Authors: Hamdi Boukamcha; Mohamed Atri; Fethi Smach
Addresses: Laboratory of Electronics and Microelectronics, Department Physique, Faculty of Science of Monastir, University of Monastir, Environment Street, Monastir, 5019, Tunisia ' Laboratory of Electronics and Microelectronics, Department Physique, Faculty of Science of Monastir, University of Monastir, Environment Street, Monastir, 5019, Tunisia ' University of Burgundy, Esplanade Erasme, Dijon, 21078, France
Abstract: Calibration of the internal and external parameters of a stereo vision camera is a well-known research problem in the computer vision. Usually, to get accurate 3D results the camera should be manually calibrated accurately as well. This paper proposes a robust approach to auto calibration stereo camera without intervention of the user. There are several methods and techniques of calibration that have been proven, in this work, we exploiting the geometric constraint, namely, the epipolar geometry. We specifically focus to use seven techniques for features extraction (SURF, BRISK, FAST, FREAK, MinEigen, MSERF, and SIFT), however, tries to establish the correspondences between points extracted in stereo images with various matching techniques (SSD, SAD, Hamming). Then we exploit the fundamental matrix to estimate the epipolar line by choosing the perfect eight-point algorithms (Norm8Point, LMedS, RANSAC, MSAC, and LTS). A large number of experiments have been carried out, and very good results have been obtained by comparison and choice of the perfect technique in every stage.
Keywords: auto calibration; robust matching; epipolar geometry; fundamental matrix; matching technique.
International Journal of Computer Aided Engineering and Technology, 2020 Vol.12 No.1, pp.74 - 94
Received: 27 May 2017
Accepted: 29 Jun 2017
Published online: 27 Nov 2019 *