TY - CONF
T1 - Building damage mapping via transfer learning
AU - Xia, Junshi
AU - Adriano, Bruno
AU - Baier, Gerald
AU - Yokoya, Naoto
N1 - Funding Information:
The authors would like to thank ESA (SENTINEL Missions) and DigitalGlobe Inc. (through their Open data for disaster response program) for providing, Sentinel-1 and WorldView-3 images, respectively. The authors also would like to thank Prof. Junwei Han and Dr. Aito Fujita to provide NWPU and ABCD datasets, respectively. This research was funded by the Japan Society for the Promotion of Science (KAKENHI 18K18067 and 19K20309)
Publisher Copyright:
© 2019 IEEE.
PY - 2019
Y1 - 2019
N2 - This paper presents building damage mapping based on transfer learning techniques. Due to the different spatial resolutions of optical (WorldView, 0.5m) and SAR (Sentinel-1, 10m), we adopt different methods: pixel-level for moderate-resolution SAR images, and patch-level for very high-resolution optical images. For SAR images, the performance of fast unsupervised transfer learning methods, such as overall centroid alignment (OCA) and CORrelation ALignment (CORAL), are investigated. For the optical images, two public databases are used to predict the building damage mapping of Palu with WorldView-3 images via ResNet50. Experimental results indicate the effectiveness of transfer learning on the building damage mapping using different data sources.
AB - This paper presents building damage mapping based on transfer learning techniques. Due to the different spatial resolutions of optical (WorldView, 0.5m) and SAR (Sentinel-1, 10m), we adopt different methods: pixel-level for moderate-resolution SAR images, and patch-level for very high-resolution optical images. For SAR images, the performance of fast unsupervised transfer learning methods, such as overall centroid alignment (OCA) and CORrelation ALignment (CORAL), are investigated. For the optical images, two public databases are used to predict the building damage mapping of Palu with WorldView-3 images via ResNet50. Experimental results indicate the effectiveness of transfer learning on the building damage mapping using different data sources.
KW - Building damage mapping
KW - Optical
KW - SAR
KW - Transfer learning
UR - http://www.scopus.com/inward/record.url?scp=85113864612&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85113864612&partnerID=8YFLogxK
U2 - 10.1109/IGARSS.2019.8900447
DO - 10.1109/IGARSS.2019.8900447
M3 - Paper
AN - SCOPUS:85113864612
SP - 4841
EP - 4844
T2 - 39th IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2019
Y2 - 28 July 2019 through 2 August 2019
ER -