An unsupervised augmentation framework for deep learning based geospatial object detection: A summary of results

Yiqun Xie, Rahul Bhojwani, Shashi Shekhar, Joseph Knight

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Scopus citations

Abstract

Given remote sensing datasets in a spatial domain, we aim to detect geospatial objects with minimum bounding rectangles (i.e., angle-aware) leveraging deep learning frameworks. Geospatial objects (e.g., buildings, vehicles, farms) provide meaningful information for a variety of societal applications, including urban planning, census, sustainable development, security surveillance, agricultural management, etc. The detection of these objects are challenging because their directions are often heavily mixed and not parallel to the orthogonal directions of an image frame due to topography, planning, etc. In addition, there is very limited training data with angle information for most types of objects. In related work, state-of-the-art deep learning frameworks detect objects using orthogonal bounding rectangles (i.e., sides are parallel to the sides of an input image), so they cannot identify the directions of objects and generate loose rectangular bounds on objects. We propose an Unsupervised Augmentation (UA) framework to detect geospatial objects with general minimum bounding rectangles (i.e., with angles). The UA framework contains two schemes, namely a ROtation-Vector (ROV) based scheme and a context-based scheme. The schemes completely avoid the need for: (1) additional ground-truth data with annotated angles; (2) restructuring of existing network architectures; and (3) re-training. Experimental results show that the UA framework can well approximate the angles of objects and generate much tighter bounding boxes on objects.

Original languageEnglish (US)
Title of host publication26th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, ACM SIGSPATIAL GIS 2018
EditorsLi Xiong, Roberto Tamassia, Kashani Farnoush Banaei, Ralf Hartmut Guting, Erik Hoel
PublisherAssociation for Computing Machinery
Pages349-358
Number of pages10
ISBN (Electronic)9781450358897
DOIs
StatePublished - Nov 6 2018
Event26th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, ACM SIGSPATIAL GIS 2018 - Seattle, United States
Duration: Nov 6 2018Nov 9 2018

Publication series

NameGIS: Proceedings of the ACM International Symposium on Advances in Geographic Information Systems

Other

Other26th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, ACM SIGSPATIAL GIS 2018
CountryUnited States
CitySeattle
Period11/6/1811/9/18

    Fingerprint

Keywords

  • Deep learning
  • Geospatial objects
  • Rectangles
  • Remote sensing
  • Rotations

Cite this

Xie, Y., Bhojwani, R., Shekhar, S., & Knight, J. (2018). An unsupervised augmentation framework for deep learning based geospatial object detection: A summary of results. In L. Xiong, R. Tamassia, K. F. Banaei, R. H. Guting, & E. Hoel (Eds.), 26th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, ACM SIGSPATIAL GIS 2018 (pp. 349-358). (GIS: Proceedings of the ACM International Symposium on Advances in Geographic Information Systems). Association for Computing Machinery. https://doi.org/10.1145/3274895.3274901