Calibration of stereo-digital image correlation for large field of view measurement based on photogrammetry

Abstract

Calibration of a stereo-digital image correlation (stereo-DIC) system is essential for three-dimensional (3D) shape and deformation measurement. Although the traditional planar calibration method is flexible in most application scenarios, it still has difficulties in large field of view (FOV) calibration due to the limited size of calibration board. In this paper, a stereo-vision calibration method is proposed for large FOV measurement based on the close-range photogrammetry. Specifically, a certain number of coded targets are first arranged in the area to be measured. Then, the left and right cameras are used to capture images at the calibration and measurement positions. With these images, accurate camera intrinsic parameters can be obtained by close-range photogrammetry methods through encoded targets. The extrinsic parameters can be obtained by the 3D coordinates of these corresponding points and further optimized using corresponding points obtained from both coded targets and speckle patterns. Laboratory experiments with field-of-view ranges of 25 cm by 25 cm and 2.8 m by 2.8 m demonstrate that this method can obtain precise calibration parameters and has good deformation measurement accuracy. Compared to traditional planar calibration methods, this method does not require high-quality or large-sized calibration objects, and all parameters can be calibrated on-site. A large shaking table experiment demonstrated the potential of this method in the field of engineering measurements.