A method and system for extracting building areas based on sar images

Abstract

The present invention provides a method and system for extracting a building area based on a SAR image. The method for extracting a building area based on a SAR image includes: acquiring a first image based on a SAR image; generating a local spatial index image and a variogram based on the first image image; based on the local space index image, variogram image, and the first image, respectively filter the sampling points to obtain the corresponding sets; process the sampling points in each set according to the preset growth criteria to construct the images of the building areas to be selected, and divide each Select the image of the built-up area to superimpose to obtain the final image of the built-up area. The method or system provided by the invention can quickly and accurately extract building areas from Sentinel-1 SAR images, realize automatic extraction of building land in large areas, and lay a foundation for global building area mapping.

Description

technical field [0001] This application relates to the field of remote sensing, in particular to a method and system for extracting building areas based on SAR images. Background technique [0002] Research shows that 66% of the world's population will live in urban areas by 2050. Although urban areas occupy only a small fraction of the global landmass, they largely influence ecological changes due to their concentration of traffic and energy consumption. Therefore, accurate and timely acquisition of building land information is of great significance for urban planning and post-earthquake disaster relief. [0003] Optical remote sensing satellites provide valuable data sources for building land extraction, but compared with optical sensors, radar systems have the characteristics of all-weather and all-weather observation and imaging. SAR data has a wider application space. Although the interpretation of SAR images is difficult, many scholars have carried out research on t...

AI Technical Summary

Problems solved by technology

However, these methods only rely on the intensity information of the image. For example, for a low-resolution SAR image such as 75m, the settlement information can be accurately obtained, but for a medium-resolution wide SAR image (such as a Sentinel-1 SAR image with a resolution of 20m) , the building area has more obvious light and dark texture features, therefore, only using the intensity information cannot obtain a complete and accurate distribution of the building area

Images