Sangam: A Confluence of Knowledge Streams

Improving Deep Learning for Maritime Remote Sensing through Data Augmentation and Latent Space

Show simple item record

dc.creator Sobien, Daniel
dc.creator Higgins, Erik
dc.creator Krometis, Justin
dc.creator Kauffman, Justin
dc.creator Freeman, Laura
dc.date 2022-07-08T12:05:15Z
dc.date 2022-07-08T12:05:15Z
dc.date 2022-07-07
dc.date 2022-07-08T11:55:12Z
dc.date.accessioned 2023-03-01T18:53:33Z
dc.date.available 2023-03-01T18:53:33Z
dc.identifier Sobien, D.; Higgins, E.; Krometis, J.; Kauffman, J.; Freeman, L. Improving Deep Learning for Maritime Remote Sensing through Data Augmentation and Latent Space. Mach. Learn. Knowl. Extr. 2022, 4, 665-687.
dc.identifier http://hdl.handle.net/10919/111176
dc.identifier https://doi.org/10.3390/make4030031
dc.identifier.uri http://localhost:8080/xmlui/handle/CUHPOERS/281737
dc.description Training deep learning models requires having the right data for the problem and understanding both your data and the models’ performance on that data. Training deep learning models is difficult when data are limited, so in this paper, we seek to answer the following question: how can we train a deep learning model to increase its performance on a targeted area with limited data? We do this by applying rotation data augmentations to a simulated synthetic aperture radar (SAR) image dataset. We use the Uniform Manifold Approximation and Projection (UMAP) dimensionality reduction technique to understand the effects of augmentations on the data in latent space. Using this latent space representation, we can understand the data and choose specific training samples aimed at boosting model performance in targeted under-performing regions without the need to increase training set sizes. Results show that using latent space to choose training data significantly improves model performance in some cases; however, there are other cases where no improvements are made. We show that linking patterns in latent space is a possible predictor of model performance, but results require some experimentation and domain knowledge to determine the best options.
dc.description Published version
dc.format application/pdf
dc.format application/pdf
dc.language en
dc.publisher MDPI
dc.rights Creative Commons Attribution 4.0 International
dc.rights http://creativecommons.org/licenses/by/4.0/
dc.title Improving Deep Learning for Maritime Remote Sensing through Data Augmentation and Latent Space
dc.title Machine Learning and Knowledge Extraction
dc.type Article - Refereed
dc.type Text


Files in this item

Files Size Format View
make-04-00031.pdf 3.426Mb application/pdf View/Open

This item appears in the following Collection(s)

Show simple item record

Search DSpace


Advanced Search

Browse