Computer Vision-Based Foot Measurement Using Modified Canny Edge Algorithm

Algorithms of the Canny Edge Detection have been used to highlight the edges of an object in an image. It has been used to determine the sizes of things by applying machine visual or reference things. In this study, we developed an image capture device to obtain an image of the foot by implementing an image processing algorithm to measure the foot size. We evaluated the proposed system by measuring foot sizes using a t-test. We used the modified canny edge algorithm using the foot size measuring system with a median filter. Additionally, we determined the different factors that hindered the accuracy of the system. The factors included the contrast between the foot and the background, the deviation of the foot from the center of the platform, and the thresholds used in the different parts of the canny edge algorithm. The difference between the actual and the system-measured foot size was found to be insignificant. The computed p-value, 0.1809, was less than the critical value, 0.05, and the mean absolute error, 0.3729cm, was less than the margin of equivalence, 0.5cm.