A Fractal Approach for Pavement Distress Detection Based on Differential Box-counting Algorithm

摘要

Presents a fractal approach to the detection of asphalt pavement cracks, which is employed in a pavement distress detection system. The utilization of fractal dimension is investigated for discriminating crack images of asphalt pavement. The fractal dimension(FD) of images captured by CCD camera is calculated with differential box-counting (DBC) algorithm that is computationally cheaper than other algorithms. The experimental results show that FD of images with cracks is larger than that with non-crack. The proposed approach can efficiently divide pavement images into two classes, one class of crack images and the other class of non-crack images.