Calibration of central catadioptric cameras based on veronese mapping by using checkerboard
摘要
The use of primitive geometric elements to calibrate central catadioptric cameras has become an important research topic in computer vision. By extending the 3D spatial coordinates and their corresponding image coordinates through a Veronese map, a projection transformation matrix between the extended spatial point and its image can be estimated by the direct linear transformation method. From this matrix, we can calculate the camera intrinsic parameters and the mirror parameter through the singular value decomposition method. Using a checkerboard as the calibration template, both simulated and real experiments were performed to verify the validity of our calibration method.