摘要
计算机视觉逐渐成为热门,由于视觉成像是射影几何,而交比是射影几何中基本的不变量,论文设计了一种基于交比不变量的新型人工标识物,有着可靠的检测、快速的识别、精确的自我定位和辨别方向等功能。在目标检测和识别中,提出了一种基于Harris角点检测的快速识别算法,实现对模型的检测识别。通过室内复杂环境下的实际实验和仿真实验,验证了该方法的可行性。实验结果表明,所提出的新型人工标识物在复杂环境下有足够的应用前景,复杂背景下的路标识别在不同视角下都是实时鲁棒的,可以满足工业机器视觉的精度要求。
Computer vision has gradually become popular, for the visual imaging is the projective geometry and the ratio is the basic invariant in projective geometry, this paper designs a new type of artificial markers based on cross ratio invariability, with identification of reliable detection and rapid and accurate self-localization function. In target detection and recognition, a fast recog- nition algorithm based on Harris corner detection is proposed to realize the detection and recognition of the model. The feasibility of the method is verified by practical experiments and simulation experiments in complex indoor environment. The experimental results show that the proposed new artificial logo has enough application prospects in complex environment, and landmark recognition un- der complex background is robust in different perspectives, which can meet the accuracy requirement of industrial machine vision.
作者
王海雯
乔嘉希
WANG Haiwen;QIAO Jiaxi(Department of Mathematics,Northeastern University,Shenyang 110819)
出处
《计算机与数字工程》
2018年第10期2106-2110,共5页
Computer & Digital Engineering
基金
国家级大学生创新创业训练计划资助项目(编号:201710145021)资助
关键词
交比不变量
人工标识物
目标识别
cross ratio invariant
artificial markers
target recognition
