featurebiases.pdf (11.35 MB)
Deep learning robustness to domain shifts during seasonal variations
conference contribution
posted on 2023-06-10, 02:37 authored by Georgios Voulgaris, Andy PhilippidesAndy Philippides, Novi QuadriantoNovi QuadriantoIn certain geographic locations like South Asia, the landscape changes dramatically between dry and wet seasons. The main factor responsible for this variation is the flora that transforms the landscape between seasons. These transformations can affect the performance of deep learning models trained to analyse satellite images, especially if there are domain shifts between training and testing data distributions. The current work shows that an architecture which employs a Gabor convolutional layer as the first layer of a deep network input focuses on more salient parts of the image than one which uses a standard convolutional layer meaning that removing colour information is less damaging than for the standard network. Further we show that the proposed architecture is robust in the presence of domain shifts due to seasonal data variations.
History
Publication status
- Published
File Version
- Accepted version
Journal
IEEE International Symposium on Geoscience and Remote Sensing (IGARSS) proceedingsISSN
2153-7003Publisher
IEEEExternal DOI
Page range
1-4Event name
International Geoscience and Remote Sensing Symposium (IGARSS)Event location
Kuala Lumpur, MalaysiaEvent type
conferenceEvent date
17th - 22nd July 2022ISBN
9781665427920Department affiliated with
- Informatics Publications
Notes
© 2022 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.Full text available
- Yes
Peer reviewed?
- Yes