License: Creative Commons Attribution 3.0 Unported license (CC BY 3.0)
When quoting this document, please refer to the following
DOI: 10.4230/LIPIcs.GIScience.2021.I.3
URN: urn:nbn:de:0030-drops-130387
URL: https://drops.dagstuhl.de/opus/volltexte/2020/13038/
Go to the corresponding LIPIcs Volume Portal


Deng, Xueqing ; Tian, Yuxin ; Newsam, Shawn

Generalizing Deep Models for Overhead Image Segmentation Through Getis-Ord Gi* Pooling

pdf-format:
LIPIcs-GIScience-2021-I-3.pdf (1 MB)


Abstract

That most deep learning models are purely data driven is both a strength and a weakness. Given sufficient training data, the optimal model for a particular problem can be learned. However, this is usually not the case and so instead the model is either learned from scratch from a limited amount of training data or pre-trained on a different problem and then fine-tuned. Both of these situations are potentially suboptimal and limit the generalizability of the model. Inspired by this, we investigate methods to inform or guide deep learning models for geospatial image analysis to increase their performance when a limited amount of training data is available or when they are applied to scenarios other than which they were trained on. In particular, we exploit the fact that there are certain fundamental rules as to how things are distributed on the surface of the Earth and these rules do not vary substantially between locations. Based on this, we develop a novel feature pooling method for convolutional neural networks using Getis-Ord Gi* analysis from geostatistics. Experimental results show our proposed pooling function has significantly better generalization performance compared to a standard data-driven approach when applied to overhead image segmentation.

BibTeX - Entry

@InProceedings{deng_et_al:LIPIcs:2020:13038,
  author =	{Xueqing Deng and Yuxin Tian and Shawn Newsam},
  title =	{{Generalizing Deep Models for Overhead Image Segmentation Through Getis-Ord Gi* Pooling}},
  booktitle =	{11th International Conference on Geographic Information Science (GIScience 2021) - Part I},
  pages =	{3:1--3:14},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-166-5},
  ISSN =	{1868-8969},
  year =	{2020},
  volume =	{177},
  editor =	{Krzysztof Janowicz and Judith A. Verstegen},
  publisher =	{Schloss Dagstuhl--Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/opus/volltexte/2020/13038},
  URN =		{urn:nbn:de:0030-drops-130387},
  doi =		{10.4230/LIPIcs.GIScience.2021.I.3},
  annote =	{Keywords: Remote sensing, convolutional neural networks, pooling function, semantic segmentation, generalization}
}

Keywords: Remote sensing, convolutional neural networks, pooling function, semantic segmentation, generalization
Collection: 11th International Conference on Geographic Information Science (GIScience 2021) - Part I
Issue Date: 2020
Date of publication: 25.09.2020


DROPS-Home | Fulltext Search | Imprint | Privacy Published by LZI