Ta-YOLO: overcoming target blocked challenges in greenhouse tomato detection and counting

Screening and cultivating healthy small tomatoes, along with accurately predicting their yields, are crucial for sustaining the economy of tomato industry. However, in field scenarios, counting small tomato fruits is often hindered by environmental factors such as leaf shading. To address this chall...

Full description

Saved in:
Bibliographic Details
Main Authors: Yun Zhao, Yijia Chen, Xing Xu, Yong He, Hao Gan, Na Wu, Zhechen Wang, Xi Sun, Yali Wang, Petr Skobelev, Yanan Mi
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-07-01
Series:Frontiers in Plant Science
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fpls.2025.1618214/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849428547159982080
author Yun Zhao
Yijia Chen
Xing Xu
Yong He
Hao Gan
Na Wu
Zhechen Wang
Xi Sun
Yali Wang
Petr Skobelev
Yanan Mi
author_facet Yun Zhao
Yijia Chen
Xing Xu
Yong He
Hao Gan
Na Wu
Zhechen Wang
Xi Sun
Yali Wang
Petr Skobelev
Yanan Mi
author_sort Yun Zhao
collection DOAJ
description Screening and cultivating healthy small tomatoes, along with accurately predicting their yields, are crucial for sustaining the economy of tomato industry. However, in field scenarios, counting small tomato fruits is often hindered by environmental factors such as leaf shading. To address this challenge, this study proposed the Ta-YOLO modeling framework, aimed at improving the efficiency and accuracy of small tomato fruit detection. We captured images of small tomatoes at various stages of ripeness in real-world settings and compiled them into datasets for training and testing the model. First, we utilized the Space-to-Depth module to efficiently leverage the implicit features of the images while ensuring a lightweight operation of the backbone network. Next, we developed a novel pyramid pooling module(DASPPF) to capture global information through average pooling, effectively reducing the impact of edge and background noise on detection. We also introduced an additional tiny target detection head alongside the original detection head, enabling multi-scale detection of small tomatoes. To further enhance the model’s focus on relevant information and improve its ability to recognize small targets, we designed a multi-dimensional attention structure(CSAM) that generated feature maps with more valuable information. Finally, we proposed the EWDIoU bounding box loss function, which leveraged a 2D Gaussian distribution to enhance the model’s accuracy and robustness. The experimental results showed that the number of parameters, FLOPs, and FPS of our designed Ta-YOLO were 10.58M, 14.4G, and 131.58, respectively, and its mean average precision(mAP) reached 84.4%. It can better realize the counting of tomatoes with different maturity levels, which helps to improve the efficiency of the small tomato production and planting process.
format Article
id doaj-art-5647bd4ac2ca467ca8a3b09391d0e08a
institution Kabale University
issn 1664-462X
language English
publishDate 2025-07-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Plant Science
spelling doaj-art-5647bd4ac2ca467ca8a3b09391d0e08a2025-08-20T03:28:40ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2025-07-011610.3389/fpls.2025.16182141618214Ta-YOLO: overcoming target blocked challenges in greenhouse tomato detection and countingYun Zhao0Yijia Chen1Xing Xu2Yong He3Hao Gan4Na Wu5Zhechen Wang6Xi Sun7Yali Wang8Petr Skobelev9Yanan Mi10School of Artificial Intelligence and Information Engineering, Zhejiang University of Science and Technology, Hangzhou, ChinaSchool of Artificial Intelligence and Information Engineering, Zhejiang University of Science and Technology, Hangzhou, ChinaSchool of Artificial Intelligence and Information Engineering, Zhejiang University of Science and Technology, Hangzhou, ChinaCollege of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou, ChinaDepartment of Biosystems Engineering and Soil Science, University of Tennessee, Knoxville, TN, United StatesSchool of Artificial Intelligence and Information Engineering, Zhejiang University of Science and Technology, Hangzhou, ChinaSchool of Artificial Intelligence and Information Engineering, Zhejiang University of Science and Technology, Hangzhou, ChinaSchool of Artificial Intelligence and Information Engineering, Zhejiang University of Science and Technology, Hangzhou, ChinaCardiovascular Medicine, Zhejiang Hospital, Hangzhou, ChinaSamara Federal Research Scientific Center, Russian Academy of Sciences, Samara, RussiaDepartment of Business Development, Pegasor Oy, Tampere, FinlandScreening and cultivating healthy small tomatoes, along with accurately predicting their yields, are crucial for sustaining the economy of tomato industry. However, in field scenarios, counting small tomato fruits is often hindered by environmental factors such as leaf shading. To address this challenge, this study proposed the Ta-YOLO modeling framework, aimed at improving the efficiency and accuracy of small tomato fruit detection. We captured images of small tomatoes at various stages of ripeness in real-world settings and compiled them into datasets for training and testing the model. First, we utilized the Space-to-Depth module to efficiently leverage the implicit features of the images while ensuring a lightweight operation of the backbone network. Next, we developed a novel pyramid pooling module(DASPPF) to capture global information through average pooling, effectively reducing the impact of edge and background noise on detection. We also introduced an additional tiny target detection head alongside the original detection head, enabling multi-scale detection of small tomatoes. To further enhance the model’s focus on relevant information and improve its ability to recognize small targets, we designed a multi-dimensional attention structure(CSAM) that generated feature maps with more valuable information. Finally, we proposed the EWDIoU bounding box loss function, which leveraged a 2D Gaussian distribution to enhance the model’s accuracy and robustness. The experimental results showed that the number of parameters, FLOPs, and FPS of our designed Ta-YOLO were 10.58M, 14.4G, and 131.58, respectively, and its mean average precision(mAP) reached 84.4%. It can better realize the counting of tomatoes with different maturity levels, which helps to improve the efficiency of the small tomato production and planting process.https://www.frontiersin.org/articles/10.3389/fpls.2025.1618214/fullmachine visionTa-YOLOtarget detectiontomato countingtarget blocked
spellingShingle Yun Zhao
Yijia Chen
Xing Xu
Yong He
Hao Gan
Na Wu
Zhechen Wang
Xi Sun
Yali Wang
Petr Skobelev
Yanan Mi
Ta-YOLO: overcoming target blocked challenges in greenhouse tomato detection and counting
Frontiers in Plant Science
machine vision
Ta-YOLO
target detection
tomato counting
target blocked
title Ta-YOLO: overcoming target blocked challenges in greenhouse tomato detection and counting
title_full Ta-YOLO: overcoming target blocked challenges in greenhouse tomato detection and counting
title_fullStr Ta-YOLO: overcoming target blocked challenges in greenhouse tomato detection and counting
title_full_unstemmed Ta-YOLO: overcoming target blocked challenges in greenhouse tomato detection and counting
title_short Ta-YOLO: overcoming target blocked challenges in greenhouse tomato detection and counting
title_sort ta yolo overcoming target blocked challenges in greenhouse tomato detection and counting
topic machine vision
Ta-YOLO
target detection
tomato counting
target blocked
url https://www.frontiersin.org/articles/10.3389/fpls.2025.1618214/full
work_keys_str_mv AT yunzhao tayoloovercomingtargetblockedchallengesingreenhousetomatodetectionandcounting
AT yijiachen tayoloovercomingtargetblockedchallengesingreenhousetomatodetectionandcounting
AT xingxu tayoloovercomingtargetblockedchallengesingreenhousetomatodetectionandcounting
AT yonghe tayoloovercomingtargetblockedchallengesingreenhousetomatodetectionandcounting
AT haogan tayoloovercomingtargetblockedchallengesingreenhousetomatodetectionandcounting
AT nawu tayoloovercomingtargetblockedchallengesingreenhousetomatodetectionandcounting
AT zhechenwang tayoloovercomingtargetblockedchallengesingreenhousetomatodetectionandcounting
AT xisun tayoloovercomingtargetblockedchallengesingreenhousetomatodetectionandcounting
AT yaliwang tayoloovercomingtargetblockedchallengesingreenhousetomatodetectionandcounting
AT petrskobelev tayoloovercomingtargetblockedchallengesingreenhousetomatodetectionandcounting
AT yananmi tayoloovercomingtargetblockedchallengesingreenhousetomatodetectionandcounting