A Comparison of Image Texture Analysis Methods for Automatic Recognition and Classification of Asphalt Pavement Distresses

Evaluation of pavement performance plays a major role in pavement management systems for determination of optimum strategy in repair and maintenance of the road. One of the most prominent assets in evaluation of the pavement at network and project level is identification and survey of surface pavement distresses. In the past two decades, extensive studies have been carried out in order to develop automatic methods for pavement evaluation. Most of these methods are based on machine vision and image processing techniques. One of the most important components of machine vision system is feature extraction. In the present study, after acquisition of six different group of asphalt pavement distresses under controlled condition, in order to compare different image texture processing methods for automatic recognition and categorising of distresses, first order statistical indices based on image intensity histogram, second order statistical indices based on grey level co-occurrence matrix and third and higher order statistics based on grey level run-length matrix have been used. Based on the results of the classification of distress images acquired by Mahalanobis minimum distance method, it can be concluded that although statistics extracted from histogram and grey level co-occurrence matrix have superior sensitivity in detection of alligator cracks, but third and higher order statistical indices based on grey level run-length matrix provide better results, with 80% classification accuracy, compared to other texture analysis approaches applied in the present research.