Automatic Crack Recognition System for Concrete Structures Using Image Processing Approach

摘要

Recently, digital image processing technologies have been widely applied in different industries, because computerized technologies were widely utilized. In performance evaluation of deteriorating concrete bridges, the defect recognition such as crack pattern, crack width, crack depth, etc. On concrete surface is a significant advance and may lead to automation of the existing concrete bridge inspection at primary stage of the diagnosis. This study describes a novel approach for developing a system that extracts a crack pattern from digital images. A characteristic feature of the system is the use of an interactive genetic algorithm to optimize some parameters involved in the digital image processing. The algorithm prevents the system user from adjusting the parameters by trial and error. The user only evaluates some images produced by the system. The effectiveness of the system is verified by comparison of the results processed by trial and error and those obtained by using the proposed system. Once the above mentioned crack pattern on the concrete surface can be recognized, this study also introduces, in addition to an automatic measuring method of the maximum crack width, a practical recognition expert system which performs automatically with a certain level of accuracy.