Fresh Tea Leaf-Grading Detection: An Improved YOLOv8 Neural Network Model Utilizing Deep Learning
To facilitate the realization of automated tea picking and enhance the speed and accuracy of tea leaf grading detection, this study proposes an improved YOLOv8 network for fresh tea leaf grading recognition. This approach integrates a Hierarchical Vision Transformer using Shifted Windows to replace...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Horticulturae |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2311-7524/10/12/1347 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850242097765941248 |
|---|---|
| author | Zejun Wang Yuxin Xia Houqiao Wang Xiaohui Liu Raoqiong Che Xiaoxue Guo Hongxu Li Shihao Zhang Baijuan Wang |
| author_facet | Zejun Wang Yuxin Xia Houqiao Wang Xiaohui Liu Raoqiong Che Xiaoxue Guo Hongxu Li Shihao Zhang Baijuan Wang |
| author_sort | Zejun Wang |
| collection | DOAJ |
| description | To facilitate the realization of automated tea picking and enhance the speed and accuracy of tea leaf grading detection, this study proposes an improved YOLOv8 network for fresh tea leaf grading recognition. This approach integrates a Hierarchical Vision Transformer using Shifted Windows to replace segments of the original YOLOv8’s network architecture, thereby alleviating the computational load of dense image processing tasks and reducing computational expenses. The incorporation of an Efficient Multi-Scale Attention Module with Cross-Spatial Learning serves to attenuate the influence of irrelevant features in complex backgrounds, which in turn, elevates the model’s detection Precision. Additionally, the substitution of the loss function with SIoU facilitates a more rapid model convergence and a more precise pinpointing of defect locations. The empirical findings indicate that the enhanced YOLOv8 algorithm has achieved a marked improvement in metrics such as Precision, Recall, F1, and mAP, with increases of 3.39%, 0.86%, 2.20%, and 2.81% respectively, when juxtaposed with the original YOLOv8 model. Moreover, in external validations, the FPS enhancements over the original YOLOv8, YOLOv5, YOLOX, Faster RCNN, and SSD deep-learning models are 6.75 Hz, 10.84 Hz, 12.79 Hz, 28.24 Hz, and 21.57 Hz, respectively, and the mAP improvements in practical detection are 2.79%, 2.92%, 3.08%, 7.07%, and 3.84% respectively. The refined model not only ensures efficient and accurate tea-grading recognition but also boasts high recognition rates and swift detection capabilities, thereby establishing a foundation for the development of tea-picking robots and tea quality grading devices. |
| format | Article |
| id | doaj-art-56cade4839424bb8b221cb69c4eab568 |
| institution | OA Journals |
| issn | 2311-7524 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Horticulturae |
| spelling | doaj-art-56cade4839424bb8b221cb69c4eab5682025-08-20T02:00:23ZengMDPI AGHorticulturae2311-75242024-12-011012134710.3390/horticulturae10121347Fresh Tea Leaf-Grading Detection: An Improved YOLOv8 Neural Network Model Utilizing Deep LearningZejun Wang0Yuxin Xia1Houqiao Wang2Xiaohui Liu3Raoqiong Che4Xiaoxue Guo5Hongxu Li6Shihao Zhang7Baijuan Wang8College of Tea Science, Yunnan Agricultural University, Kunming 650201, ChinaKey Laboratory of Intelligent Organic Tea Garden Construction in University of Yunnan Province, Kunming 650201, ChinaCollege of Tea Science, Yunnan Agricultural University, Kunming 650201, ChinaCollege of Tea Science, Yunnan Agricultural University, Kunming 650201, ChinaCollege of Tea Science, Yunnan Agricultural University, Kunming 650201, ChinaCollege of Mechanical and Electrical Engineering, Wuhan Donghu University, Wuhan 430212, ChinaCollege of Tea Science, Yunnan Agricultural University, Kunming 650201, ChinaCollege of Mechanical and Electrical Engineering, Wuhan Donghu University, Wuhan 430212, ChinaCollege of Tea Science, Yunnan Agricultural University, Kunming 650201, ChinaTo facilitate the realization of automated tea picking and enhance the speed and accuracy of tea leaf grading detection, this study proposes an improved YOLOv8 network for fresh tea leaf grading recognition. This approach integrates a Hierarchical Vision Transformer using Shifted Windows to replace segments of the original YOLOv8’s network architecture, thereby alleviating the computational load of dense image processing tasks and reducing computational expenses. The incorporation of an Efficient Multi-Scale Attention Module with Cross-Spatial Learning serves to attenuate the influence of irrelevant features in complex backgrounds, which in turn, elevates the model’s detection Precision. Additionally, the substitution of the loss function with SIoU facilitates a more rapid model convergence and a more precise pinpointing of defect locations. The empirical findings indicate that the enhanced YOLOv8 algorithm has achieved a marked improvement in metrics such as Precision, Recall, F1, and mAP, with increases of 3.39%, 0.86%, 2.20%, and 2.81% respectively, when juxtaposed with the original YOLOv8 model. Moreover, in external validations, the FPS enhancements over the original YOLOv8, YOLOv5, YOLOX, Faster RCNN, and SSD deep-learning models are 6.75 Hz, 10.84 Hz, 12.79 Hz, 28.24 Hz, and 21.57 Hz, respectively, and the mAP improvements in practical detection are 2.79%, 2.92%, 3.08%, 7.07%, and 3.84% respectively. The refined model not only ensures efficient and accurate tea-grading recognition but also boasts high recognition rates and swift detection capabilities, thereby establishing a foundation for the development of tea-picking robots and tea quality grading devices.https://www.mdpi.com/2311-7524/10/12/1347grading recognitionimproved YOLOv8Hierarchical Vision Transformer using Shifted WindowsEfficient Multi-Scale Attention Module with Cross-Spatial LearningSIoU |
| spellingShingle | Zejun Wang Yuxin Xia Houqiao Wang Xiaohui Liu Raoqiong Che Xiaoxue Guo Hongxu Li Shihao Zhang Baijuan Wang Fresh Tea Leaf-Grading Detection: An Improved YOLOv8 Neural Network Model Utilizing Deep Learning Horticulturae grading recognition improved YOLOv8 Hierarchical Vision Transformer using Shifted Windows Efficient Multi-Scale Attention Module with Cross-Spatial Learning SIoU |
| title | Fresh Tea Leaf-Grading Detection: An Improved YOLOv8 Neural Network Model Utilizing Deep Learning |
| title_full | Fresh Tea Leaf-Grading Detection: An Improved YOLOv8 Neural Network Model Utilizing Deep Learning |
| title_fullStr | Fresh Tea Leaf-Grading Detection: An Improved YOLOv8 Neural Network Model Utilizing Deep Learning |
| title_full_unstemmed | Fresh Tea Leaf-Grading Detection: An Improved YOLOv8 Neural Network Model Utilizing Deep Learning |
| title_short | Fresh Tea Leaf-Grading Detection: An Improved YOLOv8 Neural Network Model Utilizing Deep Learning |
| title_sort | fresh tea leaf grading detection an improved yolov8 neural network model utilizing deep learning |
| topic | grading recognition improved YOLOv8 Hierarchical Vision Transformer using Shifted Windows Efficient Multi-Scale Attention Module with Cross-Spatial Learning SIoU |
| url | https://www.mdpi.com/2311-7524/10/12/1347 |
| work_keys_str_mv | AT zejunwang freshtealeafgradingdetectionanimprovedyolov8neuralnetworkmodelutilizingdeeplearning AT yuxinxia freshtealeafgradingdetectionanimprovedyolov8neuralnetworkmodelutilizingdeeplearning AT houqiaowang freshtealeafgradingdetectionanimprovedyolov8neuralnetworkmodelutilizingdeeplearning AT xiaohuiliu freshtealeafgradingdetectionanimprovedyolov8neuralnetworkmodelutilizingdeeplearning AT raoqiongche freshtealeafgradingdetectionanimprovedyolov8neuralnetworkmodelutilizingdeeplearning AT xiaoxueguo freshtealeafgradingdetectionanimprovedyolov8neuralnetworkmodelutilizingdeeplearning AT hongxuli freshtealeafgradingdetectionanimprovedyolov8neuralnetworkmodelutilizingdeeplearning AT shihaozhang freshtealeafgradingdetectionanimprovedyolov8neuralnetworkmodelutilizingdeeplearning AT baijuanwang freshtealeafgradingdetectionanimprovedyolov8neuralnetworkmodelutilizingdeeplearning |