Comparison of image operators for time-based photogrammetry

Abstract

Real-time deformation measurements have the potential to be used in a wide range of civil and structural engineering applications. This paper introduces an algorithm that utilizes time-based photogrammetry to measure deformations in real-time. The hardware used in the algorithm consists of a camera and a computer that runs camera-control software and a MATLAB code. The code is responsible for detecting the approximate locations of the circular targets using normalized 2D cross-correlation and performing the sub-pixel measurement of the center by fitting an ellipse on the edges of the circular target. During an experimental load test of a concrete slab outfitted with Ringed Automated Detection targets, a series of images were taken at different loads. The coordinates of the targets were measured using the edge-ellipse operator, least-squares matching, and digital image correlation methods. The edge-ellipse operator has proven to be an effective method of measuring the image coordinates of circular targets.