The method for semi-automatic extraction of residential area from high resolution remote sensing images
Residential area, as an important hazard-bearing body of earthquake disasters, usually constitutes the basis of earthquake emergency preparation.In this paper, 2 m resolution GF-1 satellite remote sensing data were used to extract the spatial distribution information of residential area, which could provide data support for the earthquake emergency preparation.The results reveal that more accurate residential area information of the high resolution GF-1 2 m image can be extracted based on gray level co-occurrence matrix, binarization and mathematical morphology.This proposed algorithm has high accuracy and good robustness.However, higher false alarm rate factor was shown in the extraction of sparse vegetation and non-residential buildings.Thus, in order to ensure data accuracy, the authors compared and analyzed the extraction results and the images, and extracted the ultimate data results semi-automatically by the artificial intervention.
国家科技期刊平台
NSTL
万方数据
维普
