Research on Improved Microscope Calibration Method Based on Coplanar Points.

Camera calibration is the premise and basis for computer vision to obtain three-dimensional space information. Its calibration accuracy directly affects the accuracy of three-dimensional measurement and the results of three-dimensional reconstruction. However, the stereo light microscope has the problems of small depth of field and complex optical path, and the traditional macroscopic calibration method cannot be applied to microscopic calibration. In order to solve this problem, this paper proposes an improved microscope calibration method based on coplanar points. Firstly, the method of varying focal length is used to obtain the camera principal point coordinates. By shooting the same target point at different magnifications, the image coordinates are obtained respectively, then the least square method is used to fit the line, and the intersection of the fitted line is the principal point of the image. Then, based on the coplanar point method, the analytical solution of the camera scale factor is obtained, which improves the robustness of the initial value estimation of the internal parameters. Finally, the initial value estimation of internal parameters is optimized by nonlinear least square method, and the tangential distortion is considered as a factor of nonlinear optimization, so as to obtain accurate microscope calibration parameters. The experimental results show that compared with the traditional Zhang’s calibration method, the improved microscope calibration method based on coplanar points has higher calibration accuracy and is suitable for microscope calibration.