摘要
针对传统视觉基础矩阵计算中存在的噪音和误匹配问题,提出了基于射影空间的基础矩阵计算方法.首先定义了三视几何中射影标准基下基础矩阵只含有5个参数的特殊形式,并利用三视射影重建中空间点反投影图像误差最小为准则,消除了图像中误匹配的影响,然后基于RANSAC(random samp ling consensus)技术寻找出最优7个匹配点(噪音最小)来进行对极几何估算.大量仿真模拟试验和真实图像表明此方法能够高精度地估计出基础矩阵.
Image noise and outlier always exist in the estimation of fundamental matrix. As a result, a new method is proposed based on projective space. First, a special fundamental matrix is defined including only 5 parameters according to the projective standard basis of threo-view geometry. The method can cope with mismatch and search for the best 7 correspondencc points to estimate epipolar geometry based on RANSAC( random sampling consensus)technique according to the principle of the minimal re-projection error. Many experiments with both synthetic and real images show that our method can obtain fundamental matrix with high accuracy.
出处
《机器人》
EI
CSCD
北大核心
2005年第6期545-549,共5页
Robot
基金
国家863计划资助项目(2001AA421160)
关键词
基础矩阵
射影重建
RANSAC
误匹配
fundamental matrix
projective reconstruction
RANSAC( random sampling consensus)
outlier