A new technique of camera calibration based on X-target

Abstract

A new technique of camera calibration based on X-target is proposed. Calibration steps include: judging gray saturation, adjusting verticality between optical axis and object surface, two-dimensional plane camera calibration. In the calibration process, the improved Harris operator and spatial moment are used to detect sub-pixel X-target corners, and the accuracy achieves up to 0.1 pixels. It does not need to calculate the specific internal and external camera parameters, and only calculates the relations of world coordinates and image pixel coordinates, and the distortion model. The method has better accuracy and stability, and has been applied in the industrial field of embedded machine vision. © 2010 Springer-Verlag.