Journal
REMOTE SENSING
Volume 13, Issue 6, Pages -Publisher
MDPI
DOI: 10.3390/rs13061172
Keywords
deep learning; building extraction; ensemble learning; GAN
Categories
Funding
- Beijing Municipal Science and Technology Project [Z191100001419002]
Ask authors/readers for more resources
With the progress of urbanization, the management issues in the wildland-urban interface have become more serious. Building research is crucial in this area, and methods for extracting building information can be obtained from high-resolution remote sensing images or relevant agencies.
Following the advancement and progression of urbanization, management problems of the wildland-urban interface (WUI) have become increasingly serious. WUI regional governance issues involve many factors including climate, humanities, etc., and have attracted attention and research from all walks of life. Building research plays a vital part in the WUI area. Building location is closely related with the planning and management of the WUI area, and the number of buildings is related to the rescue arrangement. There are two major methods to obtain this building information: one is to obtain them from relevant agencies, which is slow and lacks timeliness, while the other approach is to extract them from high-resolution remote sensing images, which is relatively inexpensive and offers improved timeliness. Inspired by the recent successful application of deep learning, in this paper, we propose a method for extracting building information from high-resolution remote sensing images based on deep learning, which is combined with ensemble learning to extract the building location. Further, we use the idea of image anomaly detection to estimate the number of buildings. After verification on two datasets, we obtain superior semantic segmentation results and achieve better building contour extraction and number estimation.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available