基于平面法向量的线结构光视觉传感器参数标定

Parameter Calibration of Line Structured Light Vision Sensor Based on Plane Normal Vectors

为解决线结构光传感器精度低且易受环境干扰的问题,提出一种标定方法。此方法以棋盘格为靶标,使用一种新的角点快速定位算法进行相机标定;使用Steger算法提取图像激光中心点后结合Huber损失函数拟合出光条直线;通过相机内外参标定两个平面的法向量,两法向量叉乘后获得激光光条在相机坐标系下的点向式方程;对提取的多条激光光条进行等距取点后利用最小二乘拟合出激光平面。使用3种方法进行对比实验,验证了此方法的有效性,同时对长为7 mm的标准量块在此系统下进行测量,其平均测量误差为0.036 mm。与传统方法相比,此方法具有较高的精度与较好的鲁棒性,能够适应较复杂的环境。

In order to solve the problem of low precision and easy noise interference of line structured light sensor,a calibration method is proposed.This method took the checkerboard as the target,and used a new fast corner point positioning algorithm to calibrate the camera.Then used the Steger algorithm to extract the laser center point of the image,and then used the Huber loss function to fit the light line.The normal vector of the plane,the point-wise equation of the laser light bar in the camera coordinate system was obtained after the cross product of the two normal vectors.Finally,the laser light bar was equidistant from the extracted multiple laser light bars and the least squares were used to fit the laser plane.The effectiveness of this method is verified by comparative experiments with three methods.At the same time,a standard gauge block with a length of 7 mm is measured under this system,and the average measurement error is 0.036 mm.Compared with traditional methods,this method has higher accuracy and better robustness,and can adapt to more complex environments.