A Novel Approach for the Determination of Surface Tilt Angles in Two-Dimensional Digital Image Correlation Experiments

Abstract

Two-dimensional digital image correlation (2D-DIC) is increasingly being used in a wide variety of applications. This technique is capable of measuring the in-plane deformations and strains of planner surfaces. A basic condition for using this technique is having the camera to observe the target surface perpendicularly. If the camera is not perpendicular to the surface, errors will be introduced in the measured displacements and strains. Small amounts of camera misalignment can easily go undetected during the initial setting of the experimental setup. In this paper, a novel approach for verifying the perpendicularity of the camera with respect to the target surface in 2D-DIC experiments is introduced. If the camera is not perpendicular to the surface, the tilt angles about both of the two in-plane axes can be calculated using this novel approach. A simple in-plane rigid-body-translation of the specimen in any of the two in-plane directions (horizontal or vertical direction relative to the camera image), is used for detecting camera non-perpendicularity and calculating the tilt angle(s). This approach uses the normal and shear strain errors, induced in DIC results due to camera non-perpendicularity, to determine the surface tilt angle(s), if any is present. The tilt angle(s) are calculated using simple analytical equations that are developed based on the pinhole camera model. The proposed approach is validated using experiments performed while the target surface is tilted at angles ranging from 0.5° to 4°. The experimental results show that this approach is able to confidently detect camera misalignment for tilt angles of 1° or larger and the tilt angles can be calculated with less than 0.3° error.