彭红星^1,2,3，梁啟俊¹，邹湘军^2,4，王红军^2,5，熊俊涛^1,2*，罗彦淋¹，郭尚昆¹，沈冠佳¹

(1.华南农业大学数学与信息学院，广州 510642，中国；

2. 佛山市中科农业机器人与智慧农业创新研究院，佛山 528251，广东，中国；

3. 农业农村部华南热带智慧农业技术重点实验室，广州 510642，中国；

4. 新疆大学智能制造现代产业学院，乌鲁木齐 830046，中国；

5. 华南农业大学工程学院，广州 510642，中国)

摘要：在非结构化荔枝园中，荔枝果实和采摘点的精确识别和定位对于荔枝采摘机器人至关重要。大多数研究采用多步骤方法来检测水果和定位采摘点，这些方法速度慢，而且难于应对复杂的环境。该研究提出了一种基于YOLOv8-pose的YOLOv8-iGR改进模型，该模型集成端到端网络，可同时用于对象检测和关键点检测。

首先，本研究考虑了辅助点对采摘点的影响，并设计了四种荔枝关键点策略；其次，提出了一种名为iSaE的架构，该架构结合了CNN和注意力机制的能力；随后，原模型中的C2f由通用高效层聚合网络（GELAN）取代，以减少模型冗余并提高检测效率精度；最后，基于RFAConv，RFAPoseHead被设计用于解决大型卷积核中的参数共享问题，从而更有效地提取特征信息。

实验结果表明，YOLOv8-iGR在不同场景的荔枝果实检测中，实现了95.7%的AP，采摘点的欧几里德距离误差小于8像素，满足荔枝采摘的实际要求。此外，模型的GFLOPs降低了10.71%。通过野外采摘实验测试了模型对采摘点定位的准确性。总之，YOLOv8-iGR具有出色的检测性能和较低的模型复杂度，使其更适合在机器人上实施，同时也可为荔枝采摘机器人的视觉系统提供技术支持。 

关键词：荔枝；目标检测；采摘点检测；YOLOv8-pose；采摘机器人

DOI: 10.25165/j.ijabe.20251804.9303

引用信息:  Peng H X, Liang Q J, Zou X J, Wang H J, Xiong J T, Luo Y L, et al. Synchronous detection method for litchi fruits and picking points of a litchi-picking robot based on improved YOLOv8-pose. Int J Agric & Biol Eng, 2025; 18(4): 266–274.

阅读全文链接：http://ijabe.net/article/doi/10.25165/j.ijabe.20251804.9303

Synchronous detection method for litchi fruits and picking points of a litchi-picking robot based on improved YOLOv8-pose

Hongxing Peng^1,2,3, Qijun Liang¹, Xiangjun Zou^2,4, Hongjun Wang^2,5, Juntao Xiong^1,2*, Yanlin Luo¹, Shangkun Guo¹, Guanjia Shen¹

(1. College of Mathematics and Informatics, South China Agricultural University, Guangzhou 510642, China;

2. Foshan-Zhongke Innovation Research Institute of Intelligent Agriculture and Robotics, Foshan528251, Guangdong, China;

3. Key Laboratory of Smart Agricultural Technology in Tropical South China, Ministry of Agriculture and Rural Affairs, Guangzhou 510642, China;

4. College of Intelligent Manufacturing and Modern Industry, Xinjiang University, Urumqi 830046, Xinjiang, China;

5. College of Engineering, South China Agricultural University, Guangzhou 510642, China)

Abstract: In the unstructured litchi orchard, precise identification and localization of litchi fruits and picking points are crucial for litchi-picking robots. Most studies adopt multi-step methods to detect fruit and locate picking points, which are slow and struggle to cope with complex environments. This study proposes a YOLOv8-iGR model based on YOLOv8n-pose improvement, integrating end-to-end network for both object detection and key point detection. Specifically, this study considers the influence of auxiliary points on picking point and designs four litchi key point strategies. Secondly, the architecture named iSaE is proposed, which combines the capabilities of CNN and attention mechanism. Subsequently, C2f is replaced by Generalized Efficient Layer Aggregation Network (GELAN) to reduce model redundancy and improve detection accuracy. Finally, based on RFAConv, RFAPoseHead is designed  to  address  the  issue  of  parameter sharing in large convolutional kernels, thereby more effectively extracting feature information. Experimental results demonstrate that YOLOv8-iGR achieves an AP of 95.7% in litchi fruit detection, and the Euclidean distance error of picking points is less than 8 pixels across different scenes, meeting the requirements of litchi picking. Additionally, the GFLOPs of the model are reduced by 10.71%. The accuracy of the model’s localization for picking  points was  tested  through  field  picking  experiments. In conclusion, YOLOv8-iGR exhibits outstanding detection performance along with lower model complexity, making it more feasible for implementation on robots. This will provide technical support for the vision system of the litchi-picking robot.

Keywords: litchi, object detection, picking point detection, YOLOv8-pose, picking robot

DOI: 10.25165/j.ijabe.20251804.9303

Citation:  Peng H X, Liang Q J, Zou X J, Wang H J, Xiong J T, Luo Y L, et al. Synchronous detection method for litchi fruits and picking points of a litchi-picking robot based on improved YOLOv8-pose. Int J Agric & Biol Eng, 2025; 18(4): 266–274.

支持单位：