Crack in any structure is a destructive phenomenon. It can cause the early failure of any structure or component. It may lead to severe accidents and may harm humans too. So, it is very essential to detect these cracks effectively and do the required measures to avoid failure or accidents. Previously, this was done by skilled inspectors but it is relatively difficult for them to observe inaccessible areas like walls of dams or to observe every component on the assembly line, etc. So, a reliable

more »

... and automated method of crack detection is required. In this paper, a method using MATLAB software and with the help of image processing techniques like thresholding, morphological operation proper detection of crack is proposed. The edge detection technique is used for extracting the crack boundaries and for better visualisation of cracks, these extracted boundaries are superimposed on the original image. The parameter of the cracklike percentage area of the crack is also determined. Also, as the thresholding value is calculated automatically for every sample, the system developed can be used for all similar kinds of samples without altering the parameters.