Calibration method for binocular vision with large FOV based on normalized 1D homography

Abstract Calibration method for binocular vision with large field of view (FOV) is the key technology in visual measurement system. The method based on 1D targets has been widely studied due to its high flexibility and convenience. In this paper, a calibration method based on normalized 1D homography matrix is proposed for robust camera internal parameters calculation with 1D target. Furthermore, based on combined one-dimensional targets, the calibration method for external parameter of binocular vision system is proposed. The numerical simulation is carried out to verify our method and to analyze the noise sensitivity, and the real data experiments are also developed. The results show that the internal parameter based on the normalized homography matrix is better than 0.1mm when the noise level is 2, and the precision of the measurement system is about 0.1mm in the depth direction of 3m. The method based on 1D target in this paper is effective and practical in the calibration of large FOV measurement system.