A Lightweight Citrus Object Detection Method in Complex Environments

Aiming at the limitations of current citrus detection methods in complex orchard environments, especially the problems of poor model adaptability and high computational complexity under different lighting, multiple occlusions, and dense fruit conditions, this study proposes an improved citrus detect...

Full description

Saved in:
Bibliographic Details
Main Authors: Qiurong Lv, Fuchun Sun, Yuechao Bian, Haorong Wu, Xiaoxiao Li, Xin Li, Jie Zhou
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Agriculture
Subjects:
Online Access:https://www.mdpi.com/2077-0472/15/10/1046
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849327673591988224
author Qiurong Lv
Fuchun Sun
Yuechao Bian
Haorong Wu
Xiaoxiao Li
Xin Li
Jie Zhou
author_facet Qiurong Lv
Fuchun Sun
Yuechao Bian
Haorong Wu
Xiaoxiao Li
Xin Li
Jie Zhou
author_sort Qiurong Lv
collection DOAJ
description Aiming at the limitations of current citrus detection methods in complex orchard environments, especially the problems of poor model adaptability and high computational complexity under different lighting, multiple occlusions, and dense fruit conditions, this study proposes an improved citrus detection model, YOLO-PBGM, based on You Only Look Once v7 (YOLOv7). First, to tackle the large size of the YOLOv7 network model and its deployment challenges, the PC-ELAN module is constructed by introducing Partial Convolution (PConv) for lightweight improvement, which reduces the model’s demand for computing resources and parameters. At the same time, the Bi-Former attention module is embedded to enhance the perception and processing of citrus fruit information. Secondly, a lightweight neck network is constructed using Grouped Shuffle Convolution (GSConv) to simplify computational complexity. Finally, the minimum-point-distance-based IoU (MPDIoU) loss function is utilized to optimize the boundary return mechanism, which speeds up model convergence and reduces the redundancy of bounding box regression. Experimental results indicate that for the citrus dataset collected in a natural environment, the improved model reduces Params and GFLOPs by 15.4% and 23.7%, respectively, while improving precision, recall, and mAP by 0.3%, 4%, and 3.5%, respectively, thereby outperforming other detection networks. Additionally, an analysis of citrus object detection under varying lighting and occlusion conditions reveals that the YOLO-PBGM network model demonstrates good adaptability, effectively coping with variations in lighting and occlusions while exhibiting high robustness. This model can provide a technical reference for uncrewed intelligent picking of citrus.
format Article
id doaj-art-e22eb86ed5044432bd1bc2d5947f685c
institution Kabale University
issn 2077-0472
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Agriculture
spelling doaj-art-e22eb86ed5044432bd1bc2d5947f685c2025-08-20T03:47:48ZengMDPI AGAgriculture2077-04722025-05-011510104610.3390/agriculture15101046A Lightweight Citrus Object Detection Method in Complex EnvironmentsQiurong Lv0Fuchun Sun1Yuechao Bian2Haorong Wu3Xiaoxiao Li4Xin Li5Jie Zhou6School of Mechanical Engineering, Chengdu University, Chengdu 610106, ChinaSchool of Mechanical Engineering, Chengdu University, Chengdu 610106, ChinaSchool of Mechanical Engineering, Chengdu University, Chengdu 610106, ChinaSchool of Electronic Information and Electrical Engineering, Chengdu University, Chengdu 610106, ChinaSchool of Mechanical Engineering, Chengdu University, Chengdu 610106, ChinaShengzhong Water Conservancy Project Operation and Management Center of Sichuan Province, Nanchong 623300, ChinaSchool of Mechanical Engineering, Chengdu University, Chengdu 610106, ChinaAiming at the limitations of current citrus detection methods in complex orchard environments, especially the problems of poor model adaptability and high computational complexity under different lighting, multiple occlusions, and dense fruit conditions, this study proposes an improved citrus detection model, YOLO-PBGM, based on You Only Look Once v7 (YOLOv7). First, to tackle the large size of the YOLOv7 network model and its deployment challenges, the PC-ELAN module is constructed by introducing Partial Convolution (PConv) for lightweight improvement, which reduces the model’s demand for computing resources and parameters. At the same time, the Bi-Former attention module is embedded to enhance the perception and processing of citrus fruit information. Secondly, a lightweight neck network is constructed using Grouped Shuffle Convolution (GSConv) to simplify computational complexity. Finally, the minimum-point-distance-based IoU (MPDIoU) loss function is utilized to optimize the boundary return mechanism, which speeds up model convergence and reduces the redundancy of bounding box regression. Experimental results indicate that for the citrus dataset collected in a natural environment, the improved model reduces Params and GFLOPs by 15.4% and 23.7%, respectively, while improving precision, recall, and mAP by 0.3%, 4%, and 3.5%, respectively, thereby outperforming other detection networks. Additionally, an analysis of citrus object detection under varying lighting and occlusion conditions reveals that the YOLO-PBGM network model demonstrates good adaptability, effectively coping with variations in lighting and occlusions while exhibiting high robustness. This model can provide a technical reference for uncrewed intelligent picking of citrus.https://www.mdpi.com/2077-0472/15/10/1046citrusattention mechanismmachine visionYOLOv7deep learning
spellingShingle Qiurong Lv
Fuchun Sun
Yuechao Bian
Haorong Wu
Xiaoxiao Li
Xin Li
Jie Zhou
A Lightweight Citrus Object Detection Method in Complex Environments
Agriculture
citrus
attention mechanism
machine vision
YOLOv7
deep learning
title A Lightweight Citrus Object Detection Method in Complex Environments
title_full A Lightweight Citrus Object Detection Method in Complex Environments
title_fullStr A Lightweight Citrus Object Detection Method in Complex Environments
title_full_unstemmed A Lightweight Citrus Object Detection Method in Complex Environments
title_short A Lightweight Citrus Object Detection Method in Complex Environments
title_sort lightweight citrus object detection method in complex environments
topic citrus
attention mechanism
machine vision
YOLOv7
deep learning
url https://www.mdpi.com/2077-0472/15/10/1046
work_keys_str_mv AT qiuronglv alightweightcitrusobjectdetectionmethodincomplexenvironments
AT fuchunsun alightweightcitrusobjectdetectionmethodincomplexenvironments
AT yuechaobian alightweightcitrusobjectdetectionmethodincomplexenvironments
AT haorongwu alightweightcitrusobjectdetectionmethodincomplexenvironments
AT xiaoxiaoli alightweightcitrusobjectdetectionmethodincomplexenvironments
AT xinli alightweightcitrusobjectdetectionmethodincomplexenvironments
AT jiezhou alightweightcitrusobjectdetectionmethodincomplexenvironments
AT qiuronglv lightweightcitrusobjectdetectionmethodincomplexenvironments
AT fuchunsun lightweightcitrusobjectdetectionmethodincomplexenvironments
AT yuechaobian lightweightcitrusobjectdetectionmethodincomplexenvironments
AT haorongwu lightweightcitrusobjectdetectionmethodincomplexenvironments
AT xiaoxiaoli lightweightcitrusobjectdetectionmethodincomplexenvironments
AT xinli lightweightcitrusobjectdetectionmethodincomplexenvironments
AT jiezhou lightweightcitrusobjectdetectionmethodincomplexenvironments