A comparison of multi-resolution methods for detection and isolation of pavement distress

摘要: The research presented in this article is aimed at the development of an automated imaging system for distress detection and isolation asphalt pavement obtained from image acquisition (PIAS). This focuses on comparing discriminating power several multi-resolution texture analysis techniques using wavelet, ridgelet, curvelet-based descriptors. approach consists four steps: Image collection, segmentation regions interest (ROI), extraction most discriminative features, creation a classifier that automatically identifies distress, storage. Tests curvelet features indicated signatures outperform all other pothole yielding accuracy rates 97.9%. Ridgelet-based cracking 93.6-96.4% rate.