Deep convolutional segmentation of remote sensing imagery: A simple and efficient alternative to stitching output labels

Bohao Huang, Leslie M. Collins, Kyle Bradbury, Jordan M. Malof

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

8 Scopus citations

Abstract

In this work we consider the application of convolutional neural networks (CNNs) for the semantic segmentation of remote sensing imagery (e.g., aerial color or hyperspectral imagery). In segmentation the goal is to provide a dense pixel-wise labeling of the input imagery. However, remote sensing imagery is usually stored in the form of very large images, called “tiles”, which are too large to be segmented directly using most CNNs and their associated hardware. During label inference (i.e., obtaining labels for a new large tile) smaller sub-images, called “patches”, are extracted uniformly over a tile and the resulting label maps are “stitched” (or concatenated) to create a tile-sized label map. This approach suffers from computational inefficiency and risks of discontinuities at the boundaries between the output of individual patches. In this work we propose a simple alternative approach in which the input size of the CNN is dramatically increased only during label inference. We evaluate the performance of the proposed approach against a standard stitching approach using two popular segmentation CNN models on the INRIA building labeling dataset. The results suggest that the proposed approach substantially reduces label inference time, while also yielding modest overall label accuracy increases. This approach also contributed to our winning entry (overall performance) in the INRIA building labeling competition.

Original languageEnglish
Title of host publication2018 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages6899-6902
Number of pages4
ISBN (Electronic)9781538671504
DOIs
StatePublished - Oct 31 2018
Event38th Annual IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018 - Valencia, Spain
Duration: Jul 22 2018Jul 27 2018

Publication series

NameInternational Geoscience and Remote Sensing Symposium (IGARSS)
Volume2018-July

Conference

Conference38th Annual IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018
Country/TerritorySpain
CityValencia
Period07/22/1807/27/18

Keywords

  • Aerial imagery
  • Building detection
  • Convolutional neural networks
  • Deep learning
  • Semantic segmentation

Fingerprint

Dive into the research topics of 'Deep convolutional segmentation of remote sensing imagery: A simple and efficient alternative to stitching output labels'. Together they form a unique fingerprint.

Cite this