您的位置:首页  > 论文页面

3D立体图像的生成及深度信息的提取

发表时间:2013-02-28  浏览量:17101  下载量:8519
全部作者: 张芳萍,刘杰
作者单位: 北京交通大学计算机与信息技术学院
摘 要: 在对3D立体显示理论研究的基础上,采用佩戴快门眼镜的主动快门式立体显示的方法进行3D立体显示,采用图像拼接的方法生成了可采用NVIDIA 3D VISION立体幻境观看的3D立体图像。并在双目立体视觉理论的基础上,采用圆形特征平面标定板Zhang平面标定法进行摄像机标定,采用霍夫变换(Hough transformation, HT)提取目标立体图像对十字特征点;通过基于特征点的匹配得到匹配点对;采用三角测量被动测距的方法,恢复匹配的特征点的深度信息。通过实验将实际测量的深度信息和恢复的深度信息进行对比,验证计算深度信息的可靠性。最后给出搭建双目立体视觉系统的构建模块,根据面向对象的设计方法,在VC++环境下开发摄像机标定、特征提取、特征匹配、深度信息提取等相应算法的实现软件。
关 键 词: 计算机应用;摄像机标定;立体匹配;深度信息
Title: 3D stereo images generation and depth information extraction
Author: ZHANG Fangping, LIU Jie
Organization: Computer Science and Information Technology School, Beijing Jiaotong University
Abstract: Based on the studies of 3D stereoscopic display methods, the active shutter stereoscopic display method was used for 3D stereo vision in the paper. 3D stereo images were generated by stitching images and watched by NVIDIA 3D VISION stereo vision. A system of depth information extraction was proposed in this paper, which is based on the research of binocular stereo vision theory. With the system, the camera was calibrated wsing circle feature tablet and Zhang�s flexible new method. Feature point was selected and the circle centre point was gotten by Hough transformation (HT), matching double points was gotten with feature match and depth information was recovered by triangle measuring method. With the comparison of the recovery depth information and the measuring depth information, the experiment showed that the recovery method was reliable. With an object-oriented design methd in VC++, related algorithm software including camera calibration, feature extract, stereo matching and depth information extraction was developed.
Key words: computer application; camera calibration; stereo matching; depth information
发表期数: 2013年2月第4期
引用格式: 张芳萍,刘杰. 3D立体图像的生成及深度信息的提取[J]. 中国科技论文在线精品论文,2013,6(4):331-338.
 
0 评论数 0
暂无评论
友情链接