WSEAS Transactions on Computers
Print ISSN: 1109-2750, E-ISSN: 2224-2880
Volume 14, 2015
Novel Planar Self-Calibration Approach of a Camera having Variable Intrinsic Parameters
Authors: , , ,
Abstract: The self-calibration of a camera is an important computer vision problem it allows us to estimate the camera intrinsic parameters without any knowledge of the used scene. In this work we proposed a novel method of camera self-calibration with varying intrinsic parameters from an unknown planar scene. The principle of the present approach consists of using some projection points of the scene in only two image planes to obtain a system of equations according to the intrinsic camera parameters and the image of absolute conic in the both views. From this system of equations we formulated a non-linear cost function whose minimization permits us to estimate the intrinsic camera parameters in the both images. The experimental results on synthetic and real data justify the effectiveness of our method.
Search Articles
Keywords: Self-calibration, Computer vision, Varying intrinsic parameters, Image of the absolute conic, nonlinear cost function, Unknown planar scene
Pages: 444-454
WSEAS Transactions on Computers, ISSN / E-ISSN: 1109-2750 / 2224-2880, Volume 14, 2015, Art. #45