Improving the accuracy of crack length measurement using machine vision

In the current paper a new method is introduced to analyze and measure the cracks dimensions in solid materials such as mechanical tools and bricks. Since the cracks do not have a regular or predictable shape, in order to achieve the exact dimensions of such cracks, the conventional mathematical formulas are by no means applicable. Hence, while studying different crack analyzing methods, we argue on their faults and limits and propose our method which aims to measure the crack dimensions in a solid object by utilizing machine vision, image capturing and image processing techniques. We define new algorithms and perform picture scaling in real dimensions and analyze the acquired data to obtain the most precise results. This optimal machine vision technique is performed in two different ways. The first method is based on the measurement of the length of the image skeleton, while the second technique is based on measuring the half of the perimeter of the crack's image. After obtaining the measurements, we optimize the results with the help of some pre-defined algorithms. It is shown that our proposed algorithms provide a reliable method which can be used to measure any crack dimensions. Also, we apply these techniques on a sample brick with some random cracks on it. After gaining the binary images, filters are applied to gain the best results among all images. (C) 2015 The Authors. Published by Elsevier Ltd.