Fine-scale impervious surface extraction by fusing high resolution remote sensing images and vehicle digital measurable images

Abstract

Impervious surface area (ISA) as the indicator of urbanization has a significance for urban ecological environment evaluation. Impervious surface extracting methods based on high-resolution remote sensing imagery can extract ISA in a fine-scale. However, a series of consequent problems cannot be ignored, such as shadows from tall building and canopies. In order to solve the shadow problem, high-resolution remote sensing image from Wuhan city is divided into three classes which are shadows, impervious surface and pervious. Due to the shadow and vegetation area along the street, the vehicle mounted digital measurable images are introduced to obtain fine-scale impervious surface information. D-S theory of evidence was employed to detect the shadow area according to spectral information. Then the impervious surface was extracted by fusing VHR remote sensing image and vehicle digital measurable images. The results showed that the proposed method was capable to provide the extraction of fine-scale impervious surface.