Calibrating stereo-digital image correlation system using synthetic speckle-pattern calibration target
Calibrating stereo-digital image correlation system using synthetic speckle-pattern calibration target
复制标题
使用合成散斑图案校准目标校准立体数字图像相关系统
DOI：
10.1088/1361-6501/ab8dfb
复制
发表时间：
2020-09-01
影响因子：
2.4
通讯作者：
Pan, Bing
中科院分区：
文献类型：
--
作者：
Chen, Bin;Liu, Haibo;Pan, Bing
我们描述了一种立体数字图像相关（stereo- dic）的替代立体校准方法，该方法使用计算机模拟散斑模式校准目标和基于dic的控制点匹配技术。本文将最近提出的摄像机标定方法改进和扩展到立体标定，立体标定是立体dic测量的重要组成部分。在所提出的校准方法中，通过在商用平板上显示合成散斑图案来生成校准目标。在模拟的散斑图上，指定一组具有灵活位置和数量的控制点，并使用成熟的基于子集的DIC算法与捕获的校准图像进行精确匹配。利用这些精确匹配的控制点，利用立体标定算法对立体- dic系统的内外参数进行标定。为了获得更好的精度，还考虑了校准目标制造过程中的误差并进行了补偿。由于控制点匹配数量灵活，匹配精度高，鲁棒性强，因此采用合成散斑模式标定目标的立体标定方法比基于棋盘目标和圆形点目标等指定特征数量有限的标定目标具有更高的标定精度。通过一组模拟标定试验验证了所提出的立体标定方法，并将其应用于实际立体- dic系统的标定，用于三维全场形状和位移的测量。结果证实了所提出的立体校正方法在立体dic应用中的有效性和准确性。
We describe an alternative stereo calibration method for stereo-digital image correlation (stereo-DIC) using computer-simulated speckle-pattern calibration targets and a DIC-based control points matching technique. The present work refines and extends the recently proposed camera calibration method to stereo calibration, which is an essential part of stereo-DIC measurement. In the proposed calibration method, the calibration target is generated by displaying a synthetic speckle pattern on a commercial tablet. On the simulated speckle pattern, a group of control points is specified with flexible location and number, which is further precisely matched to the captured calibration images using the well-developed subset-based DIC algorithm. These precisely matched control points are subsequently used to calibrate the intrinsic and extrinsic parameters of the stereo-DIC system with stereo calibration algorithm. For better accuracy, the error in the manufacture of the calibration target is also considered and compensated. As a flexible number of control point matches are registered with high accuracy and strong robustness, the presented stereo calibration method using synthetic speckle pattern calibration target offers higher calibration accuracy than that based on a calibration target with a limited number of designated features, such as checkerboard target and circular point target. After being demonstrated by a set of simulated calibration tests, the proposed stereo calibration method was applied to calibrate a real stereo-DIC system, which was then used to measure 3D full-field shapes and displacements. The results confirm the effectiveness and accuracy of the proposed stereo calibration method for stereo-DIC applications.