Pavement crack image detection algorithm under nonuniform illuminance

The goal of this paper is to develop an algorithm to automatically detect and classify pavement cracks. Firstly, background subset interpolation method is used to adjust the nonuniform background illumination and decrease the shadows influence. The iterated threshold segmentation method, the closing operation and the sequential labeling of connected components are applied sequentially to segment the crack from the background accurately. Projection feature, total sum of crack pixel and distribution density are selected to classify the cracks. Lastly, the main parameters for crack, such as length, width and area are calculated. The results show that the proposed method is efficient.