Digital image correlation (DIC) has become a valuable tool in the evaluation of mechanical experiments, particularly fatigue crack growth experiments. The evaluation requires accurate information of the crack path and crack tip position, which is difficult to obtain due to inherent noise and artefacts. Machine learning models have been extremely successful in recognizing this relevant information given labelled DIC displacement data. For the training of robust models, which generalize well, big data is needed. However, data is typically scarce in the field of material science and engineering because experiments are expensive and time-consuming. We present a method to generate synthetic DIC displacement data using generative adversarial networks with a physics-guided discriminator. To decide whether data samples are real or fake, this discriminator additionally receives the derived von Mises equivalent strain. We show that this physics-guided approach leads to improved results in terms of visual quality of samples, sliced Wasserstein distance, and geometry score.

翻译：数字图像相关（DIC）已成为评估力学实验（尤其是疲劳裂纹扩展实验）的重要工具。该类评估需要精确的裂纹路径和裂纹尖端位置信息，但由于固有噪声和伪影的存在，这类信息的获取十分困难。机器学习模型在基于标注DIC位移数据识别这些相关信息方面已取得极大成功。为训练具有良好泛化能力的鲁棒模型，需要大量数据。然而，由于实验成本高昂且耗时长，材料科学与工程领域的数据通常较为匮乏。本文提出一种方法，利用生成对抗网络与物理引导的判别器生成合成DIC位移数据。该判别器在判定数据样本真实与否时，额外接收导出的冯·米塞斯等效应变信息。研究表明，这种物理引导方法在样本视觉质量、切片Wasserstein距离以及几何评分方面均取得了更优结果。