We propose and demonstrate the basic computational standards for quantifying the accumulated plastic deformation of crystalline materials through the use of non-intrusive, cheap and high-throughput mechanical tests such as Digital Image Correlation (DIC). We demonstrate the utility of machine learning approaches for classifying and distinguishing DIC-data samples from materials with varying degrees of mechanical deformation.While such plastic deformation should be apparent in an electron microscope study of the tested material, we develop a standard for optical digital-image correlation and machine learning that should serve as a substitute for electron microscopy, with the idea that the DIC strategy is both cheaper and higher-throughput. In order to provide a clear, quantitative example of our techniques and standards, we use samples of dislocation patterns on material surfaces that have been mechanically deformed at varying degrees. These samples have been generated by the use of two dimensional discrete dislocation dynamics. Such data is directly comparable to the mechanical deformation of thin films.