Circular coded target system for industrial applications

摘要

Coded targets are used as reference targets with a known location during camera calibration, for robust searching of corresponding features between images during various applications of machine vision like object tracking, robot navigation or 3D measurement. In this paper, a target system for industrial photogrammetric applications is outlined. The methods which have been chosen emphasize maximum robustness, along with accuracy at the expense of computational efficiency, since photogrammetric measurements are mainly evaluated offline. The outlined system combines widely used photogrammetric circular coded target design with an automatic library generator. It also utilizes robust methods of target detection and recognition with error correction (in case of 60 15-bit targets, up to 1 bit confused or 2 bits occluded) along with preserving the low false-positive or confusion rate. Any error correction method for this type of targets was not introduced before. The solution also allows for the creation of versatile target libraries or to work with the existing target libraries of a number of commercial photogrammetric systems. The properties of the target system were tested under challenging conditions (including heavy noise, blur, occlusion and geometrical image transformations) and compared to state-of-the-art systems, e.g. TRI-TOP (GOM), or ArUco, which it outperforms. The target system is already used for the camera calibration of a specialized photogrammetric system utilized in the heavy industry environment.