The uploaded pothole road images in the online server’s database are retrieved one at a time and various preprocessing
algorithms and the proposed pothole detection algorithm are applied. Fig. 3 shows the output of various pre-processing steps of the pothole image. During the first step, the input image is converted into
binary image using gray scale algorithm. Then median filer is applied on binary image to remove noise. The
next step is applying Canny Edge detection algorithm to extract the useful structural information from the
image. Then dilation is applied to the image, which adds the pixels to the boundaries of object in the image
enlarging the boundaries of regions of foreground pixels to grow in size and makes the holes within those
regions to get reduced.