Automated road distress detection

摘要

Potholes, cracks and patches are a few types of road surface distresses which are to be assessed India. Presently, road distress data assessment is reported to be done through distress data collection and processing the collected data. The data analysis and collection are to be automated these days, since manual assessment is expensive, time consuming and slows down the road maintenance management. The presence of shadows has been responsible for reducing the reliability of the system due to wrong detection results. Therefore, shadow detection and removal is important for improving performance. Many techniques have been proposed over the years, but shadow detection still remains an extremely challenging problem, particularly from a single image. In this paper, a robust method for automated detection and assessment of potholes, cracks and patches from images of Indian local roads is proposed, where effect of shadow is rectified. For testing its performance, the proposed method has been implemented using MATLAB. The results are evaluated through accuracy and precision recall metrics and compared with the methods presented by earlier researchers as well as current practices in the field.