摘要
针对混合现实技术在识别标志物时易发生抖动,且识别过程易受到遮挡影响的问题,从标志物特征点提取与匹配的角度入手,改进混合现实技术对标志物的识别算法。通过构造尺度空间,结合加速稳健特征(speeded up robust features,SURF)算法提取特征点,对ORB(oriented FAST and rotated BRIEF)算法的特征点提取和匹配进行改进。改进后的算法在特征点匹配的过程中精度更高,比SURF算法提升了38.8%,比ORB算法提升了28.3%,有效地提高了目标识别的效率。结果表明:把改进后的算法运用在混合现实系统中,可以在标志物被遮挡50%时,成功把虚拟模型叠加在标志物上,解决了模型抖动的问题。
The mixed reality technique in the identification of landmarks is prone to jitter and the recognition process is susceptible to occlusion.Aiming at the problems with starting of the feature point extraction and matching of landmarks to improve the mixed reality based landmark recognition algorithm,by constructing the scale space,combined with the SURF algorithm to extract feature points,the feature point extraction and matching of the ORB algorithm was improved.The improved algorithm has higher accuracy in the process of feature point matching,38.8%higher than the SURF algorithm and 28.3%higher than the ORB algorithm,which effectively improves the efficiency of target recognition.The results show that the application of the improved algorithm to mixed the reality systems could successfully superimpose a virtual model on a marker when the marker was blocked by 50%of the area,and solved the problem of model jitter.
作者
张笑宇
汤汶
万韬阮
朱耀麟
武桐
ZHANG Xiaoyu;TANG Wen;WAN Taoruan;ZHU Yaolin;WU Tong(School of Electronics and Information, Xi’an Polytechnic University, Xi’an 710048, China;Faculty of Science and Technology,Bournemouth University,Poole BH12 5BB,United Kingdom;Faulty of Engineering and Informatics,University of Bradford,Bradford BD7 1DP,United Kingdom;School of Electronics and Information, Northwestern Polytechnical University, Xi’an 710072,China)
出处
《西安工程大学学报》
CAS
2020年第4期57-63,共7页
Journal of Xi’an Polytechnic University
基金
陕西省重点研发计划项目(2018SF-351)
陕西省教育厅服务地方科学研究计划项目(18JC012)
陕西省科技厅重点研发计划一般项目(2019GY-098)
榆林市科技局科创新城项目(2018-2-24)。
关键词
混合现实
ORB算法
特征识别
尺度空间
单应性矩阵
匹配精度
mixed reality(MR)
oriented FAST and rotated BRIEF(ORB)algorithm
feature recognition
scale space
homography matrix
matching accuracy
