Automatic control point segmentation and localization for online camera calibration

Abstract

Camera calibration is an essential step for 3D object recognition and reconstruction. In this work, we propose a novel method for the automatic localization of control points and present a simple and yet robust online camera calibration system. First a planar calibration pattern with circular control points is designed. These points are then automatically segmented and localized by integrating information resulted from two complementary segmentation methods: region extraction and contour detection. After the calibration pattern has been shown to the system at a few locations, both the intrinsic and extrinsic camera parameters can be determined. The main advantage of this approach is that the metric measurement of the calibration pattern in the image plane is done purely automatically. We have implemented the whole system on a windows platform. Tests with USB cameras in different configurations and applications show the efficiency as well as the accuracy of the proposed approach.