Road Damage Information Extraction Based on High-Resolution SAR Imagery

In this paper, an approach for road damage information extraction based on multi-temporal synthetic aperture radar (SAR) images is proposed. The road centerline is obtained by using particle filtering. According to complex mountainous terrains, the image pre-processing and the filtering process are adjusted and the road centerline can identified correctly. A road damage model is then developed in terms of the features of the road damage in SAR images, thus the road damage information is attained. The experimental results indicate that the proposed method can achieve the accuracy of road damage information above 80%.