Tomato detection in natural environment based on improved YOLOv8 network
In this paper, an improved lightweight YOLOv8 method is proposed to detect the ripeness of tomato fruits, given the problems of subtle differences between neighboring stages of ripening and mutual occlusion of branches, leaves, and fruits. The method replaces the backbone network of the original YO...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
PAGEPress Publications
2025-07-01
|
| Series: | Journal of Agricultural Engineering |
| Subjects: | |
| Online Access: | https://www.agroengineering.org/jae/article/view/1732 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839582398486937600 |
|---|---|
| author | Wancheng Dong Yipeng Zhao Jiaxing Pei Zuolong Feng Zhikai Ma Leilei Wang Simon Shemin Wang |
| author_facet | Wancheng Dong Yipeng Zhao Jiaxing Pei Zuolong Feng Zhikai Ma Leilei Wang Simon Shemin Wang |
| author_sort | Wancheng Dong |
| collection | DOAJ |
| description |
In this paper, an improved lightweight YOLOv8 method is proposed to detect the ripeness of tomato fruits, given the problems of subtle differences between neighboring stages of ripening and mutual occlusion of branches, leaves, and fruits. The method replaces the backbone network of the original YOLOv8 with a more lightweight MobileNetV3 structure to reduce the number of parameters of the model; at the same time, it integrates the convolutional attention mechanism module (CBAM) in the feature extraction network, which enhances the network's capability of extracting features of tomato fruits. At the same time, it introduces the SCYLLA-IoU (SIoU) as a bounded YOLOv8 frame regression loss function, effectively solving the mismatch problem between the predicted frame and the actual frame and improving recognition accuracy. Compared with the current mainstream models Resnet50, VGG16, YOLOv3, YOLOv5, YOLOv7, etc., the model is in an advantageous position regarding precision rate, recall rate, and detection accuracy. The research and experimental results show that the mean values of precision, recall rate, and average precision of the improved MCS-YOLOv8 model under the test set are 91.2%, 90.2%, and 90.3%, respectively. The detection speed of a single image is 5.4ms, and the model occupies less memory by 8.7 M. The model has a clear advantage in both detection speed and precision rate and also shows that the improved MCS-YOLOv8 model can provide strong technical support for tomato-picking robots in complex environments in the field.
|
| format | Article |
| id | doaj-art-ee9bb511f76b4f2c84b3b46ccb915a0a |
| institution | Matheson Library |
| issn | 1974-7071 2239-6268 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | PAGEPress Publications |
| record_format | Article |
| series | Journal of Agricultural Engineering |
| spelling | doaj-art-ee9bb511f76b4f2c84b3b46ccb915a0a2025-08-04T06:16:24ZengPAGEPress PublicationsJournal of Agricultural Engineering1974-70712239-62682025-07-0110.4081/jae.2025.1732Tomato detection in natural environment based on improved YOLOv8 networkWancheng Dong0Yipeng Zhao1Jiaxing Pei2Zuolong Feng3Zhikai Ma4Leilei Wang5Simon Shemin Wang6School of Mechanical and Equipment Engineering, Hebei University of Engineering, HandanSchool of Mechanical and Equipment Engineering, Hebei University of Engineering, HandanSchool of Mechanical and Equipment Engineering, Hebei University of Engineering, HandanHebei Provincial Agricultural Mechanization Technology Promotion Station, LangfangCollege of Mechanical and Electrical Engineering, Hebei Agricultural University, HandanSchool of Mechanical and Equipment Engineering, Hebei University of Engineering, HandanSchool of Mechanical and Equipment Engineering, Hebei University of Engineering, Handan In this paper, an improved lightweight YOLOv8 method is proposed to detect the ripeness of tomato fruits, given the problems of subtle differences between neighboring stages of ripening and mutual occlusion of branches, leaves, and fruits. The method replaces the backbone network of the original YOLOv8 with a more lightweight MobileNetV3 structure to reduce the number of parameters of the model; at the same time, it integrates the convolutional attention mechanism module (CBAM) in the feature extraction network, which enhances the network's capability of extracting features of tomato fruits. At the same time, it introduces the SCYLLA-IoU (SIoU) as a bounded YOLOv8 frame regression loss function, effectively solving the mismatch problem between the predicted frame and the actual frame and improving recognition accuracy. Compared with the current mainstream models Resnet50, VGG16, YOLOv3, YOLOv5, YOLOv7, etc., the model is in an advantageous position regarding precision rate, recall rate, and detection accuracy. The research and experimental results show that the mean values of precision, recall rate, and average precision of the improved MCS-YOLOv8 model under the test set are 91.2%, 90.2%, and 90.3%, respectively. The detection speed of a single image is 5.4ms, and the model occupies less memory by 8.7 M. The model has a clear advantage in both detection speed and precision rate and also shows that the improved MCS-YOLOv8 model can provide strong technical support for tomato-picking robots in complex environments in the field. https://www.agroengineering.org/jae/article/view/1732CBAMloss functionMobileNetV3ripeningtomatoYOLOv8 |
| spellingShingle | Wancheng Dong Yipeng Zhao Jiaxing Pei Zuolong Feng Zhikai Ma Leilei Wang Simon Shemin Wang Tomato detection in natural environment based on improved YOLOv8 network Journal of Agricultural Engineering CBAM loss function MobileNetV3 ripening tomato YOLOv8 |
| title | Tomato detection in natural environment based on improved YOLOv8 network |
| title_full | Tomato detection in natural environment based on improved YOLOv8 network |
| title_fullStr | Tomato detection in natural environment based on improved YOLOv8 network |
| title_full_unstemmed | Tomato detection in natural environment based on improved YOLOv8 network |
| title_short | Tomato detection in natural environment based on improved YOLOv8 network |
| title_sort | tomato detection in natural environment based on improved yolov8 network |
| topic | CBAM loss function MobileNetV3 ripening tomato YOLOv8 |
| url | https://www.agroengineering.org/jae/article/view/1732 |
| work_keys_str_mv | AT wanchengdong tomatodetectioninnaturalenvironmentbasedonimprovedyolov8network AT yipengzhao tomatodetectioninnaturalenvironmentbasedonimprovedyolov8network AT jiaxingpei tomatodetectioninnaturalenvironmentbasedonimprovedyolov8network AT zuolongfeng tomatodetectioninnaturalenvironmentbasedonimprovedyolov8network AT zhikaima tomatodetectioninnaturalenvironmentbasedonimprovedyolov8network AT leileiwang tomatodetectioninnaturalenvironmentbasedonimprovedyolov8network AT simonsheminwang tomatodetectioninnaturalenvironmentbasedonimprovedyolov8network |