Aleatoric Uncertainty Embedded Transfer Learning for SEA-ICE Classification in SAR Images

Ying Liu, Zhongling Huang, Junwei Han

科研成果: 书/报告/会议事项章节会议稿件同行评审

4 引用 (Scopus)

摘要

Fine-grained sea-ice classification in SAR images is challenging due to the scarce labeled data and the imperfect annotation. Pre-training strategies are commonly carried out to prevent severe overfitting with limited labeled data. In spite of this, the observation noise still exists in the transferred features, which can be captured by aleatoric uncertainty. In this paper, we propose an aleatoric uncertainty embedded sea-ice classification method together with transfer learning of two different pre-training strategies. Instead of representing the transferred feature as a deterministic embedding, the proposed method concerns the feature uncertainty and models the embedding as a Gaussian distribution with variance. The experiments demonstrate that the proposed aleatoric uncertainty estimation is beneficial to improving the classification result of transfer learning. Based on the measured feature uncertainty, we analyze the potential of integrating two different pre-trained models to further enhance the performance.

源语言英语
主期刊名IGARSS 2022 - 2022 IEEE International Geoscience and Remote Sensing Symposium
出版商Institute of Electrical and Electronics Engineers Inc.
4980-4983
页数4
ISBN(电子版)9781665427920
DOI
出版状态已出版 - 2022
活动2022 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2022 - Kuala Lumpur, 马来西亚
期限: 17 7月 202222 7月 2022

出版系列

姓名International Geoscience and Remote Sensing Symposium (IGARSS)
2022-July

会议

会议2022 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2022
国家/地区马来西亚
Kuala Lumpur
时期17/07/2222/07/22

指纹

探究 'Aleatoric Uncertainty Embedded Transfer Learning for SEA-ICE Classification in SAR Images' 的科研主题。它们共同构成独一无二的指纹。

引用此