Register      Login
Crop and Pasture Science Crop and Pasture Science Society
Plant sciences, sustainable farming systems and food quality
Shopping Cart: ( empty )
You are here: Home > Journals > CP > CP24073
RESEARCH ARTICLE
Contents Vol 76(2)

Detecting rice (Oryza sativa) panicle using an improved YOLOv5 model

Xiaoyue Seng A B , Xue Yang A B , Tonghai Liu https://orcid.org/0000-0002-7390-7098 A C * , Rui Zhang A B , Chuangchuang Yuan A B , TianTian Guo A B and Wenzheng Liu A B
+ Author Affiliations
- Author Affiliations

A Tianjin Key Laboratory of Intelligent Breeding of Major Crops, Tianjin 300392, China.

B College of Computer and Information Engineering, Tianjin Agricultural University, Tianjin 300392, China.

C College of Engineering and Technology, Tianjin Agricultural University, Tianjin 300392, China.

* Correspondence to: tonghai_1227@126.com

Handling Editor: Davide Cammarano

Crop & Pasture Science 76, CP24073 https://doi.org/10.1071/CP24073
Submitted: 27 March 2024  Accepted: 20 January 2025  Published: 14 February 2025

© 2025 The Author(s) (or their employer(s)). Published by CSIRO Publishing

Abstract

Context

Rice (Oryza sativa) panicle provides important information to improve production efficiency, optimise resources, and aid in successful breeding of high-performing rice varieties.

Aims

In order to efficiently count rice panicles, a rice panicle recognition model based on YOLOv5s-Slim Neck-GhostNet was evaluated.

Methods

We used the developmental stages from heading to maturity as the time period to collect data for testing and validating the model. The GSConv convolution module from the YOLOv5 (You Only Look Once) model was compared with the original Conv convolution. We improved the original C3 module and replaced it with VoVGSCSP module, which further enhanced the detection ability of the model for small targets, such as rice panicles. To further optimise the performance of the model and reduce the computational complexity, we replaced the original backbone network of the model with a lightweight and efficient GhostNet structure.

Key results

Our results showed that the precision of the test set was 96.5%, the recall was 94.6%, the F1-score was 95.5%, and the mAP@0.5 was 97.2%. Compared with the original YOLOv5s model, mAP@0.5 increased by 1.8%, and the model size is reduced by 5.7M.

Conclusions

The improved YOLOv5 model had increased capability to detect and count rice panicles in real time. Our method reduced the size of the model while maintaining an acceptable level of accuracy.

Implications

The technology provides an intelligent and automated solution to better monitor rice panicle development, and has the potential for practical application in agricultural settings.

Keywords: agricultural production, count, deep learning, image processing, lightweighting, rice panicle, swamp rice, YOLOv5s.

References

Amrani A, Sohel F, Diepeveen D, Murray D, Jones MGK (2022) Insect detection from imagery using YOLOv3-based adaptive feature fusion convolution network. Crop & Pasture Science 74, 615-627.
| Crossref | Google Scholar |

Bao W, Zhang X, Hu G, Huang L, Liang D, Lin Z (2020) Estimation and counting of wheat ears density in field based on deep convolutional neural network. Transactions of the Chinese Society of Agricultural Engineering 36, 186-193.
| Crossref | Google Scholar |

Bao W, Yang X, Liang D, Hu G, Yang X (2021) Lightweight convolutional neural network model for field wheat ear disease identification. Computers and Electronics in Agriculture 189, 106367.
| Crossref | Google Scholar |

Cao Y, Liu Y, Ma D, Li A, Xu T (2020) Rice panicle UAV image segmentation method based on optimal subset selection. Transactions of the Chinese Society of Agricultural Machinery 51, 171-177, 188.
| Crossref | Google Scholar |

Chen H, Lu X, Tian F, Dong Q, Wang K, Han X (2013) Segmentation of wheat ear image based on Sobel operator edge detection. Journal of Agricultural Mechanization Research 35, 33-36.
| Crossref | Google Scholar |

Chen Y, Xin R, Jiang H, Liu Y, Zhang X, Yu J (2023) Refined feature fusion for in-field high-density and multi-scale rice panicle counting in UAV images. Computers and Electronics in Agriculture 211, 108032.
| Crossref | Google Scholar |

Dandrifosse S, Ennadifi E, Carlier A, Gosselin B, Dumont B, Mercatoris B (2022) Deep learning for wheat ear segmentation and ear density measurement: from heading to maturity. Computers and Electronics in Agriculture 199, 107161.
| Crossref | Google Scholar |

Dong Y, Liu Y, Li W, Liu Z, Li C (2023) An efficient and lightweight wheat ear detection algorithm based on the three-eye attention mechanism. Computer Engineering and Design 44, 2062-2069.
| Crossref | Google Scholar |

Du Z, Xu X, Bai Z, Liu X, Hu Y, Li W, Wang C, Li D (2023) Feature fusion strategy and improved GhostNet for accurate recognition of fish feeding behavior. Computers and Electronics in Agriculture 214, 108310.
| Crossref | Google Scholar |

Fang Y, Wu Q, Li S, Guan J, Cui Y (2024) Enhanced YOLOv5s-based algorithm for industrial part detection. Sensors 24, 1183.
| Crossref | Google Scholar |

Gündüz MŞ, Işık G (2023) A new YOLO-based method for real-time crowd detection from video and performance analysis of YOLO models. Journal of Real-Time Image Processing 20, 5.
| Crossref | Google Scholar |

Hayat MA, Wu J, Cao Y (2020) Unsupervised Bayesian learning for rice panicle segmentation with UAV images. Plant Methods 16, 18.
| Crossref | Google Scholar |

Huang W, Huo Y, Yang S, Liu M, Li H, Zhang M (2023) Detection of Laodelphax striatellus (small brown planthopper) based on improved YOLOv5. Computers and Electronics in Agriculture 206, 107657.
| Crossref | Google Scholar |

Kou L, Zhang H (2023) Research on multi-target recognition technology of apple picking robot based on improved YOLOv5. Journal of Chinese Agricultural Mechanization 44, 162-168.
| Crossref | Google Scholar |

Lei Y, Pan D, Feng Z, Qian J (2023) Lightweight YOLOv5s human ear recognition based on MobileNetV3 and Ghostnet. Applied Sciences 13, 6667.
| Crossref | Google Scholar |

Liao H, Yan W, Liu D (2023) Lightweight semantic segmentation method based on GhostNet and Atrous spatial pyramid pooling. Journal of Physics: Conference Series 2477, 012080.
| Crossref | Google Scholar |

Liu T, Sun C, Wang L, Zhong X, Zhu X, Guo W (2014) Field wheat spike counting based on image processing technology. Transactions of the Chinese Society of Agricultural Machinery 45, 282-290.
| Crossref | Google Scholar |

Liu Q, Dong L, Zeng Z, Zhu W, Zhu Y, Meng C (2023) SSD with multi-scale feature fusion and attention mechanism. Scientific Reports 13, 21387.
| Crossref | Google Scholar |

Mahmudul Hasan ASM, Sohel F, Diepeveen D, Laga H, Jones MGK (2022) Weed recognition using deep learning techniques on class-imbalanced imagery. Crop & Pasture Science 74, 628-644.
| Crossref | Google Scholar |

Ramachandran A, K.S. SK (2023) Tiny Criss-Cross Network for segmenting paddy panicles using aerial images. Computers and Electrical Engineering 108, 108728.
| Crossref | Google Scholar |

Wang X, Yang W, Lv Q, Huang C, Liang X, Chen G, Xiong L, Duan L (2022) Field rice panicle detection and counting based on deep learning. Frontiers in Plant Science 13, 966495.
| Crossref | Google Scholar |

Wang M, Sun K, Guo A (2023a) Wheat ear detection using anchor-free ObjectBox model with attention mechanism. Signal, Image and Video Processing 17, 3435-3432.
| Crossref | Google Scholar |

Wang B, Yang G, Yang H, Gu J, Xu S, Zhao D, Xu B (2023b) Multiscale maize tassel identification based on improved RetinaNet model and UAV images. Remote Sensing 15, 2530.
| Crossref | Google Scholar |

Xiao L, Pan Z, Du X, Chen W, Qu W, Bai Y, Xu T (2023) Weighted skip-connection feature fusion: a method for augmenting UAV oriented rice panicle image segmentation. Computers and Electronics in Agriculture 207, 107754.
| Crossref | Google Scholar |

Xu W, Wang Y (2024) A lightweight neural network for accurate rice panicle detection and counting in field conditions. International Journal of Advanced Computer Science and Applications 15(2), 561-569.
| Crossref | Google Scholar |

Xu C, Jiang H, Yuen P, Zaki Ahmad K, Chen Y (2020) MHW-PD: A robust rice panicles counting algorithm based on deep learning and multi-scale hybrid window. Computers and Electronics in Agriculture 173, 105375.
| Crossref | Google Scholar |

Yang W, Duan L, Yang W (2021) Research on rice phenotypic feature extraction and panicle quality prediction based on deep learning. Journal of Huazhong Agricultural University 40, 227-235.
| Crossref | Google Scholar |

Yuan C, Liu T, Gao F, Zhang R, Seng X (2023) YOLOv5s-CBAM-DMLHead: A lightweight identification algorithm for weedy rice (Oryza sativa f. spontanea) based on improved YOLOv5. Crop Protection 172, 106342.
| Crossref | Google Scholar |

Zhao F, Wang K, Yuan Y (2014) Study on wheat spike identification based on color features and AdaBoost algorithm. Crops Journal 1, 141-144.
| Crossref | Google Scholar |

Zhong H, Ma X, Li Z, Wang X, Liu S, Liu W, Wang C, Lin Y (2023) Rapid grading detection on hybrid rice bud seeds based on improved YOLOv5 model. Journal of South China Agricultural University 44, 960-967.
| Crossref | Google Scholar |