# Reasearch on Kiwi Fruit Flower Recognition for Efficient Pollination Based on an Improved YOLOv5 Algorithm

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Materials and Methods

#### 2.1. Image Acquisition

#### 2.2. Based on YOLOv5 Algorithm

#### 2.3. Methods Used in This Study

#### 2.3.1. Use of K-Means++ to Cluster out New Anchor Boxes

#### 2.3.2. Adding CBAM Mechanism

#### 2.3.3. CIOU Replaces GIOU

#### 2.3.4. Detection Function for Flower Angle Calculation Module

#### 2.3.5. Search for Pollination Points Based on Flower Overlap and Flower Angle Identification

_{(i-1)}, x

_{i}, x

_{(i+1)}is the abscissa of the overlapping flower centroid, y

_{(i-1)}, y

_{i}, y

_{(i+1)}is the ordinate of the overlapping flower centroid.

_{x}is the abscissa of the pollination point, c

_{y}is the ordinate of pollination point.

_{1}and x

_{2}are the horizontal coordinates of the pollination point and the flower centre point, respectively, and H is the pollination distance.

#### 2.4. Model Evaluation

#### 2.5. Data Set Construction and Model Parameter

## 3. Results

#### 3.1. Training Results

#### 3.2. Accuracy Rate of Flower Angle Recognition

#### 3.3. Comparison Experiment Conducted to Identify Kiwi Flower Overlap

#### 3.4. Comparative Test of Four YOLOv5 Models

## 4. Discussion

## Author Contributions

## Funding

## Data Availability Statement

## Conflicts of Interest

## References

- Guan, L. Japan’s kiwi artificial pollination technology. J. Deciduous Fruit Trees
**2002**, 5, 60. [Google Scholar] - Jiang, Z.J. Optimization of Double-Flow Spray Parameters and Development of Kiwifruit Pollination Device. Master’s Thesis, Northwest Agriculture and Forestry University of Science and Technology, Xi’an, China, 2020. [Google Scholar]
- Krizhevsky, A.; Sutskever, I.; Hinton, G.E. ImageNet classification with deep convolutional neural networks. Adv. Neural Inf. Process. Syst.
**2012**, 25, 1097–1105. [Google Scholar] [CrossRef][Green Version] - Liu, T.; Teng, G.; Yuan, Y.; Liu, B.; Liu, Z. Recognition method of winter jujube fruit in natural scene based on improved YOLO-v3. Trans. Chin. Soc. Agric. Mach.
**2021**, 52, 17–25. [Google Scholar] - Zhang, H.; Fu, Z.; Han, W.; Yang, G.; Niu, D.; Zhou, X. Maize seedling number acquisition method based on improved YOLO. Trans. Chin. Soc. Agric. Mach.
**2021**, 52, 221–229. [Google Scholar] - Li, K.J. Research and Application of Weed Detection Algorithm in Cotton Field based on YOLOv3. Master’s Thesis, Xinjiang University, Ürümqi, China, 2021. [Google Scholar]
- Yue, Y.; Li, X.; Zhao, H.; Wang, H. Crop disease image recognition based on improved VGG networks. J. Agric. Mech. Res.
**2022**, 44, 18–24. [Google Scholar] - Li, H.; Li, C.; Li, G.; Chen, L. A real-time table grape detection method based on improved YOLOv4-tiny network in complex background. Biosyst. Eng.
**2021**, 212, 347–359. [Google Scholar] [CrossRef] - Wang, H.; Xu, Y.; He, Y.; Cai, Y.; Chen, L.; Li, Y.; Angel Sotelo, M.; Li, Z. YOLOv5-Fog: A Multiobjective Visual Detection Algorithm for Fog Driving Scenes Based on Improved YOLOv5. IEEE Trans. Instrum. Meas.
**2022**, 71, 1–12. [Google Scholar] [CrossRef] - Lv, J.; Xu, H.; Han, Y.; Lu, W.; Xu, L.; Rong, H.; Yang, B.; Li, Z.; Ma, Z. A visual identification method for the apple growth forms in the orchard. Comput. Electron. Agric.
**2022**, 197, 106954. [Google Scholar] [CrossRef] - Chaschatzis, C.; Karaiskou, C.; Mouratidis, E.G.; Karagiannis, E.; Sarigiannidis, P.G. Detection and Characterization of Stressed Sweet Cherry Tissues Using Machine Learning. Drones
**2021**, 6, 3. [Google Scholar] [CrossRef] - Ficzere, M.; Mészáros, L.A.; Kállai-Szabó, N.; Kovács, A.; Antal, I.; Nagy, Z.K.; Galata, D.L. Real-time coating thickness measurement and defect recognition of film coated tablets with machine vision and deep learning. Int. J. Pharm.
**2022**, 623, 121957. [Google Scholar] [CrossRef] - Jin, Y.; Gao, H.; Fan, X.; Khan, H.; Chen, Y. Defect Identification of Adhesive Structure Based on DCGAN and YOLOv5. IEEE Access
**2022**, 10, 79913–79924. [Google Scholar] [CrossRef] - Li, H.; Yang, G. Dietary Nutritional Information Autonomous Perception Method Based on Machine Vision in Smart Homes. Entropy
**2022**, 24, 868. [Google Scholar] [CrossRef] - Xue, J.; Zheng, Y.; Dong-Ye, C.; Wang, P.; Yasir, M. Improved YOLOv5 network method for remote sensing image-based ground objects recognition. Soft Comput.
**2022**, 26, 10879–10889. [Google Scholar] [CrossRef] - Yao, J.; Qi, J.; Zhang, J.; Shao, H.; Yang, J.; Li, X. A Real-Time Detection Algorithm for Kiwifruit Defects Based on YOLOv5. Electronics
**2021**, 10, 1711. [Google Scholar] [CrossRef] - Zhao, W.; Liu, S.; Li, X.; Han, X.; Yang, H. Fast and accurate wheat grain quality detection based on improved YOLOv5. Comput. Electron. Agric.
**2022**, 202, 107426. [Google Scholar] [CrossRef] - Zhang, H.; Tian, M.; Shao, G.; Cheng, J.; Liu, J. Target Detection of Forward-Looking Sonar Image Based on Improved YOLOv5. IEEE Access
**2022**, 10, 18023–18034. [Google Scholar] [CrossRef] - Zhang, Z.; Qiao, Y.; Guo, Y.; He, D. Deep Learning Based Automatic Grape Downy Mildew Detection. Front. Plant Sci.
**2022**, 13, 872107. [Google Scholar] [CrossRef] - Dai, G.; Hu, L.; Fan, J.; Yan, S.; Li, R. A Deep Learning-Based Object Detection Scheme by Improving YOLOv5 for Sprouted Potatoes Datasets. IEEE Access
**2022**, 10, 85416–85428. [Google Scholar] [CrossRef] - Gao, G.; Wang, S.; Shuai, C.; Zhang, Z.; Zhang, S.; Feng, Y. Recognition and Detection of Greenhouse Tomatoes in Complex Environment. Traitement Signal.
**2022**, 39, 291–298. [Google Scholar] [CrossRef] - Ye, Z.; Guo, Q.; Wei, J.; Zhang, J.; Zhang, H.; Bian, L.; Guo, S.; Zheng, X.; Cao, S. Recognition of terminal buds of densely-planted Chinese fir seedlings using improved YOLOv5 by integrating attention mechanism. Front. Plant Sci.
**2022**, 13, 991929. [Google Scholar] [CrossRef] - Li, D.H.; Zhao, H.; Yu, X. Overlapping green apple recognition based on improved spectral clustering. Spectrosc. Spect. Anal.
**2019**, 39, 2974–2981. [Google Scholar] - Uijlings, J.R.; Van De Sande, K.E.; Gevers, T.; Smeulders, A.W. Selective search for object recognition. Int. J. Comput. Vis.
**2013**, 104, 154–171. [Google Scholar] [CrossRef][Green Version] - Miao, R.; Shan, Z.; Zhou, Q.; Wu, Y.; Ge, L.; Zhang, J.; Hu, H. Real-time defect identification of narrow overlap welds and application based on convolutional neural networks. J. Manuf. Syst.
**2022**, 62, 800–810. [Google Scholar] [CrossRef]

**Figure 6.**Schematic of pollination points under overlapping conditions: (

**a**) single flower; (

**b**) overlap of two; (

**c**) overlap of three; (

**d**) four or more overlapping.

Data Set | Place | Condition of Overlap | Set of Training | Set of Verification | Set of Tests |
---|---|---|---|---|---|

D1 | Orchard greenhouses | Single flower | 800 | 200 | 200 |

D2 | Overlap of two | 800 | 200 | 200 | |

D3 | Overlap of three | 800 | 200 | 200 | |

D4 | Four or more overlapping | 800 | 200 | 200 |

Parameters | Value |
---|---|

Input size/pixels | 640 × 640 |

Initial learning rate | 0.032 |

Momentum | 0.843 |

Cyclical learning rate | 0.12 |

Iteration | 200 |

Model | Predicted Value (Flowers) | Rate of Recall (Flowers) | Predicted Value (Stamens) | Rate of Recall (Stamens) | mAP@0.5 |
---|---|---|---|---|---|

YOLOv5s | 96.7% | 89.1% | 91.1% | 78.3% | 90.1% |

Faster-RCNN-ResNet50 | 57.4% | 98.9% | 58.4% | 97.9% | 92.6% |

Faster-RCNN-VGG | 68.5% | 98.9% | 67.9% | 98.0% | 95.6% |

SSD-VGG | 76.6% | 87.4% | 82.8% | 65.6% | 82.3% |

SSD-MobileNetv2 | 86.7% | 70.2% | 89.6% | 55.0% | 81.1% |

Model | F1 | Average Time Per Frame (Milliseconds) | Memory (MB) |
---|---|---|---|

YOLOv5s | 90.12% | 8.64 | 20 |

YOLOv5m | 93.97% | 40.5 | 134 |

YOLOv5l | 85.74% | 425 | 278 |

YOLOv5x | 91.24% | 13.87 | 60.8 |

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |

© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Zhou, H.; Ou, J.; Meng, P.; Tong, J.; Ye, H.; Li, Z.
Reasearch on Kiwi Fruit Flower Recognition for Efficient Pollination Based on an Improved YOLOv5 Algorithm. *Horticulturae* **2023**, *9*, 400.
https://doi.org/10.3390/horticulturae9030400

**AMA Style**

Zhou H, Ou J, Meng P, Tong J, Ye H, Li Z.
Reasearch on Kiwi Fruit Flower Recognition for Efficient Pollination Based on an Improved YOLOv5 Algorithm. *Horticulturae*. 2023; 9(3):400.
https://doi.org/10.3390/horticulturae9030400

**Chicago/Turabian Style**

Zhou, Haili, Junlang Ou, Penghao Meng, Junhua Tong, Hongbao Ye, and Zhen Li.
2023. "Reasearch on Kiwi Fruit Flower Recognition for Efficient Pollination Based on an Improved YOLOv5 Algorithm" *Horticulturae* 9, no. 3: 400.
https://doi.org/10.3390/horticulturae9030400