A Robust Tomato Counting Framework for Greenhouse Inspection Robots Using YOLOv8 and Inter-Frame Prediction

Accurate tomato yield estimation and ripeness monitoring are critical for optimizing greenhouse management. While manual counting remains labor-intensive and error-prone, this study introduces a novel vision-based framework for automated tomato counting in standardized greenhouse environments. The p...

Full description

Saved in:
Bibliographic Details
Main Authors: Wanli Zheng, Guanglin Dai, Miao Hu, Pengbo Wang
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Agronomy
Subjects:
Online Access:https://www.mdpi.com/2073-4395/15/5/1135
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850254720239665152
author Wanli Zheng
Guanglin Dai
Miao Hu
Pengbo Wang
author_facet Wanli Zheng
Guanglin Dai
Miao Hu
Pengbo Wang
author_sort Wanli Zheng
collection DOAJ
description Accurate tomato yield estimation and ripeness monitoring are critical for optimizing greenhouse management. While manual counting remains labor-intensive and error-prone, this study introduces a novel vision-based framework for automated tomato counting in standardized greenhouse environments. The proposed method integrates YOLOv8-based detection, depth filtering, and an inter-frame prediction algorithm to address key challenges such as background interference, occlusion, and double-counting. Our approach achieves 97.09% accuracy in tomato cluster detection, with mature and immature single fruit recognition accuracies of 92.03% and 91.79%, respectively. The multi-target tracking algorithm demonstrates a MOTA (Multiple Object Tracking Accuracy) of 0.954, outperforming conventional methods like YOLOv8 + DeepSORT. By fusing odometry data from an inspection robot, this lightweight solution enables real-time yield estimation and maturity classification, offering practical value for precision agriculture.
format Article
id doaj-art-60c86e1ae1ff4317bd8eab9122a293f5
institution OA Journals
issn 2073-4395
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Agronomy
spelling doaj-art-60c86e1ae1ff4317bd8eab9122a293f52025-08-20T01:57:04ZengMDPI AGAgronomy2073-43952025-05-01155113510.3390/agronomy15051135A Robust Tomato Counting Framework for Greenhouse Inspection Robots Using YOLOv8 and Inter-Frame PredictionWanli Zheng0Guanglin Dai1Miao Hu2Pengbo Wang3Jiangsu Key Laboratory of Embodied Intelligent Robot Technology, College of Mechanical And Electrical Engineering, Soochow University, Suzhou 215123, ChinaJiangsu Key Laboratory of Embodied Intelligent Robot Technology, College of Mechanical And Electrical Engineering, Soochow University, Suzhou 215123, ChinaJiangsu Key Laboratory of Embodied Intelligent Robot Technology, College of Mechanical And Electrical Engineering, Soochow University, Suzhou 215123, ChinaJiangsu Key Laboratory of Embodied Intelligent Robot Technology, College of Mechanical And Electrical Engineering, Soochow University, Suzhou 215123, ChinaAccurate tomato yield estimation and ripeness monitoring are critical for optimizing greenhouse management. While manual counting remains labor-intensive and error-prone, this study introduces a novel vision-based framework for automated tomato counting in standardized greenhouse environments. The proposed method integrates YOLOv8-based detection, depth filtering, and an inter-frame prediction algorithm to address key challenges such as background interference, occlusion, and double-counting. Our approach achieves 97.09% accuracy in tomato cluster detection, with mature and immature single fruit recognition accuracies of 92.03% and 91.79%, respectively. The multi-target tracking algorithm demonstrates a MOTA (Multiple Object Tracking Accuracy) of 0.954, outperforming conventional methods like YOLOv8 + DeepSORT. By fusing odometry data from an inspection robot, this lightweight solution enables real-time yield estimation and maturity classification, offering practical value for precision agriculture.https://www.mdpi.com/2073-4395/15/5/1135agricultural roboticscomputer visionmultiple objects trackingtomato countingyield estimation
spellingShingle Wanli Zheng
Guanglin Dai
Miao Hu
Pengbo Wang
A Robust Tomato Counting Framework for Greenhouse Inspection Robots Using YOLOv8 and Inter-Frame Prediction
Agronomy
agricultural robotics
computer vision
multiple objects tracking
tomato counting
yield estimation
title A Robust Tomato Counting Framework for Greenhouse Inspection Robots Using YOLOv8 and Inter-Frame Prediction
title_full A Robust Tomato Counting Framework for Greenhouse Inspection Robots Using YOLOv8 and Inter-Frame Prediction
title_fullStr A Robust Tomato Counting Framework for Greenhouse Inspection Robots Using YOLOv8 and Inter-Frame Prediction
title_full_unstemmed A Robust Tomato Counting Framework for Greenhouse Inspection Robots Using YOLOv8 and Inter-Frame Prediction
title_short A Robust Tomato Counting Framework for Greenhouse Inspection Robots Using YOLOv8 and Inter-Frame Prediction
title_sort robust tomato counting framework for greenhouse inspection robots using yolov8 and inter frame prediction
topic agricultural robotics
computer vision
multiple objects tracking
tomato counting
yield estimation
url https://www.mdpi.com/2073-4395/15/5/1135
work_keys_str_mv AT wanlizheng arobusttomatocountingframeworkforgreenhouseinspectionrobotsusingyolov8andinterframeprediction
AT guanglindai arobusttomatocountingframeworkforgreenhouseinspectionrobotsusingyolov8andinterframeprediction
AT miaohu arobusttomatocountingframeworkforgreenhouseinspectionrobotsusingyolov8andinterframeprediction
AT pengbowang arobusttomatocountingframeworkforgreenhouseinspectionrobotsusingyolov8andinterframeprediction
AT wanlizheng robusttomatocountingframeworkforgreenhouseinspectionrobotsusingyolov8andinterframeprediction
AT guanglindai robusttomatocountingframeworkforgreenhouseinspectionrobotsusingyolov8andinterframeprediction
AT miaohu robusttomatocountingframeworkforgreenhouseinspectionrobotsusingyolov8andinterframeprediction
AT pengbowang robusttomatocountingframeworkforgreenhouseinspectionrobotsusingyolov8andinterframeprediction