分析了基于随机抽样检验思想的现有鲁棒算法在基本矩阵的求解问题中存在的不足,提出一种获得基本矩阵最优解的算法.利用各种鲁棒技术获得内点集,以点到极线的距离作为最优量度标准,采用贪心策略在内点集中寻找最优子集,并利用最优子集来计算基本矩阵.合成数据与真实图像的实验结果表明,该算法在基本矩阵的求解精度、抗噪声能力、对极点的稳定性等方面优于现有的鲁棒方法.
By analyzing the shortcoming of existing robust algorithms based on random sampling employed in estimating the fundamental matrix, a novel algorithm is proposed for optimal estimation of fundamental matrix. The algorithm firstly uses some robust techniques to construct an inliers set of matching points. Then it takes the epipolar distance as the optimal criterion and search optimal subset in the inliers set under the greedy strategy. Finally, the fundamental matrix is calculated with the optimal subset. Experimental results on real image and synthetic data show that the proposed algorithm is superior over other robust methods in terms of estimation accuracy, anti-noise ability and stability of epipoles.