A Fitting Method of Central Catadioptric Line Images

Under central catadioptric camera, the projection of a line in space is a conic curve. Due to the partial occlusion, only a small arc of the conic curve is visible on the image plane, thus it is very difficult to correctly estimate the line image from its visible part, which affects the calibration accuracy of the camera. What’s more, the existing method in the literatures cannot still solve this problem effectively. In addition, according to the image formation of central catadioptric camera, we find that if the antipodal points of image points on the visible arc are known, the fitting accuracy can be improved greatly. In this paper, we propose a new catadioptric line fitting method, which is suitable to all kinds of central catadioptric cameras. Firstly, a new relationship between a pair of antipodal image points and the camera principal point are derived. Next, based on the relationship, a new objection function is established to estimate the line images. Finally, these estimated line images are used to calibrate camera intrinsic parameters to evaluate the performance of the fitting method. Experimental results on both simulated and real image data have demonstrated the effectiveness of the fitting method. That is, our method not only has good robustness but also can improve the fitting accuracy of central catadioptric line images and the calibration accuracy of the camera intrinsic parameters.