AITP-YOLO: improved tomato ripeness detection model based on multiple strategies

IntroductionThis paper offers a multi-scale AITP-YOLO model, based on the enhanced YOLOv10s model, to address the challenges of difficult identification and frequent misdetection of tomatoes, facilitating ripeness detection under realistic conditions.MethodsA four-head detector incorporates a small...

Full description

Saved in:
Bibliographic Details
Main Authors: Wenyuan Huang, Yiran Liao, Peiling Wang, Ziao Chen, Ziqi Yang, Lijia Xu, Jiong Mu
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.2025.1596739/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849731833131958272
author Wenyuan Huang
Wenyuan Huang
Yiran Liao
Peiling Wang
Ziao Chen
Ziao Chen
Ziqi Yang
Ziqi Yang
Lijia Xu
Jiong Mu
Jiong Mu
author_facet Wenyuan Huang
Wenyuan Huang
Yiran Liao
Peiling Wang
Ziao Chen
Ziao Chen
Ziqi Yang
Ziqi Yang
Lijia Xu
Jiong Mu
Jiong Mu
author_sort Wenyuan Huang
collection DOAJ
description IntroductionThis paper offers a multi-scale AITP-YOLO model, based on the enhanced YOLOv10s model, to address the challenges of difficult identification and frequent misdetection of tomatoes, facilitating ripeness detection under realistic conditions.MethodsA four-head detector incorporates a small target detection layer, enhancing the model's capacity to identify small targets. Secondly, a multi-scale feature fusion technique employing cross-level features is implemented in the feature fusion layer to amalgamate convolutions of varying sizes, enhancing the model's fusion capacity and generalization proficiency for features of diverse scales. The bounding box loss function is modified to Shape-IoU, with the loss computed by emphasizing the shape and scale of the bounding box, hence enhancing the precision of bounding box regression, expediting model convergence, and augmenting model correctness. Ultimately, the model is compressed via Network Slimming puring,which removes redundant channels while mataining detection accuracy.ResultsThe experimental findings indicate that the enhanced model achieves average precision, accuracy, and recall of 92.6%, 89.7%, and 87.4%, respectively. In comparison to the baseline network YOLOv10s, the model weights are compressed by 7.64%, while average precision, accuracy, and recall are elevated by 4.6%, 5.8%, and 7.3%, respectively.DiscussionThe enhanced model features a reduced model size while exhibiting superior detection capabilities, enabling more efficient and precise recognition of tomato stages amidst complicated backgrounds, hence offering a valuable technical reference for automated tomato harvesting technology.
format Article
id doaj-art-e4bd931b41224e53828fc486a150741d
institution DOAJ
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-e4bd931b41224e53828fc486a150741d2025-08-20T03:08:25ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2025-05-011610.3389/fpls.2025.15967391596739AITP-YOLO: improved tomato ripeness detection model based on multiple strategiesWenyuan Huang0Wenyuan Huang1Yiran Liao2Peiling Wang3Ziao Chen4Ziao Chen5Ziqi Yang6Ziqi Yang7Lijia Xu8Jiong Mu9Jiong Mu10College of Information Engineering, Sichuan Agricultural University, Yaan, ChinaSichuan Key Laboratory of Agricultural Information Engineering, Yaan, ChinaCollege of Mechanical and Electrical Engineering, Sichuan Agricultural University, Yaan, ChinaCollege of Information Engineering, Sichuan Agricultural University, Yaan, ChinaSichuan Key Laboratory of Agricultural Information Engineering, Yaan, ChinaCountry College of Law, Sichuan Agricultural University, Yaan, ChinaCollege of Information Engineering, Sichuan Agricultural University, Yaan, ChinaSichuan Key Laboratory of Agricultural Information Engineering, Yaan, ChinaCollege of Mechanical and Electrical Engineering, Sichuan Agricultural University, Yaan, ChinaCollege of Information Engineering, Sichuan Agricultural University, Yaan, ChinaSichuan Key Laboratory of Agricultural Information Engineering, Yaan, ChinaIntroductionThis paper offers a multi-scale AITP-YOLO model, based on the enhanced YOLOv10s model, to address the challenges of difficult identification and frequent misdetection of tomatoes, facilitating ripeness detection under realistic conditions.MethodsA four-head detector incorporates a small target detection layer, enhancing the model's capacity to identify small targets. Secondly, a multi-scale feature fusion technique employing cross-level features is implemented in the feature fusion layer to amalgamate convolutions of varying sizes, enhancing the model's fusion capacity and generalization proficiency for features of diverse scales. The bounding box loss function is modified to Shape-IoU, with the loss computed by emphasizing the shape and scale of the bounding box, hence enhancing the precision of bounding box regression, expediting model convergence, and augmenting model correctness. Ultimately, the model is compressed via Network Slimming puring,which removes redundant channels while mataining detection accuracy.ResultsThe experimental findings indicate that the enhanced model achieves average precision, accuracy, and recall of 92.6%, 89.7%, and 87.4%, respectively. In comparison to the baseline network YOLOv10s, the model weights are compressed by 7.64%, while average precision, accuracy, and recall are elevated by 4.6%, 5.8%, and 7.3%, respectively.DiscussionThe enhanced model features a reduced model size while exhibiting superior detection capabilities, enabling more efficient and precise recognition of tomato stages amidst complicated backgrounds, hence offering a valuable technical reference for automated tomato harvesting technology.https://www.frontiersin.org/articles/10.3389/fpls.2025.1596739/fulltarget detectionimage recognitionYOLOv10small target detection headmulti-scaletomato
spellingShingle Wenyuan Huang
Wenyuan Huang
Yiran Liao
Peiling Wang
Ziao Chen
Ziao Chen
Ziqi Yang
Ziqi Yang
Lijia Xu
Jiong Mu
Jiong Mu
AITP-YOLO: improved tomato ripeness detection model based on multiple strategies
Frontiers in Plant Science
target detection
image recognition
YOLOv10
small target detection head
multi-scale
tomato
title AITP-YOLO: improved tomato ripeness detection model based on multiple strategies
title_full AITP-YOLO: improved tomato ripeness detection model based on multiple strategies
title_fullStr AITP-YOLO: improved tomato ripeness detection model based on multiple strategies
title_full_unstemmed AITP-YOLO: improved tomato ripeness detection model based on multiple strategies
title_short AITP-YOLO: improved tomato ripeness detection model based on multiple strategies
title_sort aitp yolo improved tomato ripeness detection model based on multiple strategies
topic target detection
image recognition
YOLOv10
small target detection head
multi-scale
tomato
url https://www.frontiersin.org/articles/10.3389/fpls.2025.1596739/full
work_keys_str_mv AT wenyuanhuang aitpyoloimprovedtomatoripenessdetectionmodelbasedonmultiplestrategies
AT wenyuanhuang aitpyoloimprovedtomatoripenessdetectionmodelbasedonmultiplestrategies
AT yiranliao aitpyoloimprovedtomatoripenessdetectionmodelbasedonmultiplestrategies
AT peilingwang aitpyoloimprovedtomatoripenessdetectionmodelbasedonmultiplestrategies
AT ziaochen aitpyoloimprovedtomatoripenessdetectionmodelbasedonmultiplestrategies
AT ziaochen aitpyoloimprovedtomatoripenessdetectionmodelbasedonmultiplestrategies
AT ziqiyang aitpyoloimprovedtomatoripenessdetectionmodelbasedonmultiplestrategies
AT ziqiyang aitpyoloimprovedtomatoripenessdetectionmodelbasedonmultiplestrategies
AT lijiaxu aitpyoloimprovedtomatoripenessdetectionmodelbasedonmultiplestrategies
AT jiongmu aitpyoloimprovedtomatoripenessdetectionmodelbasedonmultiplestrategies
AT jiongmu aitpyoloimprovedtomatoripenessdetectionmodelbasedonmultiplestrategies