Abstract
The WorldView-2 high spatial resolution satellite with eight multispectral imaging bands is ideally suited for extracting built-up areas (BUs) from remote sensing images. In this study, an object-based automatic multi-index BUs extraction method was developed. First, several indices, including BUs extraction index (NBEIr-c), vegetation extraction index(NDVInir2-r) and water extraction index (NDWI b-nir1), were developed to obtain the BUs, vegetation and water maps, and then the fractional-order Darwinian particle swarm optimization (FODPSO) algorithm was employed to automatically segment the multi-index images and obtained BUs, water, vegetation and bare soil (BS) information. Finally, the extracted BUs results were optimized via an object-based analysis method and the results were compared with those of two other relevant indices, which confirmed the proposed method had a higher accuracy and exhibited higher performance when separating the BS from the BUs.
Acknowledgement
We would like to thank the WorldView-2 data were obtained from the Digital Globe.
Disclosure statement
No potential conflict of interest was reported by the authors.
Funding
This work was supported by Hainan Provincial Department of Science and Technology (Grant No. ZDKJ2016021) and the Major Special Project-the China High-Resolution Earth Observation System (Grant 30-Y20A07-9003-17/18).