摘要
利用计算机图形学、模式识别等技术,计算机视觉得到了迅速发展。摄像机标定是从二维图像获取三维信息必不可少的步骤,目前,非线性标定已成为计算机视觉领域的研究热点之一。利用计算机视觉中外部参数集合是旋转矩阵的性质,给出一种外部参数标定的一个保结构算法。把求外部参数问题转化为求基于特殊正交群所组成的流形上的优化问题。这种方法与以前的方法不同的是通过流形上的测地线搜索来求解,使所求解正交性保持不变,并且明显的减少了迭代的步数。
Computer vision techniques are developed with the computer graphics and pattern recognition technology.The calibration of cameras is an indispensable part to obtain 3D geometric information from 2D images.The calibration of a nonlinear parameters calibration has become one of the major research areas in computer vision technology.By using the properties that the outer parametric set is orthogonality matrices,a structure preserving algorithm for outer parametric in computer vision is presented.The outer parameters estimate problem in computer vision is formulated as an optimization over special orthogonal matrices of the manifolds.A key feature of the pro-posed approach,not used in earlier studies,is a geodesic search,which ensures the orthogonality of optimization solutions and significantly decreasing the iterations.
出处
《计算机工程与设计》
CSCD
北大核心
2009年第9期2336-2338,共3页
Computer Engineering and Design
基金
国家自然科学基金项目(60703118)
关键词
计算机视觉
参数标定
保结构算法
特殊正交矩阵
测地线搜索
computer vision
parameters calibration
structure preserving algorithm
special orthogonal matrices
geodesic search
