Automatic Extrinsic Self-Calibration for Fusing Data from Monocular Vision and 3-D Laser Scanner

In this paper, an extrinsic self-calibration approach is proposed for solving the problem of real-time data fusion between monocular vision and 3-D laser scanner. A novel calibration board is designed for data fusion so that the extrinsic parameters can be obtained automatically by matching the corner features extracted from both vision and laser data. Experimental results obtained from both indoor and outdoor RGB-D scene reconstruction demonstrate the validity and good performance of the proposed approach.