单目摄像机线性模型可靠性分析与研究

要利用摄像机进行高精度的非接触式测量,通常在线性模型的基础上采用单目摄像机进行光学标定、成像、校正和测量,其中常用的针孔相机模型的可靠性及适用范围鲜有报道,为此对目前广泛采用的线性模型的可靠性进行分析与研究。根据单目摄像机成像特点,建立基于图像像素坐标与三维世界坐标的摄像机数学模型,利用自制模板进行视觉测量实验,并通过亚像素角点检测方法得到图像像素坐标,在不考虑光学畸变的情况下,对模板线段进行特征提取和线性度分析,并不断调整测量距离进行重复实验。实际针孔相机模型测量结果显示相对误差不超过1%,在该模型下的重复测量误差最大不超过0.3 mm,并进一步提出了该模型在不同区域和不同物距下存在的测量差异。实验结果显示,线性模型简单实用且在一定物距范围内具有良好的线性度。实验结果为针孔相机模型的建立提供了参考。

Reliability analysis and research of monocular camera linear model

In order to use the camera for non-contact measurement with high accuracy, the monocular camera is usually used for optical calibration, imaging, calibration and measurement, on the basis of the linear model, but there is no analysis and research on the reliability of the commonly used linear model. Thus its reliability is studied and analyzed here. According to the characteristic of the monocular camera, and in order to build the camera model based on relationship of the image pixel coordinates and the three-dimensional world coordinates, the vision measurement experiment is completed by using our own custom template. Then, we get image pixel coordinates by the sub-pixel corner detection method. Neglecting the distortion, the feature extraction and linearity analysis are realized. Many repeated experiments are operated by constantly adjusting the measurement distance. The relative error of measurement by the linear model is less than 1%, and the measurement error is not more than 0.3 mm. Meanwhile, it is found that the model has measurement difference for different regions and object distances. The experiment results show this linear model is simple and practical, and has good linearity within a certain object distance range. These experiment results provide a powerful basis for establishment of the linear model. The work has the potential in practical engineering.