Affine reprojection hybrid virtual-reality registration method

YONG Jiu; WANG Yang-ping; LEI Xiao-mei

doi:10.13873/J.1000-9787(2016)09-0049-05

Chinese

您当前的位置：

首页 >

文章列表页 >

Affine reprojection hybrid virtual-reality registration method

更新时间：2025-09-11

- Affine reprojection hybrid virtual-reality registration method
- Vol. 35, Issue 9, Pages: 49-53(2016)
- 作者机构：
  
  1. 兰州交通大学电子与信息工程学院
  2. 兰州宇信信息技术有限责任公司
- 作者简介：
- 基金信息：
- DOI：10.13873/J.1000-9787(2016)09-0049-05
  CLC： TP391.41
- Published：2016
- 稿件说明：
移动端阅览
[1]雍玖,王阳萍,雷晓妹.仿射重投影混合虚实注册方法[J].传感器与微系统,2016,35(09):49-53.

YONG Jiu, WANG Yang-ping, LEI Xiao-mei. Affine reprojection hybrid virtual-reality registration method[J]. 2016, 35(9): 49-53.
[1]雍玖,王阳萍,雷晓妹.仿射重投影混合虚实注册方法[J].传感器与微系统,2016,35(09):49-53. DOI： 10.13873/J.1000-9787(2016)09-0049-05.

YONG Jiu, WANG Yang-ping, LEI Xiao-mei. Affine reprojection hybrid virtual-reality registration method[J]. 2016, 35(9): 49-53. DOI： 10.13873/J.1000-9787(2016)09-0049-05.

摘要

针对特征图形结合自然纹理特征的虚实注册误差较大、运行速度慢的问题

提出仿射重投影混合虚实注册方法。通过建立摄像机、真实场景以及特征坐标系之间对应关系的仿射坐标系

且采用ORB算法进行特征检测及后续特征模板的构造

并利用仿射重投影矩阵来恢复特征模板在当前图像上的投影位置信息以计算虚实注册矩阵。采用Open GL生成的立方体模型与真实场景之间进行虚实注册仿真实验。仿真结果表明:在特征图形被遮挡或不可见时

本文方法在精度、健壮性以及速率方面表现优异

提高了增强现实（AR）系统的实时性和鲁棒性。

Abstract

In order to solve the problem of large error and slow running speed of virtual-reality registration method combining feature marker with natural texture feature

affine reprojection hybrid virtual-reality registration method is proposed.Through establishing affine coordinate system with corresponding relationship between camera

real scene and feature coordinates

and using oriented FAST and rotated BRIEF（ORB） algorithm for feature detection and construction of subsequent feature template

and use affine reprojection matrix to restore the projection position information of feature template on current image to calculate virtual-reality registration matrix.Simulation is carried out using Open GL generated cube model and real scene.Simulation results certify when the feature marker is occluded or invisible

the proposed registration method is superior in precision

robustness and rate;and real-time and robustness performance of augmented reality（AR） system are improved.

关键词

Keywords

references

移动增强现实三维注册中特征提取算法改进 [J]. 王文润,王阳萍. 计算机仿真 . 2015(12)

面向增强现实的多传感器组合位姿注册算法 [J]. 王崴,彭勃宇,周诚,瞿珏,刘晓卫. 传感器与微系统 . 2015(07)

二维码在移动增强现实中的应用研究 [J]. 桂振文,王涌天,刘越,陈靖. 计算机辅助设计与图形学学报 . 2014(01)

增强现实混合跟踪技术综述 [J]. 罗斌,王涌天,沈浩,吴志杰,刘越. 自动化学报 . 2013(08)

一种基于SURF与ARToolkit的增强现实设计 [J]. 陈荣军,朱雄泳,谭洪舟,陈奇彦. 电子技术应用 . 2012(10)

增强现实研究进展与核心技术(上) [J]. 陈靖,王聪. 信息技术与标准化 . 2012(06)

模板匹配技术在图像识别中的应用 [J]. 田娟,郑郁正. 传感器与微系统 . 2008(01)

Improvement of registration accuracy of a handheld augmented reality system for urban landscape simulation [J] . Tomohiro Fukuda,Tian Zhang,Nobuyoshi Yabuki. Frontiers of Architectural Research . 2014

Speeded-Up Robust Features （SURF） [J] . Herbert Bay,Andreas Ess,Tinne Tuytelaars,Luc Van Gool. Computer Vision and Image Understanding . 2007 (3)

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Application of ORB algorithm based on K-means in recognition of front vehicle

Defect detection method of carton based on block feature matching

Image matching algorithm based on improved ORB and grid-based motion statistics

Related Author

XIAO Bin

YANG Tao

MEI Yanying

LUO Shoupin

YANG Tao

MEI Yanying

Related Institution

School of Information Engineering,Southwest University of Science&Technology

Key Laboratory of Sichuan Province for Robot Technology Used for Special Environment

School of Information Engineering,Southwest University of Science＆Technology

Key Laboratory of Sichuan Province for Robot Technology Used for Special Environment

⁰