Chapter 1 A New Method for Improving Measurement Accuracy of Digital Image Correlation Li Bang-Jian, Wang Quan-Bao, and Duan Deng-Ping Abstract Digital image correlation (DIC) method has been applied in wide fields including experimental mechanics. The displacement measurement accuracy plays an important role in these situations. The direct method to improve the measurement accuracy is to reduce the measurement error in DIC. In the paper, a new method has been developed to improve the DIC accuracy using the feature of DIC system error. The feature of DIC system error is analyzed. And the reduction of the DIC system error has been verified by experiment. Keywords Digital image correlation • System error • Translation measurement • Error reduction • Measurement accuracy 1.1 Introduction Digital image correlation (DIC) is an optical measurement method and DIC is popular in motion measurement, dimension measurement and experimental mechanics for its countless and full-field features [1–5]. In 2D DIC, a digital camera is used to capture the reference image before the specimen deformation and capture the deformed image when the loading is imposed by drawing machine. In motion measurement, the reference and deformed images are captured before and after the specimen or object movement respectively. In dimension measurement, the same speckle images in different positions are regarded as the reference and deformed images respectively. And in subset based DIC, the displacement of sampled point can be obtained by matching the selected subset. Subsequently the dynamic variables including velocity and acceleration can be calculated by post-processing with the displacement data. That is to say that displacement or translation measurement is the basic for other variables’ calculation. So the translation measurement of DIC has been studied by many researchers [6–8]. And the measurement accuracy of DIC has been one main research point recently [9, 10]. In addition, the DIC calculation speed has been another main research point. For calculation speed, many techniques are proposed. Pan [11] proposed that reliability guided searching method combined with interpolation coefficient pre-computed technique can make the calculation speed about 160 times faster than the classical DIC method. Jiang et al. [12] presented that path-independent searching method combined with graphics processing unit (GPU) based parallel calculation, inverse compositional Gauss Newton (IC-GN) [13] and Fast Fourier transform (FFT) can make about 66 times faster than non-path-independent method. For measurement accuracy, many researchers have studied it by theory, simulation and experiments. Generally measurement accuracy can be quantified with measurement error which contains system error and random error by theoretical and experimental analysis. For instance, Wang et al. [14] investigated the shape function induced random error theoretically and experimentally. Schreier et al. [15] studied the interpolation induced system error by simulation. Pan et al. [8] investigated lens distortion induced system error by theory and experiments. Ma et al. [7] researched the camera self-heating induced system error by experiments. Su and Xu et al. [16–18] derived the system error theoretically. Hu et al. [19] developed a method to evaluate 3D DIC error. Pan et al. [20, 21] used bilateral telecentric lens to improve the 2D DIC measurement accuracy. Considering the limitation of field of view of bilateral telecentric lens, Pan et al. [22] used a non-deformable sample to compensate the DIC measurement accuracy. Pan [23] and Zhou et al. [24] proposed that image pre-filtering in DIC can enhance measurement accuracy by smoothing the image noise but the pre-filtering parameters had to be selected and the random error increased due to the pre-filtering smoothing function. Zhou et al. [25] developed an adaptive subset offset method in incremental DIC to improve the measurement accuracy by avoiding interpolation of the sample points. L. Bang-Jian • W. Quan-Bao ( ) • D. Deng-Ping School of Aeronautics and Astronautics, Shanghai Jiao Tong University, Shanghai, People’s Republic of China e-mail: quanbaowang@sjtu.edu.cn © The Society for Experimental Mechanics, Inc. 2018 L. Lamberti et al. (eds.), Advancement of Optical Methods in Experimental Mechanics, Volume 3, Conference Proceedings of the Society for Experimental Mechanics Series, DOI 10.1007/978-3-319-63028-1_1 1
RkJQdWJsaXNoZXIy MTMzNzEzMQ==