Towards Out-of-Distribution Detection for Remote Sensing
| Autor: | Anna Kruspe, Jakob Gawlikowski, Xiao Xiang Zhu, Sudipan Saha |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
business.industry
Computer science Out-of- distribution Reliability (computer networking) Deep learning robustness computer.software_genre Data modeling remote sensing Identification (information) open set recognition Artificial intelligence Scenario testing business computer Remote sensing Vulnerability (computing) Test data Data integration |
| Zdroj: | IGARSS |
| DOI: | 10.1109/igarss47720.2021.9553266 |
| Popis: | In remote sensing, distributional mismatch between the training and test data may arise due to several reasons, including unseen classes in the test data, differences in the geographic area, and multi-sensor differences. Deep learning based models may behave in unexpected manners when subjected to test data that has such distributional shifts from the training data, also called out-of-distribution (OOD) examples. Vulnerability to OOD data severely reduces the reliability of deep learning based models. In this work, we address this issue by proposing a model to quantify distributional uncertainty of deep learning based remote sensing models. In particular, we adopt a Dirichlet Prior Network for remote sensing data. The approach seeks to maximize the representation gap between the in-domain and OOD examples for a better identification of unknown examples at test time. Experimental results on three exemplary test scenarios show that the proposed model can detect OOD images in remote sensing. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|