Characteristics of Color Digital Image Correlation for Deformation Measurement in Geomechanical Structures

Abstract

Digital image correlation (DIC) is a photogrammetric method that allows reconstruction of displacement and strain fields from the surface images. The displacements and strains are reconstructed by correlating sections (subsets) of the reference and deformed images generally taken before and after deformation respectively. Traditionally, DIC is applied to grayscale images taken with monochromatic cameras, but with the advent of digital color cameras there is a potential to use Color DIC using color images. It is expected that color images have more information and therefore same subset size would produce better results for Color DIC as compared to traditional grayscale DIC. This study examines various sizes of speckles in speckle patterns and different types of deformation and their effect on the performance of Color DIC and grayscale DIC. The speckle patterns and deformations were simulated and images were numerically generated. The errors are calculated by obtaining the difference between the measured and actual introduced simulated displacement or strain values. The results obtained suggest that generally Color DIC provides better results as compared to grayscale DIC. The quantity of improvement in performance of Color DIC dependent on the size of the speckles exits in the image and type of deformation. Usually, improvement is high for small subset sizes which decreases with the increase of the subset It is also found that for complex deformation scenarios, the performance of grayscale DIC is better than Color DIC if selected subset is larger than a specific subset size usually known as optimal subset size.