摘要
针对传统的增强现实(AR,augmented reality)三维注册算法耗时较长,结果准确率和鲁棒性低的问题,从高鲁棒性的角度出发,对AR三维注册算法进行改进。算法的实现过程共分为位置初推导、特征信息匹配与算法建立等3个步骤。在完成摄像头定位跟踪并确定虚拟目标在真实世界中的位置后,对特征点信息进行匹配。在此基础上,通过算法初始化和在线化过程,结合KLT追踪实现三维注册。实验结果表明,改进后的AR三维注册算法在耗时、特征点位置信息计算结果准确率和鲁棒性等方面明显优于传统算法。
Aiming at the problem that the traditional augmented reality 3D registration algorithm takes a long time,the accuracy of calculation results and low robustness,a new high robust augmented reality 3D registration algorithm is proposed.The implementation process of the algorithm is divided into three steps:initial position derivation,feature information matching and algorithm establishment.After the camera positioning tracking is completed and the position of the virtual target in the real world is determined,the feature point information is matched;on this basis,the three-dimensional registration is implemented by KLT tracking through algorithm initialization and online process.The experimental results show that the performance of the augmented reality 3D registration algorithm in terms of computation time,accuracy of the calculation results of feature point location information and performance robustness is obviously due to the traditional algorithm.
作者
黄珍
潘颖
HUANG Zhen;PAN Ying(School of Digital Media, Lanzhou University of Arts and Science, Lanzhou 730000,China)
出处
《西安工程大学学报》
CAS
2019年第6期679-684,共6页
Journal of Xi’an Polytechnic University
基金
甘肃省自然科学基金项目(17JR5RA007)
2019年甘肃省高等学校创新能力提升项目(2019B-194)
兰州文理学院种子基金(自然)项目(17XJZZ005)
关键词
高鲁棒性
增强现实
三维注册
KLT追踪
耗时
信息匹配
high robustness
augmented reality
3D registration
KLT tracking
time consuming
information matching
