Title

  • external user (warningwarning)
  • Log in as
  • language uk
Record number 2249983
Title Injecting spatial priors in Earth observation with machine vision
show extra info.
Diego Marcos Gonzalez
Author(s) Marcos Gonzalez, Diego (dissertant)
Publisher Wageningen : Wageningen University
Publication year 2019
Description XII, 130 pages figures, diagrams
Description 1 online resource (PDF, XII, 130 pages) figures, diagrams
Notes Includes bibliographical references. - With summary in English
ISBN 9789463435604; 9463435603
Tutors Tuia, Prof. dr. D.
Graduation date 2019-02-11
Dissertation no. 7153
Author abstract show abstract

Remote Sensing (RS) imagery with submeter resolution is becoming ubiquitous. Be it from satellites, aerial campaigns or Unmanned Aerial Vehicles, this spatial resolution allows to recognize individual objects and their parts from above.

This has driven, during the last few years, a big interest in the RS community on Computer Vision (CV) methods developed for the automated understanding of natural images.

A central element to the success of \\CV is the use of prior information about the image generation process and the objects these images contain: neighboring pixels are likely to belong to the same object; objects of the same nature tend to look similar with independence of their location in the image; certain objects tend to occur in particular geometric configurations; etc.

When using RS imagery, additional prior knowledge exists on how the images were formed, since we know roughly the geographical location of the objects, the geospatial prior, and the direction they were observed from, the overhead-view prior.

This thesis explores ways of encoding these priors in CV models to improve their performance on RS imagery, with a focus on land-cover and land-use mapping.

Online full textINTERNET
On paper Get the document, find related information or use other SFX services
Publication type PhD thesis
Language English
Comments
There are no comments yet. You can post the first one!
Post a comment
 

To support researchers to publish their research Open Access, deals have been negotiated with various publishers. Depending on the deal, a discount is provided for the author on the Article Processing Charges that need to be paid by the author to publish an article Open Access. A discount of 100% means that (after approval) the author does not have to pay Article Processing Charges.

For the approval of an Open Access deal for an article, the corresponding author of this article must be affiliated with Wageningen University & Research.

Please log in to use this service. Login as Wageningen University & Research user or guest user in upper right hand corner of this page.