A Lightweight Cherry Tomato Maturity Real-Time Detection Algorithm Based on Improved YOLOV5n

被引:19
作者
Wang, Congyue [1 ,2 ]
Wang, Chaofeng [1 ,2 ]
Wang, Lele [2 ]
Wang, Jing [1 ,2 ]
Liao, Jiapeng [1 ,2 ]
Li, Yuanhong [1 ,2 ]
Lan, Yubin [1 ,2 ,3 ,4 ]
机构
[1] South China Agr Univ, Coll Elect Engn, Coll Artificial Intelligence, Guangzhou 510642, Peoples R China
[2] South China Agr Univ, Natl Ctr Int Collaborat Res Precis Agr Aviat Pesti, Guangzhou 510642, Peoples R China
[3] South China Agr Univ, Guangdong Lab Lingnan Modern Agr, Guangzhou 510642, Peoples R China
[4] Texas A&M Univ, Dept Biol & Agr Engn, College Stn, TX 77843 USA
来源
AGRONOMY-BASEL | 2023年 / 13卷 / 08期
关键词
precision agriculture; YOLOv5; cherry tomato; maturity detection; CA; WIoU; COLOR;
D O I
10.3390/agronomy13082106
中图分类号
S3 [农学(农艺学)];
学科分类号
0901 ;
摘要
To enhance the efficiency of mechanical automatic picking of cherry tomatoes in a precision agriculture environment, this study proposes an improved target detection algorithm based on YOLOv5n. The improvement steps are as follows: First, the K-means++ clustering algorithm is utilized to update the scale and aspect ratio of the anchor box, adapting it to the shape characteristics of cherry tomatoes. Secondly, the coordinate attention (CA) mechanism is introduced to expand the receptive field range and reduce interference from branches, dead leaves, and other backgrounds in the recognition of cherry tomato maturity. Next, the traditional loss function is replaced by the bounding box regression loss with dynamic focusing mechanism (WIoU) loss function. The outlier degree and dynamic nonmonotonic focusing mechanism are introduced to address the boundary box regression balance problem between high-quality and low-quality data. This research employs a self-built cherry tomato dataset to train the target detection algorithms before and after the improvements. Comparative experiments are conducted with YOLO series algorithms. The experimental results indicate that the improved model has achieved a 1.4% increase in both precision and recall compared to the previous model. It achieves an average accuracy mAP of 95.2%, an average detection time of 5.3 ms, and a weight file size of only 4.4 MB. These results demonstrate that the model fulfills the requirements for real-time detection and lightweight applications. It is highly suitable for deployment in embedded systems and mobile devices. The improved model presented in this paper enables real-time target recognition and maturity detection for cherry tomatoes. It provides rapid and accurate target recognition guidance for achieving mechanical automatic picking of cherry tomatoes.
引用
收藏
页数:24
相关论文
共 48 条
[41]   Fast and precise detection of litchi fruits for yield estimation based on the improved YOLOv5 model [J].
Wang, Lele ;
Zhao, Yingjie ;
Xiong, Zhangjun ;
Wang, Shizhou ;
Li, Yuanhong ;
Lan, Yubin .
FRONTIERS IN PLANT SCIENCE, 2022, 13
[42]  
[王立舒 Wang Lishu], 2021, [农业工程学报, Transactions of the Chinese Society of Agricultural Engineering], V37, P170
[43]  
Wang QL, 2020, PROC CVPR IEEE, P11531, DOI 10.1109/CVPR42600.2020.01155
[44]   Review of smart robots for fruit and vegetable picking in agriculture [J].
Wang, Zhiheng ;
Xun, Yi ;
Wang, Yingkuan ;
Yang, Qinghua .
INTERNATIONAL JOURNAL OF AGRICULTURAL AND BIOLOGICAL ENGINEERING, 2022, 15 (01) :33-54
[45]   CBAM: Convolutional Block Attention Module [J].
Woo, Sanghyun ;
Park, Jongchan ;
Lee, Joon-Young ;
Kweon, In So .
COMPUTER VISION - ECCV 2018, PT VII, 2018, 11211 :3-19
[46]  
Yang LX, 2021, PR MACH LEARN RES, V139
[47]   A New Apple Segmentation and Recognition Method Based on Modified Fuzzy C-means and Hough Transform [J].
Yin, Shoulin ;
Meng, Lei ;
Liu, Jie .
JOURNAL OF APPLIED SCIENCE AND ENGINEERING, 2019, 22 (02) :349-354
[48]  
Zheng ZH, 2020, AAAI CONF ARTIF INTE, V34, P12993