An Improved YOLOv8 Model for Detecting Four Stages of Tomato Ripening and Its Application Deployment in a Greenhouse Environment

The ripeness of tomatoes is a critical factor influencing both their quality and yield. Currently, the accurate and efficient detection of tomato ripeness in greenhouse environments, along with the implementation of selective harvesting, has become a topic of significant research interest. In respon...

Full description

Saved in:
Bibliographic Details
Main Authors: Haoran Sun, Qi Zheng, Weixiang Yao, Junyong Wang, Changliang Liu, Huiduo Yu, Chunling Chen
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Agriculture
Subjects:
Online Access:https://www.mdpi.com/2077-0472/15/9/936
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850030570559504384
author Haoran Sun
Qi Zheng
Weixiang Yao
Junyong Wang
Changliang Liu
Huiduo Yu
Chunling Chen
author_facet Haoran Sun
Qi Zheng
Weixiang Yao
Junyong Wang
Changliang Liu
Huiduo Yu
Chunling Chen
author_sort Haoran Sun
collection DOAJ
description The ripeness of tomatoes is a critical factor influencing both their quality and yield. Currently, the accurate and efficient detection of tomato ripeness in greenhouse environments, along with the implementation of selective harvesting, has become a topic of significant research interest. In response to the current challenges, including the unclear segmentation of tomato ripeness stages, low recognition accuracy, and the limited deployment of mobile applications, this study provided a detailed classification of tomato ripeness stages. Through image processing techniques, the issue of class imbalance was addressed. Based on this, a model named GCSS-YOLO was proposed. Feature extraction was refined by introducing the RepNCSPELAN module, which is a lightweight alternative that reduces model size. A multi-dimensional feature neck network was integrated to enhance feature fusion, and three Semantic Feature Learning modules (SGE) were added before the detection head to minimize environmental interference. Further, Shape_IoU replaced CIoU as the loss function, prioritizing bounding box shape and size for improved detection accuracy. Experiments demonstrated GCSS-YOLO’s superiority, achieving an average mean average precision mAP50 of 85.3% and F1 score of 82.4%, outperforming the SSD, RT-DETR, and YOLO variants and advanced models like YOLO-TGI and SAG-YOLO. For practical deployment, this study deployed a mobile application developed using the NCNN framework on the Android platform. Upon evaluation, the model achieved an RMSE of 0.9045, an MAE of 0.4545, and an R<sup>2</sup> value of 0.9426, indicating strong performance.
format Article
id doaj-art-62a8b85e03214a368f867314cf7a0b7b
institution DOAJ
issn 2077-0472
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Agriculture
spelling doaj-art-62a8b85e03214a368f867314cf7a0b7b2025-08-20T02:59:11ZengMDPI AGAgriculture2077-04722025-04-0115993610.3390/agriculture15090936An Improved YOLOv8 Model for Detecting Four Stages of Tomato Ripening and Its Application Deployment in a Greenhouse EnvironmentHaoran Sun0Qi Zheng1Weixiang Yao2Junyong Wang3Changliang Liu4Huiduo Yu5Chunling Chen6College of Information and Electrical Engineering, Shenyang Agricultural University, Shenyang 110866, ChinaCollege of Information and Electrical Engineering, Shenyang Agricultural University, Shenyang 110866, ChinaCollege of Information and Electrical Engineering, Shenyang Agricultural University, Shenyang 110866, ChinaCollege of Information and Electrical Engineering, Shenyang Agricultural University, Shenyang 110866, ChinaCollege of Information and Electrical Engineering, Shenyang Agricultural University, Shenyang 110866, ChinaCollege of Information and Electrical Engineering, Shenyang Agricultural University, Shenyang 110866, ChinaCollege of Information and Electrical Engineering, Shenyang Agricultural University, Shenyang 110866, ChinaThe ripeness of tomatoes is a critical factor influencing both their quality and yield. Currently, the accurate and efficient detection of tomato ripeness in greenhouse environments, along with the implementation of selective harvesting, has become a topic of significant research interest. In response to the current challenges, including the unclear segmentation of tomato ripeness stages, low recognition accuracy, and the limited deployment of mobile applications, this study provided a detailed classification of tomato ripeness stages. Through image processing techniques, the issue of class imbalance was addressed. Based on this, a model named GCSS-YOLO was proposed. Feature extraction was refined by introducing the RepNCSPELAN module, which is a lightweight alternative that reduces model size. A multi-dimensional feature neck network was integrated to enhance feature fusion, and three Semantic Feature Learning modules (SGE) were added before the detection head to minimize environmental interference. Further, Shape_IoU replaced CIoU as the loss function, prioritizing bounding box shape and size for improved detection accuracy. Experiments demonstrated GCSS-YOLO’s superiority, achieving an average mean average precision mAP50 of 85.3% and F1 score of 82.4%, outperforming the SSD, RT-DETR, and YOLO variants and advanced models like YOLO-TGI and SAG-YOLO. For practical deployment, this study deployed a mobile application developed using the NCNN framework on the Android platform. Upon evaluation, the model achieved an RMSE of 0.9045, an MAE of 0.4545, and an R<sup>2</sup> value of 0.9426, indicating strong performance.https://www.mdpi.com/2077-0472/15/9/936tomato detectionripenessYOLOv8sandroid deploymentgreenhouse environment
spellingShingle Haoran Sun
Qi Zheng
Weixiang Yao
Junyong Wang
Changliang Liu
Huiduo Yu
Chunling Chen
An Improved YOLOv8 Model for Detecting Four Stages of Tomato Ripening and Its Application Deployment in a Greenhouse Environment
Agriculture
tomato detection
ripeness
YOLOv8s
android deployment
greenhouse environment
title An Improved YOLOv8 Model for Detecting Four Stages of Tomato Ripening and Its Application Deployment in a Greenhouse Environment
title_full An Improved YOLOv8 Model for Detecting Four Stages of Tomato Ripening and Its Application Deployment in a Greenhouse Environment
title_fullStr An Improved YOLOv8 Model for Detecting Four Stages of Tomato Ripening and Its Application Deployment in a Greenhouse Environment
title_full_unstemmed An Improved YOLOv8 Model for Detecting Four Stages of Tomato Ripening and Its Application Deployment in a Greenhouse Environment
title_short An Improved YOLOv8 Model for Detecting Four Stages of Tomato Ripening and Its Application Deployment in a Greenhouse Environment
title_sort improved yolov8 model for detecting four stages of tomato ripening and its application deployment in a greenhouse environment
topic tomato detection
ripeness
YOLOv8s
android deployment
greenhouse environment
url https://www.mdpi.com/2077-0472/15/9/936
work_keys_str_mv AT haoransun animprovedyolov8modelfordetectingfourstagesoftomatoripeninganditsapplicationdeploymentinagreenhouseenvironment
AT qizheng animprovedyolov8modelfordetectingfourstagesoftomatoripeninganditsapplicationdeploymentinagreenhouseenvironment
AT weixiangyao animprovedyolov8modelfordetectingfourstagesoftomatoripeninganditsapplicationdeploymentinagreenhouseenvironment
AT junyongwang animprovedyolov8modelfordetectingfourstagesoftomatoripeninganditsapplicationdeploymentinagreenhouseenvironment
AT changliangliu animprovedyolov8modelfordetectingfourstagesoftomatoripeninganditsapplicationdeploymentinagreenhouseenvironment
AT huiduoyu animprovedyolov8modelfordetectingfourstagesoftomatoripeninganditsapplicationdeploymentinagreenhouseenvironment
AT chunlingchen animprovedyolov8modelfordetectingfourstagesoftomatoripeninganditsapplicationdeploymentinagreenhouseenvironment
AT haoransun improvedyolov8modelfordetectingfourstagesoftomatoripeninganditsapplicationdeploymentinagreenhouseenvironment
AT qizheng improvedyolov8modelfordetectingfourstagesoftomatoripeninganditsapplicationdeploymentinagreenhouseenvironment
AT weixiangyao improvedyolov8modelfordetectingfourstagesoftomatoripeninganditsapplicationdeploymentinagreenhouseenvironment
AT junyongwang improvedyolov8modelfordetectingfourstagesoftomatoripeninganditsapplicationdeploymentinagreenhouseenvironment
AT changliangliu improvedyolov8modelfordetectingfourstagesoftomatoripeninganditsapplicationdeploymentinagreenhouseenvironment
AT huiduoyu improvedyolov8modelfordetectingfourstagesoftomatoripeninganditsapplicationdeploymentinagreenhouseenvironment
AT chunlingchen improvedyolov8modelfordetectingfourstagesoftomatoripeninganditsapplicationdeploymentinagreenhouseenvironment