Automatic Evaluation of Pavement Cracking Using Radon Transform and Providing A General Index of Pavement Status

The automatic evaluation of pavement condition is an important and emerging area for pavement management systems (PMS). In most of the pattern recognition methods, thresholding is used to automate the processing. In this paper, a new method is proposed to determine an effective threshold for evaluation of pavement cracking. In the proposed method, three-dimensional Radon transform data has been used to investigate cracking. Also, based on fuzzy theory, a flexible new method is presented based on image data. In order to evaluate the efficiency of the proposed method, a general comparison between the fixed threshold and specific filtering method has been performed. The proposed method was evaluated on a set of image. The results indicate that the proposed method shows better performance than other existing methods in terms of speed, accuracy, and comprehensiveness. This method can be used to determine the type, severity, extent, and general evaluation of pavement cracking behavior in automated analysis.