Abstract

This research focuses on the problem of detecting cracks on plastered brick wall surfaces. Cracks can cause damage and pose serious risks. The study proposes an automated method to address this issue, using two key algorithms: edge detection and image segmentation. The method begins by converting the color image into a grayscale image and then applies the Sobel filter to identify regions with significant changes in edge magnitude. The result is a grayscale image with the cracks' edges highlighted. Subsequently, the method uses the Otsu thresholding algorithm to automatically determine the appropriate threshold for each image, effectively separating the cracks from the background. Experimental images captured cracks of various sizes and shapes, with crack widths ranging from a maximum of 5 mm to approximately 1 mm minimum, captured by Iphone 13. The results of the study demonstrate that this method efficiently detects and segments cracks, capable of identifying both large and small cracks. The research results suggest that the proposed method has potential applications in maintenance and repair to enhance the aesthetics of building surfaces. However, the accuracy of the results depends on the resolution and angle of image capture.