Transportation Informatics: Advanced Image Processing Techniques Automated Pavement Distress Evaluation

The current project, funded by MIOH-UTC for the period 1/1/20094/30/2010, is concerned with the development of the framework for a transportation facility inspection system using advanced image processing techniques. The focus of this study is on the technical details of investigating and utilizing state-of-the-art image analysis techniques to further advance research in image processing based inspection systems in order to detect and classify the cracks in pavement. The detection of cracks and other degradations of pavement surfaces has traditionally been done by human experts conducting visual inspection while driving along the surveyed road. This manual approach is not only time consuming but also costly and subjective. To overcome these limitations we developed two different approaches for automatic crack detection and classification to speed up the process and reduce subjectivity. In the first approach, after the pavement images are captured by a digital camera, regions corresponding to cracks are detected over the acquired images by local segmentation and then represented by a matrix of square tiles. Since the crack pattern can be represented by the distribution of the crack tiles, standard deviations of both vertical and horizontal histograms are calculated to map the cracks onto a 2D feature space, where four crack types can be identified as: longitudinal cracks, transversal cracks, block cracks and alligator cracks. This new technique provides a low-cost, near real time distress analysis option. In the second approach we explore the use of a more robust multi-resolution scheme based on the beamlet transform. This method uses a pavement distress image enhancement algorithm to correct the non-uniform background illumination by calculating the multiplicative factors that eliminate the background lighting variations. To extract the linear features such as surface cracks from the pavement images, the image is partitioned into small windows and a beamlet transform based algorithm is applied. The crack segments are then linked together and classified into four types, vertical, horizontal, transversal, and block types. Simulation results show that the method is effective and robust in the extraction of cracks from a variety of pavement images. The experimental results, obtained by testing real pavement images over local asphalt roads, present the effectiveness of our algorithm for automating the process of identifying road distresses from images.