Application of Edge Detection Technique for Concrete Surface Crack Detection

Concrete is one of the most widely used construction materials in various types of infrastructure such as buildings, bridges, and so on. Concrete is preferable because it is easy to form, has high strength, is durable, and has very low maintenance costs. Given the nature of concrete, however, monitoring and maintenance must be performed regularly to avoid damage that can cause further deterioration and even collapse. Nowadays, monitoring for damage in concrete structures such as cracks is normally done by a conventional method such as direct in-field monitoring by an investigator by means of physical testing equipment used in the monitoring area. The situation may arise when the structure is in a condition that is difficult for humans to reach. For instance, in a high-rise building or an elevated bridge where the safety of the investigators is questioned. Therefore, to minimize the risk of accidents and facilitate the process of rapid monitoring of concrete structures, it is necessary to develop a practical method that is more human, safer, and more accurate thereby supporting sustainable infrastructure and smart monitoring in the future. The method that can be used is image processing technology. Image processing is a method for processing images by transforming digital images to obtain the required information from the image, in these cases in the form of damage to a concrete structure. The image processing method adopted in this work implanted the artificial intelligence which was used to classify images. The latter was preferred as it was known as a simple method and worked well with edge detection techniques. Specifically, this method worked by capturing the images from a high-resolution camera and the latter was processed under edge detection algorithms to detect the surface cracks. From this study, it is shown that the edge detection method is capable of demonstrating a clear virtual crack pattern which has a close resemblance with the actual crack pattern of a concrete surface, thereby suggesting better accuracy. It is also shown that the application of the edge detection technique provides additional information in terms of crack calculation which can be useful to quantify the damage in the concrete.