Reports Outline Robotics Study Results from Guru Nanak Dev Engineering College ( Distributed Feature Matching for Robust Object Localization in Robotic Manipulat ion)
报告概述了来自Guru Nanak Dev Engineering的机器人研究成果鲁棒对象的分布式特征匹配(College)机器人的定位
Reports Outline Robotics Study Results from Guru Nanak Dev Engineering College ( Distributed Feature Matching for Robust Object Localization in Robotic Manipulat ion)
报告概述了来自Guru Nanak Dev Engineering的机器人研究成果鲁棒对象的分布式特征匹配(College)机器人的定位
扫码查看
点击上方二维码区域，可以放大扫码查看
摘要
由一名新闻记者-机器人与机器学习日报的工作人员新闻编辑每日新闻-调查人员讨论机器人的新发现。根据新闻报道NewsRx记者从印度旁遮普报道，研究表明：“特征匹配算法是用来重新认识平面物体或平面在图像中的位置。新闻记者从Guru Nanak Dev Engi Neering学院的研究中获得了一句话：“这特别适用于自动机械臂的控制，以进行采摘和放置操作。”单目视觉导航系统。问题出现在物体表面不平坦或被检测到的物体表面特征点属于不同高度平面。如果物体是远离摄像机视图中心，导致投影离子视差和视表面几何是扭曲的。本文提出的算法识别具有不同形状的水平面在单个平面上采用分布式匹配的方法来精确地找到目标的位置物体。这种方法需要两幅物体的图像来训练并找到一幅物体单幅图像，这允许仅使用单目摄像机进行三维模型匹配，而不使用机器学习需要大量训练图像数据集的技术。该算法对多平面系统最有效三维物体，在不同高度的水平面上有几个特征点。
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning DailyNews Daily News – Investigators discuss new findings in robotics. According to news reporting originatingfrom Punjab, India, by NewsRx correspondents, research stated, “The feature matching algorithms areused to re cognize the position of flat objects or surfaces in an image.”The news journalists obtained a quote from the research from Guru Nanak Dev Engi neering College:“This is particularly used for the control of autonomous robot arms for pick and place operations undermonocular vision guidance systems. The problem arises where the object surface is not flat or the detectedfeature poin ts belong to the different height planes. The error is much more prominent if th e object isplaced away from the center of the camera view that leads to project ion parallax and the apparent surfacegeometry is distorted. The algorithm propo sed in this paper identifies horizontal planes with differentheights and uses f eature matching on individual planes in a distributed way to find accurate posit ion ofthe object. Two images of the object are required by this method to train and then find the object in asingle image, this allows 3D model matching using only monocular camera without using machine learningtechniques thatrequire a l arge dataset of training images. The algorithm works best for the multi-planar3 D objects, which have several feature pointson different height horizontal plane levels.”
Key words
Guru Nanak Dev Engineering College/Punj ab/India/Asia/Emerging Technologies/Machine Learning/Robotics/Robots
NETL
NSTL