A platform for crowdsourcing the creation of representative, accurate landcover maps

Woodrow Wilson School, Princeton University, Princeton, New Jersey, USA
Department of Civil and Environmental Engineering, Princeton University, Princeton, New Jersey, USA
Computational Science and Engineering Support, Office of Information Technology, Princeton University, Princeton, New Jersey, USA
Department of Geography, Indiana University, Bloomington, Indiana, USA
DOI
10.7287/peerj.preprints.1030v2
Subject Areas
Agricultural Science, Ecology, Environmental Sciences, Human-Computer Interaction, Coupled Natural and Human Systems
Keywords
remote sensing, landcover, crowd-sourcing, accuracy assessment, representative sampling, object extraction
Copyright
© 2015 Estes et al.
Licence
This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ PrePrints) and either DOI or URL of the article must be cited.
Cite this article
Estes L, McRitchie D, Choi J, Debats SR, Evans T, Guthe W, Luo D, Ragazzo G, Zempleni R, Caylor K. 2015. A platform for crowdsourcing the creation of representative, accurate landcover maps. PeerJ PrePrints 3:e1030v2

Abstract

Accurate landcover maps are fundamental to understanding socio-economic and environmental patterns and processes, but existing datasets contain substantial errors. Crowdsourcing map creation may substantially improve accuracy, particularly for discrete cover types, but the quality and representativeness of crowdsourced data is hard to verify. We present an open-sourced platform, DIYlandcover, that serves representative samples of high resolution imagery to an online job market, where workers delineate individual landcover features of interest. Worker mapping skill is frequently assessed, providing estimates of overall map accuracy and a basis for performance-based payments. A trial of DIYlandcover showed that novice workers delineated South African cropland with 91% accuracy, exceeding the accuracy of current generation global landcover products, while capturing important geometric data. A scaling-up assessment suggests the possibility of developing an Africa-wide vector-based dataset of croplands for $2-3 million within 1.2-3.8 years. DIYlandcover can be readily adapted to map other discrete cover types.

Author Comment

This is a revised version of the original paper following an initial peer review. The title is slightly changed, and new analyses of the potential time and cost of mapping large regions is presented. Other changes in the text have been with respect to describing the underlying rationale, prior studies, and broader applications of the software have also been made.

Supplemental Information

Appendix S2

Supplementary results figures

DOI: 10.7287/peerj.preprints.1030v2/supp-2

Appendix S3

Code used to analyze trial data and create tables and figures.

DOI: 10.7287/peerj.preprints.1030v2/supp-3