东北大学学报（自然科学版）

东北大学学报:自然科学版 ›› 2016, Vol. 37 ›› Issue (5): 614-618.DOI: 10.12068/j.issn.1005-3026.2016.05.002

• 信息与控制 • 上一篇    下一篇

基于改进随机蕨的增强现实场景实时跟踪注册算法

赵越， 李晶皎， 李海鹏， 杨丹   

  1. (东北大学 信息科学与工程学院， 辽宁 沈阳110819)
  • 收稿日期:2015-02-11 修回日期:2015-02-11 出版日期:2016-05-15 发布日期:2016-05-13
  • 通讯作者: 赵越
  • 作者简介:赵越(1979-)，女，辽宁抚顺人，东北大学博士研究生，渤海大学讲师; 李晶皎(1964-)，女，辽宁沈阳人，东北大学教授，博士生导师.
  • 基金资助:
    国家自然科学基金资助项目(60970157); 中央高校基础科研青年教师创新基金资助项目(N130404004).

Real-Time Tracking and Registration Algorithm of Scenarios of Augmented Reality Based on Improved Random Fern

ZHAO Yue， LI Jing-jiao， LI Hai-peng， YANG Dan   

  1. School of Information Science & Engineering， Northeastern University， Shenyang 110819， China.
  • Received:2015-02-11 Revised:2015-02-11 Online:2016-05-15 Published:2016-05-13
  • Contact: ZHAO Yue
  • About author:-
  • Supported by:
    -

RichHTML

0

PDF (PC)

772

摘要： 针对增强现实场景跟踪注册存在速度较慢等问题，提出了一种基于改进随机蕨的增强现实跟踪注册算法.该算法采用离线训练和在线跟踪两个模块.提出了一种嵌入式蕨分类器以提高特征点匹配精度，该分类器采用有监督的降维方法，并利用了所有可能的信息.通过该分类器进行特征匹配，进而计算摄像机位姿并渲染注册虚拟物体.实验结果证明，提出的嵌入式蕨在平均分类精度上优于其他算法.平均处理每帧图像的时间为34.22ms，基本满足实时性.

关键词: 随机蕨, 降维, 分类器, 跟踪, 增强现实

Abstract: Because of the slow speed of tracking and registration on scenarios of augmented reality， an algorithm was proposed based on the improved random fern on tracking and registration of augmented reality. The proposed algorithm has the offline training section and the online tracking section. A classifier based on the embedded fern was proposed. In the developed classifier， a supervised dimensionality reduction method and all possible information were used. What’s more， it was also used for the feature matching， and then camera pose was computed and virtual objects were rendered and registered. Experimental results show that the proposed algorithm is superior to other algorithms on the average classification accuracy. The average processing time of each frame is about 34.22ms， which can almost meet the real-time.

Key words: random fern, dimensionality reduction, classifier, tracking, augmented reality

中图分类号: