Geometry-Aware 3D Point Cloud Learning for Precise Cutting-Point Detection in Unstructured Field Environments

IF 5.2 2区 计算机科学 Q2 ROBOTICS
Hongjun Wang, Gengming Zhang, Hao Cao, Kewei Hu, Quanchao Wang, Yuqin Deng, Junfeng Gao, Yunchao Tang
{"title":"Geometry-Aware 3D Point Cloud Learning for Precise Cutting-Point Detection in Unstructured Field Environments","authors":"Hongjun Wang,&nbsp;Gengming Zhang,&nbsp;Hao Cao,&nbsp;Kewei Hu,&nbsp;Quanchao Wang,&nbsp;Yuqin Deng,&nbsp;Junfeng Gao,&nbsp;Yunchao Tang","doi":"10.1002/rob.22567","DOIUrl":null,"url":null,"abstract":"<div>\n \n <p>In automated lychee harvesting, the complex geometric structures of branches, leaves, and clustered fruits pose significant challenges for robotic cutting point detection, where even minor positioning errors can lead to harvest damage and operational failures. This study introduces the Fcaf3d-lychee network model, specifically designed for precise lychee picking point localization. The data acquisition system utilizes Microsoft's Azure Kinect DK time-of-flight camera to capture point cloud data through multi-view stitching, enabling comprehensive spatial information capture. The proposed model enhances the Fully Convolutional Anchor-Free 3D Object Detection (Fcaf3d) architecture by incorporating a squeeze-and-excitation (SE) module, which leverages human visual attention mechanisms to improve feature extraction capabilities. Experimental results demonstrate the model's superior performance, achieving an <span></span><math>\n <semantics>\n <mrow>\n \n <mrow>\n <msub>\n <mi>F</mi>\n \n <mn>1</mn>\n </msub>\n </mrow>\n </mrow>\n </semantics></math> score of 88.57% on the test data set, significantly outperforming existing approaches. Field tests in real orchard environments show robust performance under varying occlusion conditions, with detection accuracies of 0.932, 0.824, and 0.765 for unobstructed, partially obstructed, and severely obstructed scenarios, respectively. The model maintains localization errors within <span></span><math>\n <semantics>\n <mrow>\n \n <mrow>\n <mo>±</mo>\n </mrow>\n </mrow>\n </semantics></math>1.5 cm in all directions, demonstrating exceptional precision for practical harvesting applications. This research advances the field of automated fruit harvesting by providing a reliable solution for accurate picking point detection, contributing to the development of more efficient agricultural robotics systems.</p>\n </div>","PeriodicalId":192,"journal":{"name":"Journal of Field Robotics","volume":"42 7","pages":"3063-3076"},"PeriodicalIF":5.2000,"publicationDate":"2025-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Field Robotics","FirstCategoryId":"94","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/rob.22567","RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ROBOTICS","Score":null,"Total":0}
引用次数: 0

Abstract

In automated lychee harvesting, the complex geometric structures of branches, leaves, and clustered fruits pose significant challenges for robotic cutting point detection, where even minor positioning errors can lead to harvest damage and operational failures. This study introduces the Fcaf3d-lychee network model, specifically designed for precise lychee picking point localization. The data acquisition system utilizes Microsoft's Azure Kinect DK time-of-flight camera to capture point cloud data through multi-view stitching, enabling comprehensive spatial information capture. The proposed model enhances the Fully Convolutional Anchor-Free 3D Object Detection (Fcaf3d) architecture by incorporating a squeeze-and-excitation (SE) module, which leverages human visual attention mechanisms to improve feature extraction capabilities. Experimental results demonstrate the model's superior performance, achieving an F 1 score of 88.57% on the test data set, significantly outperforming existing approaches. Field tests in real orchard environments show robust performance under varying occlusion conditions, with detection accuracies of 0.932, 0.824, and 0.765 for unobstructed, partially obstructed, and severely obstructed scenarios, respectively. The model maintains localization errors within ± 1.5 cm in all directions, demonstrating exceptional precision for practical harvesting applications. This research advances the field of automated fruit harvesting by providing a reliable solution for accurate picking point detection, contributing to the development of more efficient agricultural robotics systems.

Abstract Image

几何感知三维点云学习,用于非结构化现场环境中的精确切割点检测
在自动化的荔枝收获过程中,枝叶和果实的复杂几何结构给机器人切割点检测带来了巨大的挑战,即使是很小的定位错误也会导致收获损坏和操作失败。本研究引入了专为荔枝采摘点精确定位而设计的fcaf3d -荔枝网络模型。数据采集系统采用微软Azure Kinect DK飞行时间相机,通过多视角拼接捕捉点云数据,实现全面的空间信息捕获。该模型通过加入一个挤压和激励(SE)模块来增强全卷积无锚三维目标检测(Fcaf3d)架构,该模块利用人类视觉注意机制来提高特征提取能力。实验结果表明,该模型具有优异的性能,在测试数据集上的f1得分为88.57%,显著优于现有的方法。在实际果园环境中进行的现场测试表明,在不同遮挡条件下,该方法的检测精度分别为0.932、0.824和0.765,在无遮挡、部分遮挡和严重遮挡情况下,检测精度分别为0.932、0.824和0.765。该模型在所有方向上保持±1.5厘米的定位误差,为实际收获应用展示了卓越的精度。这项研究通过提供准确采摘点检测的可靠解决方案,推动了自动化水果收获领域的发展,有助于开发更高效的农业机器人系统。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Journal of Field Robotics
Journal of Field Robotics 工程技术-机器人学
CiteScore
15.00
自引率
3.60%
发文量
80
审稿时长
6 months
期刊介绍: The Journal of Field Robotics seeks to promote scholarly publications dealing with the fundamentals of robotics in unstructured and dynamic environments. The Journal focuses on experimental robotics and encourages publication of work that has both theoretical and practical significance.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信