A generalized computer vision approach to mapping crop fields in heterogeneous agricultural landscapes

, , , ,
1 Department of Civil & Environmental Engineering, Princeton University, Princeton, NJ, United States
2 Department of Operations Research & Financial Engineering, Princeton University, Princeton, NJ, United States
3 NASA Jet Propulsion Laboratory, Pasadena, CA, United States
4 Memorial Sloan Kettering Cancer Center, New York, NY, USA
5 Department of Civil and Environmental Engineering, Princeton University, Princeton, New Jersey, United States
DOI
10.7287/peerj.preprints.1367v1
Published
Accepted
Subject Areas
Agricultural Science, Computational Science, Environmental Sciences, Human-Computer Interaction, Spatial and Geographic Information Science
Keywords
Sub-Saharan Africa, agriculture, machine learning, computer vision, land cover
Copyright
© 2015 Debats 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
Debats S, Luo D, Estes L, Fuchs TJ, Caylor KK. 2015. A generalized computer vision approach to mapping crop fields in heterogeneous agricultural landscapes. PeerJ PrePrints 3:e1367v1

Abstract

Smallholder farms dominate in many parts of the world, particularly Sub-Saharan Africa. These systems are characterized by small, heterogeneous, and often indistinct field patterns, requiring a specialized methodology to map agricultural land cover. Using a variety of sites in South Africa, we present a new approach to mapping agricultural fields, based on efficient extraction of a vast set of simple, highly correlated, and interdependent features, followed by a random forest classifier. We achieved similar high performance across agricultural types, including the spectrally indistinct smallholder fields as well as the more easily distinguishable commercial fields, and demonstrated the ability to generalize performance across large geographic areas. In sensitivity analyses, we determined multi-temporal information provided greater gains in accuracy than multi-spectral information.

Author Comment

This manuscript has been submitted to another journal for review.