Investigation of Crack Propagation of Fly Ash-Based Geopolymer Concrete Using Digital Image Segmentation Approach

Fly ash, a byproduct of burning coal, is usually used as a supplementary cementitious constituent in the creation of concrete. It has been discovered that using fly ash in a concrete has a number of advantages, including improved workability, decreased permeability, and greater strength. Using image processing techniques, this study seeks to regulate the ideal amount of fly ash in the concrete mix in respect to crack propagation. The study entails analyzing concrete cylinders samples in the lab with dimensions of 150 mm × 300 mm that contain different amounts of fly ash—5%, 15%, 25%, 30%, 40%, and 50% by weight of cement. A total of 120 concrete cylinders were prepared, and compressive tests were performed. The photographs of cracks under proper lighting conditions were taken to analyze using image processing. Different parameters were used for study: Otsu threshold value, size of area of interest of crack, and fly ash content. The Otsu threshold image segmentation was utilized as a binary thresholding method to detect the crack from the images. The parameters varied according to the histogram of each crack images. The segmentation was done by minimizing the variance on each class. With these analyzed parameters, the results signified that the percentage of fly ash content was achieved to be within 25–30% of cementitious materials. This optimum percentage fall in the category of “low crack class” as defined in this paper.