Comparative analysis of image binarization methods for crack identification in concrete structures. (September 2017)

• Record Type: Journal Article
• Authors: Kim, Hyunjun; Ahn, Eunjong; Cho, Soojin; Shin, Myoungsu; Sim, Sung-Han
• Is Part Of: Cement and concrete research. Volume 99(2017)
• Page Start: 53
• Page End: 61
• Publication Date: 2017-09
• Subjects: Crack detection (B) -- Image analysis (B) -- Surface area (B) -- Concrete (E); Cement -- Periodicals; Cement -- Research -- Periodicals; Concrete -- Periodicals; Concrete -- Research -- Periodicals; Ciment -- Périodiques; Béton -- Périodiques; Cement; Concrete; Periodicals; 620.135
• DOI: 10.1016/j.cemconres.2017.04.018
• Languages: English
• ISSNs: 0008-8846

Abstract

Abstract: Surface cracks in concrete structures are critical indicators of structural damage and durability. Manual visual inspection, the most commonly used method in practice, is inefficient from cost, time, accuracy, and safety perspectives. A promising alternative is computer vision-based methods that can automatically extract crack information from images. Image binarization, developed for text detection, is appropriate for crack identification, as texts and cracks are similar, consisting of distinguishable lines and curves. However, standardizing crack identification using image binarization is challenging, because binarization depends on the method and associated parameters. We investigate image binarization for crack identification, focusing on optimal parameter determination and comparative performance evaluation for five common binarization methods. Crack images are prepared to obtain optimal parameters by minimizing errors in estimated crack widths. Subsequently, comparative analysis is conducted using crack images with different conditions based on three performance evaluation criteria: crack width and length measurement accuracy and computation time.