Beamlet transform based technique for pavement image processing and classification

The goal of this paper is to automatically detect and classify pavement cracks using digital image processing techniques. This paper uses a pavement distress image enhancement algorithm to correct non-uniform background illumination by calculating the multiplication factors that eliminate the background lighting variations. To extract the linear features such as surface cracks from the pavement images a beamlet transform based algorithm is developed.