YOLOPears: a novel benchmark of YOLO object detectors for multi-class pear surface defect detection in quality grading systems

Pears are one of the most widely consumed fruits, and their quality directly impacts consumer satisfaction. Surface defects, such as black spots and minor blemishes, are crucial indicators of pear quality, but it is still challenging to detect them due to the similarity in visual features. This stud...

Full description

Saved in:
Bibliographic Details
Main Authors: Junsheng Chen, Haoxuan Fu, Chuhan Lin, Xian Liu, Lijin Wang, Yaohua Lin
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-02-01
Series:Frontiers in Plant Science
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fpls.2025.1483824/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850067128384749568
author Junsheng Chen
Junsheng Chen
Haoxuan Fu
Haoxuan Fu
Chuhan Lin
Chuhan Lin
Xian Liu
Lijin Wang
Lijin Wang
Yaohua Lin
Yaohua Lin
author_facet Junsheng Chen
Junsheng Chen
Haoxuan Fu
Haoxuan Fu
Chuhan Lin
Chuhan Lin
Xian Liu
Lijin Wang
Lijin Wang
Yaohua Lin
Yaohua Lin
author_sort Junsheng Chen
collection DOAJ
description Pears are one of the most widely consumed fruits, and their quality directly impacts consumer satisfaction. Surface defects, such as black spots and minor blemishes, are crucial indicators of pear quality, but it is still challenging to detect them due to the similarity in visual features. This study presents PearSurfaceDefects, a self-constructed dataset, containing 13,915 images across six categories, with 66,189 bounding box annotations. These images were captured using a custom-built image acquisition platform. A comprehensive novel benchmark of 27 state-of-the-art YOLO object detectors of seven versions Scaled-YOLOv4, YOLOR, YOLOv5, YOLOv6, YOLOv7, YOLOv8, and YOLOv9,has been established on the dataset. To further ensure the comprehensiveness of the evaluation, three advanced non YOLO object detection models, T-DETR, RT-DERTV2, and D-FINE, were also included. Through experiments, it was found that the detection accuracy of YOLOv4-P7 at mAP@0.5 reached 73.20%, and YOLOv5n and YOLOv6n also show great potential for real-time pear surface defect detection, and data augmentation can further improve the accuracy of pear surface defect detection. The pear surface defect detection dataset and software program code for model benchmarking in this study are both public, which will not only promote future research on pear surface defect detection and grading, but also provide valuable resources and reference for other fruit big data and similar research.
format Article
id doaj-art-b42c7670baeb49e4baf558e33e5cb5a9
institution DOAJ
issn 1664-462X
language English
publishDate 2025-02-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Plant Science
spelling doaj-art-b42c7670baeb49e4baf558e33e5cb5a92025-08-20T02:48:29ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2025-02-011610.3389/fpls.2025.14838241483824YOLOPears: a novel benchmark of YOLO object detectors for multi-class pear surface defect detection in quality grading systemsJunsheng Chen0Junsheng Chen1Haoxuan Fu2Haoxuan Fu3Chuhan Lin4Chuhan Lin5Xian Liu6Lijin Wang7Lijin Wang8Yaohua Lin9Yaohua Lin10Fujian Agriculture and Forestry University, Fuzhou, ChinaKey Laboratory of Smart Agriculture and Forestry (Fujian Agriculture and Forestry University), Fujian Province University, Fuzhou, ChinaFujian Agriculture and Forestry University, Fuzhou, ChinaKey Laboratory of Smart Agriculture and Forestry (Fujian Agriculture and Forestry University), Fujian Province University, Fuzhou, ChinaFujian Agriculture and Forestry University, Fuzhou, ChinaKey Laboratory of Smart Agriculture and Forestry (Fujian Agriculture and Forestry University), Fujian Province University, Fuzhou, ChinaDigital Agriculture Research Institute, Fujian Academy of Agricultural Sciences, Fuzhou, Fujian, ChinaFujian Agriculture and Forestry University, Fuzhou, ChinaKey Laboratory of Smart Agriculture and Forestry (Fujian Agriculture and Forestry University), Fujian Province University, Fuzhou, ChinaFujian Agriculture and Forestry University, Fuzhou, ChinaKey Laboratory of Smart Agriculture and Forestry (Fujian Agriculture and Forestry University), Fujian Province University, Fuzhou, ChinaPears are one of the most widely consumed fruits, and their quality directly impacts consumer satisfaction. Surface defects, such as black spots and minor blemishes, are crucial indicators of pear quality, but it is still challenging to detect them due to the similarity in visual features. This study presents PearSurfaceDefects, a self-constructed dataset, containing 13,915 images across six categories, with 66,189 bounding box annotations. These images were captured using a custom-built image acquisition platform. A comprehensive novel benchmark of 27 state-of-the-art YOLO object detectors of seven versions Scaled-YOLOv4, YOLOR, YOLOv5, YOLOv6, YOLOv7, YOLOv8, and YOLOv9,has been established on the dataset. To further ensure the comprehensiveness of the evaluation, three advanced non YOLO object detection models, T-DETR, RT-DERTV2, and D-FINE, were also included. Through experiments, it was found that the detection accuracy of YOLOv4-P7 at mAP@0.5 reached 73.20%, and YOLOv5n and YOLOv6n also show great potential for real-time pear surface defect detection, and data augmentation can further improve the accuracy of pear surface defect detection. The pear surface defect detection dataset and software program code for model benchmarking in this study are both public, which will not only promote future research on pear surface defect detection and grading, but also provide valuable resources and reference for other fruit big data and similar research.https://www.frontiersin.org/articles/10.3389/fpls.2025.1483824/fullpeardatasetpear surface defect detectionsmart agriculturedeep learningcomputer vision
spellingShingle Junsheng Chen
Junsheng Chen
Haoxuan Fu
Haoxuan Fu
Chuhan Lin
Chuhan Lin
Xian Liu
Lijin Wang
Lijin Wang
Yaohua Lin
Yaohua Lin
YOLOPears: a novel benchmark of YOLO object detectors for multi-class pear surface defect detection in quality grading systems
Frontiers in Plant Science
pear
dataset
pear surface defect detection
smart agriculture
deep learning
computer vision
title YOLOPears: a novel benchmark of YOLO object detectors for multi-class pear surface defect detection in quality grading systems
title_full YOLOPears: a novel benchmark of YOLO object detectors for multi-class pear surface defect detection in quality grading systems
title_fullStr YOLOPears: a novel benchmark of YOLO object detectors for multi-class pear surface defect detection in quality grading systems
title_full_unstemmed YOLOPears: a novel benchmark of YOLO object detectors for multi-class pear surface defect detection in quality grading systems
title_short YOLOPears: a novel benchmark of YOLO object detectors for multi-class pear surface defect detection in quality grading systems
title_sort yolopears a novel benchmark of yolo object detectors for multi class pear surface defect detection in quality grading systems
topic pear
dataset
pear surface defect detection
smart agriculture
deep learning
computer vision
url https://www.frontiersin.org/articles/10.3389/fpls.2025.1483824/full
work_keys_str_mv AT junshengchen yolopearsanovelbenchmarkofyoloobjectdetectorsformulticlasspearsurfacedefectdetectioninqualitygradingsystems
AT junshengchen yolopearsanovelbenchmarkofyoloobjectdetectorsformulticlasspearsurfacedefectdetectioninqualitygradingsystems
AT haoxuanfu yolopearsanovelbenchmarkofyoloobjectdetectorsformulticlasspearsurfacedefectdetectioninqualitygradingsystems
AT haoxuanfu yolopearsanovelbenchmarkofyoloobjectdetectorsformulticlasspearsurfacedefectdetectioninqualitygradingsystems
AT chuhanlin yolopearsanovelbenchmarkofyoloobjectdetectorsformulticlasspearsurfacedefectdetectioninqualitygradingsystems
AT chuhanlin yolopearsanovelbenchmarkofyoloobjectdetectorsformulticlasspearsurfacedefectdetectioninqualitygradingsystems
AT xianliu yolopearsanovelbenchmarkofyoloobjectdetectorsformulticlasspearsurfacedefectdetectioninqualitygradingsystems
AT lijinwang yolopearsanovelbenchmarkofyoloobjectdetectorsformulticlasspearsurfacedefectdetectioninqualitygradingsystems
AT lijinwang yolopearsanovelbenchmarkofyoloobjectdetectorsformulticlasspearsurfacedefectdetectioninqualitygradingsystems
AT yaohualin yolopearsanovelbenchmarkofyoloobjectdetectorsformulticlasspearsurfacedefectdetectioninqualitygradingsystems
AT yaohualin yolopearsanovelbenchmarkofyoloobjectdetectorsformulticlasspearsurfacedefectdetectioninqualitygradingsystems