Line Structured-Light Vision Sensor Calibration Based on Multi-Tooth Free-Moving Target and Its Application in Railway Fields

In the railway detection systems, the existing online calibration of operational outdoor structured-light vision sensors usually encounters complex lighting, camera equipped with optical filters and large fields of view (FOVs) etc. These problems seriously limit calibration precision and efficiency. A novel onsite line structured-light vision sensor calibration method independent of target reference feature points and which utilizes a multi-tooth free-moving target is proposed. The intersection points are determined as feature points by using a laser irradiated on the target, and the feature points in the target coordinate system are established on the basis of cross-ratio invariant. Then, the 3D coordinates of feature points in the camera coordinate system are calculated by combining the intrinsic parameters. Meanwhile, the method based on the uncertainty model is used to optimize the location deviation of the feature points after compensation, and the equation of the laser plane is finally obtained. High-accuracy calibration of outdoor line structured-light vision sensor can be achieved when moving freely even without the removal of the optical filter and the use of any auxiliary equipment. The measurement accuracy of our method is equivalent to that of other methods using planar target in ideal imaging conditions. In addition, our method can be applied to online wheel-size measurement system and can meet the calibration and measurement requirements in complex onsite railway environment.