Urban Trunk Roads Extraction Using Hough Transform and NDVI in Airborne Hyperspectral Remote Sensing Images

In order to extract urban main road in hyperspectral remote sensing images, this paper first employs Hough transform in three main different band of hyperspectral image datasets to identify and extract urban main road. Then, Normalized difference vegetation index (NDVI) is calculated to extract vegetation information and eliminate the disturbance brought by vegetation such as trees along roads and others vegetation landscape. The experiments use two real airborne hyperspectral datasets and results show that method proposed can effectively extract the main road in airborne hyperspectral remote sensing images by Hough transform and NDVI.