A Lightweight Model for Shine Muscat Grape Detection in Complex Environments Based on the YOLOv8 Architecture
Automated harvesting of “Sunshine Rose” grapes requires accurate detection and classification of grape clusters under challenging orchard conditions, such as occlusion and variable lighting, while ensuring that the model can be deployed on resource- and computation-constrained edge devices. This stu...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Agronomy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4395/15/1/174 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832589380271734784 |
|---|---|
| author | Changlei Tian Zhanchong Liu Haosen Chen Fanglong Dong Xiaoxiang Liu Cong Lin |
| author_facet | Changlei Tian Zhanchong Liu Haosen Chen Fanglong Dong Xiaoxiang Liu Cong Lin |
| author_sort | Changlei Tian |
| collection | DOAJ |
| description | Automated harvesting of “Sunshine Rose” grapes requires accurate detection and classification of grape clusters under challenging orchard conditions, such as occlusion and variable lighting, while ensuring that the model can be deployed on resource- and computation-constrained edge devices. This study addresses these challenges by proposing a lightweight YOLOv8-based model, incorporating DualConv and the novel C2f-GND module to enhance feature extraction and reduce computational complexity. Evaluated on the newly developed Shine-Muscat-Complex dataset of 4715 images, the proposed model achieved a 2.6% improvement in mean Average Precision (mAP) over YOLOv8n while reducing parameters by 36.8%, FLOPs by 34.1%, and inference time by 15%. Compared with the latest YOLOv11n, our model achieved a 3.3% improvement in mAP, with reductions of 26.4% in parameters, 14.3% in FLOPs, and 14.6% in inference time, demonstrating comprehensive enhancements. These results highlight the potential of our model for accurate and efficient deployment on resource-constrained edge devices, providing an algorithmic foundation for the automated harvesting of “Sunshine Rose” grapes. |
| format | Article |
| id | doaj-art-1943d94dbd344bf690b61a28e7add4d4 |
| institution | Kabale University |
| issn | 2073-4395 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Agronomy |
| spelling | doaj-art-1943d94dbd344bf690b61a28e7add4d42025-01-24T13:17:02ZengMDPI AGAgronomy2073-43952025-01-0115117410.3390/agronomy15010174A Lightweight Model for Shine Muscat Grape Detection in Complex Environments Based on the YOLOv8 ArchitectureChanglei Tian0Zhanchong Liu1Haosen Chen2Fanglong Dong3Xiaoxiang Liu4Cong Lin5School of Intelligent Systems Science and Engineering/JNU-Industry School of Artificial Intelligence, Jinan University, Zhuhai 519000, ChinaSchool of Intelligent Systems Science and Engineering/JNU-Industry School of Artificial Intelligence, Jinan University, Zhuhai 519000, ChinaSchool of Intelligent Systems Science and Engineering/JNU-Industry School of Artificial Intelligence, Jinan University, Zhuhai 519000, ChinaSchool of Intelligent Systems Science and Engineering/JNU-Industry School of Artificial Intelligence, Jinan University, Zhuhai 519000, ChinaSchool of Intelligent Systems Science and Engineering/JNU-Industry School of Artificial Intelligence, Jinan University, Zhuhai 519000, ChinaSchool of Intelligent Systems Science and Engineering/JNU-Industry School of Artificial Intelligence, Jinan University, Zhuhai 519000, ChinaAutomated harvesting of “Sunshine Rose” grapes requires accurate detection and classification of grape clusters under challenging orchard conditions, such as occlusion and variable lighting, while ensuring that the model can be deployed on resource- and computation-constrained edge devices. This study addresses these challenges by proposing a lightweight YOLOv8-based model, incorporating DualConv and the novel C2f-GND module to enhance feature extraction and reduce computational complexity. Evaluated on the newly developed Shine-Muscat-Complex dataset of 4715 images, the proposed model achieved a 2.6% improvement in mean Average Precision (mAP) over YOLOv8n while reducing parameters by 36.8%, FLOPs by 34.1%, and inference time by 15%. Compared with the latest YOLOv11n, our model achieved a 3.3% improvement in mAP, with reductions of 26.4% in parameters, 14.3% in FLOPs, and 14.6% in inference time, demonstrating comprehensive enhancements. These results highlight the potential of our model for accurate and efficient deployment on resource-constrained edge devices, providing an algorithmic foundation for the automated harvesting of “Sunshine Rose” grapes.https://www.mdpi.com/2073-4395/15/1/174grape cluster detection and classificationlightweightYOLOv8 |
| spellingShingle | Changlei Tian Zhanchong Liu Haosen Chen Fanglong Dong Xiaoxiang Liu Cong Lin A Lightweight Model for Shine Muscat Grape Detection in Complex Environments Based on the YOLOv8 Architecture Agronomy grape cluster detection and classification lightweight YOLOv8 |
| title | A Lightweight Model for Shine Muscat Grape Detection in Complex Environments Based on the YOLOv8 Architecture |
| title_full | A Lightweight Model for Shine Muscat Grape Detection in Complex Environments Based on the YOLOv8 Architecture |
| title_fullStr | A Lightweight Model for Shine Muscat Grape Detection in Complex Environments Based on the YOLOv8 Architecture |
| title_full_unstemmed | A Lightweight Model for Shine Muscat Grape Detection in Complex Environments Based on the YOLOv8 Architecture |
| title_short | A Lightweight Model for Shine Muscat Grape Detection in Complex Environments Based on the YOLOv8 Architecture |
| title_sort | lightweight model for shine muscat grape detection in complex environments based on the yolov8 architecture |
| topic | grape cluster detection and classification lightweight YOLOv8 |
| url | https://www.mdpi.com/2073-4395/15/1/174 |
| work_keys_str_mv | AT changleitian alightweightmodelforshinemuscatgrapedetectionincomplexenvironmentsbasedontheyolov8architecture AT zhanchongliu alightweightmodelforshinemuscatgrapedetectionincomplexenvironmentsbasedontheyolov8architecture AT haosenchen alightweightmodelforshinemuscatgrapedetectionincomplexenvironmentsbasedontheyolov8architecture AT fanglongdong alightweightmodelforshinemuscatgrapedetectionincomplexenvironmentsbasedontheyolov8architecture AT xiaoxiangliu alightweightmodelforshinemuscatgrapedetectionincomplexenvironmentsbasedontheyolov8architecture AT conglin alightweightmodelforshinemuscatgrapedetectionincomplexenvironmentsbasedontheyolov8architecture AT changleitian lightweightmodelforshinemuscatgrapedetectionincomplexenvironmentsbasedontheyolov8architecture AT zhanchongliu lightweightmodelforshinemuscatgrapedetectionincomplexenvironmentsbasedontheyolov8architecture AT haosenchen lightweightmodelforshinemuscatgrapedetectionincomplexenvironmentsbasedontheyolov8architecture AT fanglongdong lightweightmodelforshinemuscatgrapedetectionincomplexenvironmentsbasedontheyolov8architecture AT xiaoxiangliu lightweightmodelforshinemuscatgrapedetectionincomplexenvironmentsbasedontheyolov8architecture AT conglin lightweightmodelforshinemuscatgrapedetectionincomplexenvironmentsbasedontheyolov8architecture |