Computer-Aided Civil and Infrastructure Engineering 20 (2005) 52–61 Monitoring Crack Changes in Concrete Structures Hong-Gyoo Sohn, ∗ Yun-Mook Lim, Kong-Hyun Yun & Gi-Hong Kim School of Civil and Environmental Engineering, Yonsei University, 134 Shinchon-Dong Seodaemun-Gu, Seoul 120-749, Korea Abstract: This study proposes a crack-monitoring sys- tem to quantify the change of cracks from multitempo- ral images during the monitoring period. A series of images were taken from an off-the-shelf digital camera. Concrete cracks were extracted from the digital images by employing a series of image-processing techniques. The image coordinates and orientation of same cracks can be changed since the position and direction of the portable camera vary at every exposure time. To moni- tor the crack changes (width and length), it is critical to transform the image coordinates of cracks extracted from each image into the same object coordinates of the con- crete surface. In this study, such a geometric relationship was automatically recovered using the two-dimensional (2D) projective transformation based on the modified it- erated Hough transform (MIHT) algorithm, the result of which solved the transformation parameters. To improve the computational operation of MIHT, regions of param- eter estimation were also investigated. The developed al- gorithms were applied to monitor the crack of the concrete specimen. As a result, the change of cracks on the concrete specimen was successfully detected and accurately quan- tified. 1 INTRODUCTION Infrastructures have been extensively constructed dur- ing the last century. Concrete structures, in particular, were a major part of the constructions. Some of these concrete structures are now more than 40 years old, and their maintenance is one of the most important tasks in civil engineering. Concrete structures should be durable. In reality, the effects of inferior environmental condi- tions produce cracks on concrete structures, which could decrease their endurance. ∗ To whom correspondence should be addressed. E-mail: sohn1@ yonsei.ac.kr. The maintenance of the concrete structures began from the observation of cracks and changes in sur- faces (Bazant and Xiang, 1997; Yamazaki, 1998; Shi et al., 2001). The simplest form of quantitative anal- ysis is visual inspection of the cracks on concrete structures. Trained personnel may easily measure the length and width of cracks. The manually obtained data provide critical information and can be used to make an assessment of the concrete structures. This approach seems to be simple but it depends a lot on the knowledge and experience of each person. More- over, there are no permanent records of the examined area. The importance of image processing for visual inspec- tion has recently increased in various fields. Imagery, for- merly limited to film types but now including digital ones, is one of the efficient sources that can be utilized to ac- complish an accurate diagnosis of concrete structures. Most crack-detection algorithms rely on differences in gray values in digital images. On a typical concrete sur- face, cracks would be darker than the surrounding con- crete surface and could be easily extracted. In this case, efficient detection and accurate quantification of con- crete cracks can offer the potential for a more rational approach to maintenance. Many researchers have sought to identify methods by which cracks on concrete structures can be auto- matically detected using image processing (Dare et al., 2002; Schutter, 2002; Tung et al., 2002). These studies of- ten focus on methods that allow classification of crack types (Georgopoulos et al., 1995; Lee and Lee, 2004). However, in most digital image-processing algorithms for cracks, there has been a lack of approaches to the change detection of cracks (crack growth). Most of the researches are concentrated on the evaluation of their patterns, such as crack location, width, and types. Time- related change monitoring has remained another prob- lem to be solved. C  2005 Computer-Aided Civil and Infrastructure Engineering. Published by Blackwell Publishing, 350 Main Street, Malden, MA 02148, USA, and 9600 Garsington Road, Oxford OX4 2DQ, UK.