Stereo calibration method using mirrored images containing a camera

This paper proposes a stereo calibration method using mirrored images that contain a camera and measurement objects. In stereo methods, the measurement accuracy can be seriously affected by rotational errors in the calibration between cameras. Our method performs the calibration by using the geometrical relationship between a reference point on the camera and the reference point in its mirrored image. Compared to conventional stereo methods, this method greatly improves the accuracy of stereo measurements. Additionally, our method does not require a time-consuming setup, and it works in an environment where only a few feature points can be obtained. By performing experiments, we evaluate the measurement accuracy of our method and compare it with a conventional method.