AHG-YOLO: multi-category detection for occluded pear fruits in complex orchard scenes

IntroductionTo achieve fast detection of pear fruits in natural pear orchards and optimize path planning for harvesting robots, this study proposes the AHG-YOLO model for multi-category detection of pear fruit occlusion in complex orchard environments.MethodsUsing the Red Delicious pear as the resea...

Full description

Saved in:
Bibliographic Details
Main Authors: Na Ma, Yile Sun, Chenfei Li, Zonglin Liu, Haiyan Song
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-05-01
Series:Frontiers in Plant Science
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fpls.2025.1580325/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850153309581606912
author Na Ma
Na Ma
Na Ma
Yile Sun
Yile Sun
Chenfei Li
Chenfei Li
Zonglin Liu
Zonglin Liu
Haiyan Song
Haiyan Song
author_facet Na Ma
Na Ma
Na Ma
Yile Sun
Yile Sun
Chenfei Li
Chenfei Li
Zonglin Liu
Zonglin Liu
Haiyan Song
Haiyan Song
author_sort Na Ma
collection DOAJ
description IntroductionTo achieve fast detection of pear fruits in natural pear orchards and optimize path planning for harvesting robots, this study proposes the AHG-YOLO model for multi-category detection of pear fruit occlusion in complex orchard environments.MethodsUsing the Red Delicious pear as the research object, the pears are classified into three categories based on different occlusion statuses: non-occluded fruits (NO), fruits occluded by leaves/branches (OBL), and fruits in close contact with other fruits but not obstructed by leaves/branches (FCC). The YOLOv11n model is used as the base model for a lightweight design. First, the sampling method in the backbone and neck networks is replaced with ADown downsampling to capture higher-level image features, reducing floating-point operations and computational complexity. Next, shared weight parameters are introduced in the head network, and group convolution is applied to achieve a lightweight detection head. Finally, the boundary box loss function is changed to Generalized Intersection over Union (GIoU), improving the model’s convergence speed and further enhancing detection performance.ResultsExperimental results show that the AHG-YOLO model achieves 93.5% (FCC), 95.3% (NO), and 93.4% (OBL) in AP, with an mAP@0.5 of 94.1% across all categories. Compared to the base YOLOv11n network, precision, recall, mAP@0.5, and mAP@0.5:0.95 are improved by 2.5%, 3.6%, 2.3%, and 2.6%, respectively. The model size is only 5.1MB, with a 16.9% reduction in the number of parameters.DiscussionThe improved model demonstrates enhanced suitability for deployment on pear-harvesting embedded devices, providing technical support for the path planning of fruit-picking robotic arms.
format Article
id doaj-art-2ce3933ebc484badab681db02b203164
institution OA Journals
issn 1664-462X
language English
publishDate 2025-05-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Plant Science
spelling doaj-art-2ce3933ebc484badab681db02b2031642025-08-20T02:25:45ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2025-05-011610.3389/fpls.2025.15803251580325AHG-YOLO: multi-category detection for occluded pear fruits in complex orchard scenesNa Ma0Na Ma1Na Ma2Yile Sun3Yile Sun4Chenfei Li5Chenfei Li6Zonglin Liu7Zonglin Liu8Haiyan Song9Haiyan Song10College of Agricultural Engineering, Shanxi Agricultural University, Jinzhong, ChinaDryland Farm Machinery Key Technology and Equipment Key Laboratory of Shanxi Province, Jinzhong, ChinaCollege of Information Science and Engineering, Shanxi Agricultural University, Jinzhong, ChinaCollege of Agricultural Engineering, Shanxi Agricultural University, Jinzhong, ChinaDryland Farm Machinery Key Technology and Equipment Key Laboratory of Shanxi Province, Jinzhong, ChinaCollege of Agricultural Engineering, Shanxi Agricultural University, Jinzhong, ChinaDryland Farm Machinery Key Technology and Equipment Key Laboratory of Shanxi Province, Jinzhong, ChinaCollege of Agricultural Engineering, Shanxi Agricultural University, Jinzhong, ChinaDryland Farm Machinery Key Technology and Equipment Key Laboratory of Shanxi Province, Jinzhong, ChinaCollege of Agricultural Engineering, Shanxi Agricultural University, Jinzhong, ChinaDryland Farm Machinery Key Technology and Equipment Key Laboratory of Shanxi Province, Jinzhong, ChinaIntroductionTo achieve fast detection of pear fruits in natural pear orchards and optimize path planning for harvesting robots, this study proposes the AHG-YOLO model for multi-category detection of pear fruit occlusion in complex orchard environments.MethodsUsing the Red Delicious pear as the research object, the pears are classified into three categories based on different occlusion statuses: non-occluded fruits (NO), fruits occluded by leaves/branches (OBL), and fruits in close contact with other fruits but not obstructed by leaves/branches (FCC). The YOLOv11n model is used as the base model for a lightweight design. First, the sampling method in the backbone and neck networks is replaced with ADown downsampling to capture higher-level image features, reducing floating-point operations and computational complexity. Next, shared weight parameters are introduced in the head network, and group convolution is applied to achieve a lightweight detection head. Finally, the boundary box loss function is changed to Generalized Intersection over Union (GIoU), improving the model’s convergence speed and further enhancing detection performance.ResultsExperimental results show that the AHG-YOLO model achieves 93.5% (FCC), 95.3% (NO), and 93.4% (OBL) in AP, with an mAP@0.5 of 94.1% across all categories. Compared to the base YOLOv11n network, precision, recall, mAP@0.5, and mAP@0.5:0.95 are improved by 2.5%, 3.6%, 2.3%, and 2.6%, respectively. The model size is only 5.1MB, with a 16.9% reduction in the number of parameters.DiscussionThe improved model demonstrates enhanced suitability for deployment on pear-harvesting embedded devices, providing technical support for the path planning of fruit-picking robotic arms.https://www.frontiersin.org/articles/10.3389/fpls.2025.1580325/fullYOLOv11pear fruitsobject detectionADowngroup convolutionGIoU
spellingShingle Na Ma
Na Ma
Na Ma
Yile Sun
Yile Sun
Chenfei Li
Chenfei Li
Zonglin Liu
Zonglin Liu
Haiyan Song
Haiyan Song
AHG-YOLO: multi-category detection for occluded pear fruits in complex orchard scenes
Frontiers in Plant Science
YOLOv11
pear fruits
object detection
ADown
group convolution
GIoU
title AHG-YOLO: multi-category detection for occluded pear fruits in complex orchard scenes
title_full AHG-YOLO: multi-category detection for occluded pear fruits in complex orchard scenes
title_fullStr AHG-YOLO: multi-category detection for occluded pear fruits in complex orchard scenes
title_full_unstemmed AHG-YOLO: multi-category detection for occluded pear fruits in complex orchard scenes
title_short AHG-YOLO: multi-category detection for occluded pear fruits in complex orchard scenes
title_sort ahg yolo multi category detection for occluded pear fruits in complex orchard scenes
topic YOLOv11
pear fruits
object detection
ADown
group convolution
GIoU
url https://www.frontiersin.org/articles/10.3389/fpls.2025.1580325/full
work_keys_str_mv AT nama ahgyolomulticategorydetectionforoccludedpearfruitsincomplexorchardscenes
AT nama ahgyolomulticategorydetectionforoccludedpearfruitsincomplexorchardscenes
AT nama ahgyolomulticategorydetectionforoccludedpearfruitsincomplexorchardscenes
AT yilesun ahgyolomulticategorydetectionforoccludedpearfruitsincomplexorchardscenes
AT yilesun ahgyolomulticategorydetectionforoccludedpearfruitsincomplexorchardscenes
AT chenfeili ahgyolomulticategorydetectionforoccludedpearfruitsincomplexorchardscenes
AT chenfeili ahgyolomulticategorydetectionforoccludedpearfruitsincomplexorchardscenes
AT zonglinliu ahgyolomulticategorydetectionforoccludedpearfruitsincomplexorchardscenes
AT zonglinliu ahgyolomulticategorydetectionforoccludedpearfruitsincomplexorchardscenes
AT haiyansong ahgyolomulticategorydetectionforoccludedpearfruitsincomplexorchardscenes
AT haiyansong ahgyolomulticategorydetectionforoccludedpearfruitsincomplexorchardscenes