 2025 / Accepted: 18 November 2025 / Published: 1 December 2025

(This article belongs to the Special Issue Optimization and Machine Learning-Based Methods in Air Traffic Management and Aeronautical Domains)

Download

Browse Figure

Versions Notes

Abstract

Precision agriculture increasingly relies on unmanned aerial vehicle (UAV) imagery for high-throughput crop phenotyping, yet existing deep learning detection models face critical constraints limiting practical deployment: computational demands incompatible with edge computing platforms and insufficient accuracy for multi-scale object detection across diverse environmental conditions. We present LSM-YOLO, a lightweight detection framework specifically designed for aerial wheat head monitoring that achieves state-of-the-art performance while maintaining minimal computational requirements. The architecture integrates three synergistic innovations: a Lightweight Adaptive Extraction (LAE) module that reduces parameters by 87.3% through efficient spatial rearrangement and adaptive feature weighting while preserving critical boundary information; a P2-level high-resolution detection head that substantially improves small object recall in high-altitude imagery; and a Dynamic Head mechanism employing unified multi-dimensional attention across scale, spatial, and task dimensions. Comprehensive evaluation on the Global Wheat Head Detection dataset demonstrates that LSM-YOLO achieves 91.4% mAP@0.5 and 51.0% mAP@0.5:0.95—representing 21.1% and 37.1% improvements over baseline YOLO11n—while requiring only 1.29 M parameters and 3.4 GFLOPs, constituting 50.0% parameter reduction and 46.0% computational cost reduction compared to the baseline.

Keywords: precision agriculture; wheat head detection; lightweight deep learning; UAV remote sensing; multi-scale object detection

Share and Cite

MDPI and ACS Style

Luo, N.; Yang, Y.; Yang, X.; Yang, D.; Tang, J.; Duan, S.; Huang, H.; Zhu, H. Towards Resilient Agriculture: A Novel UAV-Based Lightweight Deep Learning Framework for Wheat Head Detection. Mathematics 2025, 13, 3844. https://doi.org/10.3390/math13233844

AMA Style

Luo N, Yang Y, Yang X, Yang D, Tang J, Duan S, Huang H, Zhu H. Towards Resilient Agriculture: A Novel UAV-Based Lightweight Deep Learning Framework for Wheat Head Detection. Mathematics. 2025; 13(23):3844. https://doi.org/10.3390/math13233844

Chicago/Turabian Style

Luo, Na, Yao Yang, Xiwei Yang, Di Yang, Jiao Tang, Siyuan Duan, Hou Huang, and He Zhu. 2025. "Towards Resilient Agriculture: A Novel UAV-Based Lightweight Deep Learning Framework for Wheat Head Detection" Mathematics 13, no. 23: 3844. https://doi.org/10.3390/math13233844

APA Style

Luo, N., Yang, Y., Yang, X., Yang, D., Tang, J., Duan, S., Huang, H., & Zhu, H. (2025). Towards Resilient Agriculture: A Novel UAV-Based Lightweight Deep Learning Framework for Wheat Head Detection. Mathematics, 13(23), 3844. https://doi.org/10.3390/math13233844

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Towards Resilient Agriculture: A Novel UAV-Based Lightweight Deep Learning Framework for Wheat Head Detection

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI