基于改进型YOLO11n算法的番茄幼苗嫁接点估计方法

李熔涛¹，袁法晖¹，Sajad Ali¹，印祥²，何勇^1,3*，刘羽飞^1,3*

(1.浙江大学生物系统工程与食品科学学院，杭州310058，中国；

2. 山东理工大学农业工程与食品科学学院，淄博255000，山东，中国；

3 农业农村部光谱检测重点实验室，杭州310058，中国)

摘要：嫁接机器人需要获取植物幼苗的位置信息以执行自动嫁接操作，准确定位嫁接点对于高质量完成嫁接任务至关重要。传统的视觉检测模型由于模型体积大，在边缘设备上的表现不佳，并且检测效率有限。为了实现快速且精确的嫁接点定位，该研究提出了一种全新的模块，称为激励上采样块（SUB）。此外，空间和通道重建卷积（SCConv）以及基于局部重要性注意力机制（LIA）被整合到 YOLO11n架构中，最终形成了YOLO11n-LSS模型。

我们的模型在实例分割任务中实现了93.2% 的平均精度均值（mAP），在关键点检测任务中达到了98.9%的mAP值。与YOLOv8n和YOLO11n相比，我们的模型分别减少了4.6% 的参数量和3.8% 的计算成本，成为一种高性能且轻量级的解决方案。新算法的成功应用将显著提高番茄自动嫁接的生产效率，推动番茄种植业的发展。

关键词：幼苗嫁接；茎秆检测；实例分割；关键点检测；YOLO模型

DOI: 10.25165/j.ijabe.20261901.10095

引用信息: Li R T, Yuan F H, Ali S, Yin X, He Y, Liu Y F. Method for the estimation of the cutting points in tomato seedling grafting based on improved YOLO11n. Int J Agric & Biol Eng, 2026; 19(1): 179–186.

阅读全文链接：http://ijabe.net/article/doi/10.25165/j.ijabe.20261901.10095

Method for the estimation of the cutting points in tomato seedling grafting based on improved YOLO11n

Rongtao Li¹, Fahui Yuan¹, Sajad Ali¹, Xiang Yin², Yong He^1,3*, Yufei Liu^1,3*

(1. College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, China; 

2. School of Agricultural Engineering and Food Science, Shandong University of Technology, Zibo 255000, Shandong, China; 

3. Key Laboratory of Spectroscopy Sensing, Ministry of Agriculture and Rural Affairs, Hangzhou 310058, China)

Abstract: A grafting robot needs to obtain the position information for the plant seedlings to perform automatic grafting operations. Accurately measuring the cutting points required during grafting plays a pivotal role in completing high-quality grafting tasks. Traditional visual detection models exhibit suboptimal performance on edge devices due to their large model size and suffer from limited detection efficiency. To achieve rapid and precise cutting point localization, this study proposes an all-new module termed the Stimulative Upsample Block (SUB). Additionally, the Spatial and Channel Reconstruction Convolution (SCConv) and a Local Importance-based Attention (LIA) mechanism are incorporated into the YOLO11n architecture, culminating in an enhanced model named YOLO11n-LSS. Our model achieved mean average precision (mAP) values of 93.2% for the instance segmentation task and 98.9% for the key point detection task. Compared to YOLOv8n and YOLO11n, our model reduces the number of parameters and computational cost by 4.6% and 3.8%, respectively, making it a high-performance and lightweight solution. The successful application of the new algorithm will significantly improve the production efficiency of automated tomato grafting and contribute to the advancement of the tomato cultivation industry.

Keywords:  seedling grafting, stem detection, instance segmentation, key point detection, YOLO algorithm

DOI: 10.25165/j.ijabe.20261901.10095

Citation: Li R T, Yuan F H, Ali S, Yin X, He Y, Liu Y F. Method for the estimation of the cutting points in tomato seedling grafting based on improved YOLO11n. Int J Agric & Biol Eng, 2026; 19(1): 179–186.

支持单位：