DIGITAL IMAGE CORRELATION FOR SMALL STRAIN MEASUREMENT IN DEFORMABLE SOLIDS AND GEOMECHANICAL STRUCTURES Nghia V. Dinh a , Ghulam Mubashar Hassan a,b , Arcady V. Dyskin a , Cara MacNish b a School of Civil, Environmental and Mining Engineering, the University of Western Australia, Australia 21235131@student.uwa.edu.au, arcady.dyskin@uwa.edu.au b School of Computer Science & Software Engineering, the University of Western Australia, Australia ghulam.hassan@research.uwa.edu.au, cara.macnish@uwa.edu.au ABSTRACT Digital image correlation (DIC) is a well-known contact- less technique offering highly accurate full-field defor- mation measurement using grayscale images. The practical implementation of DIC is still facing many challenges, es- pecially limitations of accuracy in measuring small dis- placement gradients for solids in geosciences and biomedi- cal engineering. In this paper, we introduce a novel ap- proach in which color images are employed to enhance the performance of DIC. A complete framework for Color DIC has been proposed and tested. The results show that Color DIC performs significantly better than grayscale DIC for measurement of small strains by a factor of 2. Index Terms—Image processing, digital image correla- tion, color image, deformation measurement, DIC. 1. INTRODUCTION Deformation measurement plays an important role in many fields to observe displacement and strain properties of ob- jects under load. In engineering, strain measurement plays more critical role in reflecting the state of objects or struc- tures. It helps to investigate and monitor the integrity of the structures. The traditional method for measuring strain is to use physical measuring devices, such as strain gauges, fixed directly to the surface of the object. The physical devices have limited accuracy, and the requirement to install them on the surface has practical limitations. More importantly, the measurements are limited to the points of application. In order to measure strain continuously over the area of inter- est, or ‘full field’, optical techniques are increasingly used, such as photo elasticity, holographic, grid method, particle image velocimetry (PIV) and digital image correlation (DIC) [1-5]. Currently, the most commonly used image pro- cessing technique, DIC, is considered a promising tool in measuring strain and displacement in experimental mechan- ics, geosciences and biomedical engineering. DIC is a contactless, full-field deformation measure- ment method which uses digital images to track changes in deformable solids. Its general setup is presented in Fig. 1. DIC is used in wide range of applications, to monitor de- formation in materials such as metals [6], composites [7], brittle materials [8], plastic [9], rock [10] and bio-materials [11]. A particularly important use is Structural Health Mon- itoring (SHM) in civil engineering [12], for example, moni- toring deflection and performance of bridges [13, 14]. DIC was introduced by Peters and Ranson in 1982 [15] and works by correlating small regions of reference and deformed images. This allows the determination of dis- placements and reconstruction of gradients, or strains. Since its inception there have been many improvements [4, 16]. The accuracy and sensitivity of DIC vary within orders of magnitude from 0.5 pixels [17] to 0.01 pixels [4, 18], and two dimensional DIC is reported as most effective when strain is larger than 0.003 [4]. The outcome of DIC is main- ly dependent on image pattern, mapping process, exposure time of the object, out-of-plane movement, thermal effects or imaging devices selection and capturing process [4, 19- 24]. Therefore, the development of DIC has been continuous to achieve wider range of applications and better accuracy. This paper proposed extending DIC in a novel way by using color images. It is stimulated by the fact that a number of studies have been conducted to improve DIC application to specific fields, however, usually grayscale images have been used. It is believed that color images offer DIC an ad- vantage of more information than grayscale images in im- proving the accuracy. Figure 1: General DIC setup OBJECT Camera Digital images Analysis Light source Computer Deformation