A Kind of Pavement Crack Detection Method Based on Digital Image Processing

In order to meet the requirements of rapid detection of pavement cracks, it applies the image processing technology to automatically detect cracks. Firstly, the obtained crack images were been grayed and gray-scale corrected. Secondly, an improved median filtering method based on weighted median filtering was been proposed to denoise the crack images. Then, a two-dimensional Otsu threshold segmentation method based on the optimization of mixed particle groups was been proposed to divide the crack image. Finally, the physical size of the pixels was calibrated and the characteristic parameters of the cracks were extracted. The experimental results show that the method is effective and can accurately extract the crack, and the relative error between the detected linear crack width value and the measured value is within 3%.