In order to automatically extract buildings from satellite images, we conducted experiments using Residual U-Net, one of semantic segmentation algorithms. Although limitations have been found, it has the advantage of automatically and quickly acquiring the building information which has been manually extracted by humans in the past.
|Publication status||Published - 2020|
|Event||40th Asian Conference on Remote Sensing: Progress of Remote Sensing Technology for Smart Future, ACRS 2019 - Daejeon, Korea, Republic of|
Duration: 2019 Oct 14 → 2019 Oct 18
|Conference||40th Asian Conference on Remote Sensing: Progress of Remote Sensing Technology for Smart Future, ACRS 2019|
|Country||Korea, Republic of|
|Period||19/10/14 → 19/10/18|
Bibliographical noteFunding Information:
This research was supported by a grant (2019-MOIS32-015) of Disaster-Safety Promotion Program funded by Ministry of Interior and Safety (MOIS, Korea).
All Science Journal Classification (ASJC) codes
- Information Systems