A multi-stage expert system for classification of pavement cracking

Recently, vast research attention has been put to develop automated procedures for pavement inspection and evaluation. The current work concentrates on developing a multi-stage expert system for pavement distress detection and classification. Mixture of Wavelet modulus and Three Dimensional Radon Transform (3DRT) are used for knowledge generation. The features and parameters of the peaks are finally used for training and testing the artificial neural network classifier. Experiments are conducted with distress images obtained from the GM database. High performance of the proposed approach demonstrates the advantages of this method in correct classification of pavement cracking.