The calibration of a camera is an essential step for the metric reconstruction of the world. Intrinsic and extrinsic parameters of the camera are computed using a one-to-one correspondence between 3-D and 2-D coordinates of control points on the calibration target. Three-dimensional coordinates of control points are given in advance, and 2-D coordinates are detected on the image. The calibration target uses cross lines, circles, and a chessboard pattern to improve detection of the control points on the image. We propose an algorithm to automatically detect the control points on an image, especially for a chessboard pattern. Two symmetric properties of a chessboard pattern related to geometric and brightness distribution are used, and two concentric circles are used as probes for the effective use of the two properties. The two symmetric properties of the chessboard generate the candidates of control points. Finally, a cross-ratio of four points on a line is used for verification. The experimental results using images that vary in scale, pose, and illumination demonstrate the robustness of the proposed algorithm.