Star-YOLO: A Lightweight Real-Time Wheat Grain Detection Model for Embedded Deployment

With the rapid advancement of precision agriculture, traditional object detection algorithms struggle with limited efficiency and accuracy in wheat grain detection and counting, while the need for real-time deployment of deep learning models on embedded devices becomes increasingly critical. To this...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhihang Qu, Xiao Liang, Sicheng Liang, Xiumei Guo
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11087564/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849421285254234112
author Zhihang Qu
Xiao Liang
Sicheng Liang
Xiumei Guo
author_facet Zhihang Qu
Xiao Liang
Sicheng Liang
Xiumei Guo
author_sort Zhihang Qu
collection DOAJ
description With the rapid advancement of precision agriculture, traditional object detection algorithms struggle with limited efficiency and accuracy in wheat grain detection and counting, while the need for real-time deployment of deep learning models on embedded devices becomes increasingly critical. To this end, this paper introduces Star-YOLO, a lightweight wheat grain detection model built upon YOLOv11n. The model employs StarNet to refine the C3k2 structure, reducing computational complexity without compromising detection accuracy, and integrates the MBConv module into the detection head to boost feature extraction while further minimizing computational load. A Shape-NWD loss function is designed, incorporating shape and scale information of target bounding boxes to refine regression, tackling the challenge of distinguishing overlapping wheat grains. Experimental results indicate that Star-YOLO attains a mean Average Precision (mAP) of 97.3% in wheat grain detection, outperforming models like YOLOv5n and RT-DETR in both accuracy and efficiency, with a reduced parameter count. Tests on embedded devices reveal a 36.8% performance enhancement over the baseline model, fulfilling the demands of real-time detection. The Star-YOLO model markedly enhances both the accuracy and efficiency of wheat grain detection while underscoring the potential of deep learning in agricultural image analysis and smart wheat industry advancements.
format Article
id doaj-art-18340fdbcbbf4baa94999c87d4f36f41
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-18340fdbcbbf4baa94999c87d4f36f412025-08-20T03:31:30ZengIEEEIEEE Access2169-35362025-01-011313014713015810.1109/ACCESS.2025.359137411087564Star-YOLO: A Lightweight Real-Time Wheat Grain Detection Model for Embedded DeploymentZhihang Qu0https://orcid.org/0009-0004-0701-586XXiao Liang1Sicheng Liang2Xiumei Guo3https://orcid.org/0000-0002-4953-5554School of Information Science and Engineering, Shandong Agricultural University, Taian, ChinaSchool of Information Science and Engineering, Shandong Agricultural University, Taian, ChinaSchool of Information Science and Engineering, Shandong Agricultural University, Taian, ChinaSchool of Information Science and Engineering, Shandong Agricultural University, Taian, ChinaWith the rapid advancement of precision agriculture, traditional object detection algorithms struggle with limited efficiency and accuracy in wheat grain detection and counting, while the need for real-time deployment of deep learning models on embedded devices becomes increasingly critical. To this end, this paper introduces Star-YOLO, a lightweight wheat grain detection model built upon YOLOv11n. The model employs StarNet to refine the C3k2 structure, reducing computational complexity without compromising detection accuracy, and integrates the MBConv module into the detection head to boost feature extraction while further minimizing computational load. A Shape-NWD loss function is designed, incorporating shape and scale information of target bounding boxes to refine regression, tackling the challenge of distinguishing overlapping wheat grains. Experimental results indicate that Star-YOLO attains a mean Average Precision (mAP) of 97.3% in wheat grain detection, outperforming models like YOLOv5n and RT-DETR in both accuracy and efficiency, with a reduced parameter count. Tests on embedded devices reveal a 36.8% performance enhancement over the baseline model, fulfilling the demands of real-time detection. The Star-YOLO model markedly enhances both the accuracy and efficiency of wheat grain detection while underscoring the potential of deep learning in agricultural image analysis and smart wheat industry advancements.https://ieeexplore.ieee.org/document/11087564/Wheat grainsYOLOv11lightweight architectureshape-NWDreal-time detection
spellingShingle Zhihang Qu
Xiao Liang
Sicheng Liang
Xiumei Guo
Star-YOLO: A Lightweight Real-Time Wheat Grain Detection Model for Embedded Deployment
IEEE Access
Wheat grains
YOLOv11
lightweight architecture
shape-NWD
real-time detection
title Star-YOLO: A Lightweight Real-Time Wheat Grain Detection Model for Embedded Deployment
title_full Star-YOLO: A Lightweight Real-Time Wheat Grain Detection Model for Embedded Deployment
title_fullStr Star-YOLO: A Lightweight Real-Time Wheat Grain Detection Model for Embedded Deployment
title_full_unstemmed Star-YOLO: A Lightweight Real-Time Wheat Grain Detection Model for Embedded Deployment
title_short Star-YOLO: A Lightweight Real-Time Wheat Grain Detection Model for Embedded Deployment
title_sort star yolo a lightweight real time wheat grain detection model for embedded deployment
topic Wheat grains
YOLOv11
lightweight architecture
shape-NWD
real-time detection
url https://ieeexplore.ieee.org/document/11087564/
work_keys_str_mv AT zhihangqu staryoloalightweightrealtimewheatgraindetectionmodelforembeddeddeployment
AT xiaoliang staryoloalightweightrealtimewheatgraindetectionmodelforembeddeddeployment
AT sichengliang staryoloalightweightrealtimewheatgraindetectionmodelforembeddeddeployment
AT xiumeiguo staryoloalightweightrealtimewheatgraindetectionmodelforembeddeddeployment