首页 | 本学科首页   官方微博 | 高级检索  
     

基于单幅图像的三维信息提取
引用本文:刘日晨,刘怡光,尹鑫,赵亦明.基于单幅图像的三维信息提取[J].计算机科学,2010,37(5):243-246.
作者姓名:刘日晨  刘怡光  尹鑫  赵亦明
作者单位:1. 四川大学计算机学院,成都,610065;四川大学合成视觉图形图像技术国防重点学科实验室,成都,610065
2. 四川大学计算机学院,成都,610065
基金项目:国家自然科学基金(60705005);;博士点新教师基金(20070610031);;四川省应用基础研究计划(2008JY0038);;教育部新世纪人才支持计划(NCET-08-0370)资助
摘    要:针对单幅图像提取三维信息困难,设计了一种基于单幅图像的三维信息提取算法。该算法首先需要标定摄像机,得到摄像机从三维到二维的成像矩阵M3×4,并计算M3×4的伪逆矩阵M+4×3,其本质是通过最小二乘法求解矛盾方程组;利用M+4×3提取二维单幅图像中的三维信息,再把这些三维信息投影到一个平面上显示出来。实验表明,单幅图像目标物体必须满足共面等约束条件,从而可以对目标物体的三维信息做定性分析(定性视觉)。此外,还对三维空间投影到二维图像的过程进行了理论推导,得出了如果没有足够的先验知识,成像过程不可逆的结论。

关 键 词:伪逆矩阵  三维信息提取  单幅图片  定性视觉  
收稿时间:2009/6/26 0:00:00
修稿时间:2009/9/12 0:00:00

3D Information Extraction Algorithm Based on a Single Image
LIU Ri-chen,LIU Yi-guang,YIN Xin,ZHAO Yi-ming.3D Information Extraction Algorithm Based on a Single Image[J].Computer Science,2010,37(5):243-246.
Authors:LIU Ri-chen  LIU Yi-guang  YIN Xin  ZHAO Yi-ming
Affiliation:College of Computer Science/a>;Sichuan University/a>;Chengdu 610065/a>;China;Key Laboratory of Fundamental Synthetic Vision Graphics and Image for National Defence/a>;China
Abstract:A 3D information extraction algorithm based on a single image was designed in this paper for the difficulties of this research field,in this algorithm,camera imaging matrix M3×4 was obtained by camera calibration firstly,the pseudo inverse matrix M4+×3 was calculated by M3×4,the essence of this computation was solving inconsistent equations by least squares approximation,3D information would be extracted from a single image through M4+×3,the result was shown in a image by projection operator.The result could be used for qualitative analysis(qualitative vision)currently due to the limited information from a single image.The experiment indicates that the extracted object should be coplanar.Moreover,the essential theories of 3D to 2D's projection were deduced,the conclusion of the deduce is that imaging process is an irreversible process.
Keywords:Pseudo inverse matrix  3D information extraction  Single image  Qualitative vision  
本文献已被 CNKI 万方数据 等数据库收录!
点击此处可从《计算机科学》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号