Camera Self-Calibration Based on Multiple View Images

Camera calibration is an essential part of 3D reconstruction. Conventional calibration methods require high precision equipment and sophisticated operations. Compared with that, Camera self-calibration is quite simple but with low precision, which leads to significant performance degradation of 3D reconstruction. Therefore, a high precision self-calibration method with simple operations is necessary. In this paper, by using the bundle adjustment algorithm and SIFT points matching relationship, a local-global hybrid iterative optimization method is proposed. Considering large number of matching features, a neighborhood image matching method is proposed that can significantly reduce the matching time under the premise of maintaining accuracy. Experimental results show that the proposed method is an effective method for high accuracy, and it can reduce the time consumption of image matching. Based on the relationship between corresponded matching points in multiply view images, our method makes full use of the local-global hybrid idea to compute the parameters of camera. Compared with existed other methods, it is more robust with higher precision.