论文部分内容阅读
摄像机标定的一个重要问题是既要保证参数估计的精确性,又要满足旋转矩阵正交约束条件.本文在线性优化和遗传算法的基础上提出了一种新的双段标定方案.第一阶段,在忽略畸变参数的情况下,利用最小二乘法估计摄像机内外参数.为减少畸变影响,本阶段仅考虑图像中心区域的标定点;第二阶段,以第一阶段的估计值为初始值,并考虑摄像机畸变的影响,根据整个图像范围的标定点,利用遗传算法优化摄像机的所有参数以获得精确解.由于该算法直接优化摄像机的旋转角度,所以本文的标定方案能够容易地满足旋转矩阵正交约束条件.此外,采用遗传算法也简化了摄像机标定过程
An important problem of camera calibration is that it not only guarantees the accuracy of parameter estimation but also satisfies the orthogonal constraint of rotation matrix.This paper proposes a new two-stage calibration scheme based on linear optimization and genetic algorithm.The first stage , Ignoring the distortion parameters, using the least square method to estimate the camera internal and external parameters.To reduce the distortion effect, this stage only consider the calibration point of the image center area; the second stage, the first stage of the estimated value as the initial value and Considering the influence of camera distortion, using genetic algorithm to optimize all the parameters of the camera to obtain the exact solution, based on the calibration points of the whole image range, the proposed calibration scheme can easily satisfy the orthogonal rotation matrix because the algorithm directly optimizes the rotation angle of the camera. Constraints In addition, the use of genetic algorithms also simplifies the camera calibration process