›› 2014, Vol. 27 ›› Issue (5): 119-.

• 论文 • 上一篇    下一篇

基于FPGA的增强现实人工标识识别

李聪   

  1. (西安电子科技大学 电子工程学院,陕西 西安 710071)
  • 出版日期:2014-05-15 发布日期:2014-05-14
  • 作者简介:李聪(1987—),女,硕士研究生。研究方向:增强现实,基于FPGA的嵌入式系统等。E-mail:309630611@qq.com

Artificial Mark Recognition in Augmented Reality Based on FPGA

LI Cong   

  1. (School of Electronic Engineering,Xidian University,Xi'an 710071,China)
  • Online:2014-05-15 Published:2014-05-14

摘要:

基于标志点的增强现实系统中,需要将计算机生成的虚拟物体注册到对应的标志点上,以达到对现实的增强效果,因此对标识的识别成为该系统的关键。文中设计了以FPGA为核心的人工标识识别系统,采用TRDB-D5M摄像头和DE2-115 FPGA开发板设计方案,实现了对高速图像的视频采集,并通过对含有人工标识的彩色图像进行二值化处理、连通域分析和模板匹配,实现了对人工标识的识别。通过实验验证和分析,表明该方法能够准确的识别无遮挡的标识,对增强现实移动系统的进一步完善具有实际意义。

关键词: FPGA, 增强现实, 人工标识, 二值化处理, 模板匹配

Abstract:

In a mark points based augmented reality system,one need to register a computer-generated virtual object to the corresponding mark point in order to achieve the effect of augmented reality,which makes mark recognition the key process of the system.An artificial mark recognition system with FPGA as the core is designed in the paper.Video capture is implemented by adopting TRDB-D5M camera and DE2-115 FPGA development board;then a binary processing,an algorithm based on the mark abstraction of connectivity area and a method based on template match are adopted on the color image with marks,which realizes artificial mark recognition.The experiments results and analysis show the method can recognize the marks accurately and has important practical significance for the further development of the augmented reality system.

Key words: FPGA;augmented reality;artificial mark;binary processing;template match

中图分类号: 

  • TP391.41