摘要
提出一种基于摄像机视角的立体视觉定位方法,避免了摄像机标定环节,直接利用摄像机视角参数分析摄像机在不同位置对同一目标拍摄两幅图像,根据图像中目标点的象素坐标定位出该点在实际三维空间中的坐标。与传统方法相比,该方法没有利用摄像机的焦距而是利用了视角信息,计算简单,从而使定位工作更简便。
Based on the view angle of the camera, a binocular vision localization method was proposed. To avoid the camera calibration, the two images were analyzed from difference locations using the camera view angles and the actual coordinates were calculated in a three-dimensional space through the coordinates of target point pixel in the image. Compared with traditional methods, the view angles rather than the focal length were used using this method. The results show that calculation is very simple and thus it makes the location very convenient.
出处
《中南大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2009年第S1期80-83,共4页
Journal of Central South University:Science and Technology
基金
国家自然科学基金资助项目(90820302
60805027)
高等学校博士学科点专项科研基金项目(200805330005)
关键词
视角
立体视觉
定位
view angle
binocular vision
localization
