通信与信息技术Issue(4)：7-12,6.
基于YOLO模型的成熟葡萄簇识别定位系统设计
Design of mature grape cluster recognition and localization system based on YOLO model
摘要
Abstract
With the development of modern agricultural technology,automated grape-picking robots have demonstrated immense po-tential in enhancing picking efficiency and reducing labor costs.However,the complexity of the orchard environment poses a significant challenge for robots to accurately identify and localize mature grape clusters,thereby affecting the accuracy and efficiency of harvesting.To address this challenge,this study proposes the utilization of an advanced deep learning object detection algorithm,specifically the YO-LO model.By selecting and comparing the performance of three versions,YOLOv5,YOLOv8,and YOLOv10,a rapid and efficient system for identifying the location and category of mature grape clusters is achieved.A comprehensive grape cluster target detection system has been developed,utilizing the latest Python binding library PySide6,which enables real-time image and video processing capabilities.The user-friendly GUI interface simplifies the operation process and supports target detection for static images,dynamic videos,and real-time camera inputs.The system automatically labels grape clusters in images or videos and displays the detection results in an intuitive man-ner on the UI interface.Additionally,the system incorporates a function to save detection results locally,facilitating subsequent analysis and record-keeping.This comprehensive detection and display capability not only improves the accuracy of recognizing and localizing mature grape clusters but also provides an effective solution for the localization of picking points in grape-picking robots.Experimental results demonstrate that this method exhibits high accuracy and robustness,offering robust technical support for automated grape harvesting.关键词
葡萄簇/成熟/YOLO/UI/识别/定位Key words
Grape cluster/Maturity/YOLO/UI/Recognition/Localization分类
信息技术与安全科学引用本文复制引用
胡艳茹,田苏慧敏,王旭阳..基于YOLO模型的成熟葡萄簇识别定位系统设计[J].通信与信息技术,2025,(4):7-12,6.基金项目
宁夏自然科学基金项目:基于深度学习的变压器油老化状态间接检测研究(项目编号:2023AAC03353). （项目编号:2023AAC03353）
