Single Camera Calibration Method Based On Dual-mirror Imaging

Camera calibration is the process which calculates camera calibration parameters using object's two-dimensional image information and the real three-dimensional information. Getting precise camera calibration matrixes is the first step to achieve the binocular vision system. The camera calibration parameters can be used to restore the three-dimensional coordinates of the target object in order to achieve accurate positioning of the target object and it is widely used in practice.The dual-mirror with single camera calibration method is a new calibration method, it uses two mirrors and a camera to simulate two cameras through the mirror reflection. Two mirrors are fixed side by side, when the positions of the mirrors and the camera maintain certain relationship, we can get two images of the target object in one picture to simulate the effect of two cameras . Software system can be divided into two parts, calibration model selection and corner detection. In the part of calibration model selection, it proposes the model which only consider the first-order distortion coefficient based on the linear parameter model. In the part of corner detection, it detects the picture with Hough Line Detection Method in order to find the boundary line equation , finally, make the intersection of the boundary lines as the corner coordinates. Then get camera calibration parameters matrix and distortion parameters matrix after the corner detection with iterative method. Finally, reconstruction of the target object with back projection algorithm to verify the calibration parameters.This calibration method can reduce the cost of the visual system effectively and make the visual system be more widely used. The results show that the current system under the experimental conditions, the parameter matrixes obtained by the use of double-mirror camera calibration system can accurately restore the three-dimensional coordinates of the target object. The actual error controls below 0.3cm, which can better meet the needs of binocular vision system.