Title: Automatic Identification of agricultural terraces through object- oriented analysis of very high resolution DSM and multispectral imagery obtained from an unmanned aerial vehicle
Authors: DIAZ VARELA RAMONZARCO TEJADA PABLO JESUSANGILERI VincenzoLOUDJANI Philippe
Citation: JOURNAL OF ENVIRONMENTAL MANAGEMENT vol. 134 p. 117-126
Publisher: ACADEMIC PRESS LTD- ELSEVIER SCIENCE LTD
Publication Year: 2014
JRC N°: JRC81588
ISSN: 0301-4797
URI: http://dx.doi.org/10.1016/j.jenvman.2014.01.006
http://publications.jrc.ec.europa.eu/repository/handle/JRC81588
DOI: 10.1016/j.jenvman.2014.01.006
Type: Articles in periodicals and books
Abstract: Agricultural features such as terraces provide a number of environmental services, and therefore its maintenance is supported by measures implemented in the European Common Agricultural Policy (CAP). In the framework of the CAP implementation and control there is a current and future need for the development of robust, repeatable and cost effective methodologies for the automatic identification and monitoring of these features at farm scale. This is a complex task, particularly when terraces are associated to complex vegetation cover patterns, as in the case of terraced permanent crops (e.g. olive trees). In the present work we present a novel methodology for the automatic and cost-efficient identification of terraces in this scenario using exclusively imagery from commercial off-the-shelf (COTS) cameras on board of unmanned aerial vehicles (UAV). Using state of the art computer vision techniques, we generated ortho imagery and digital surface models (DSM) at 11 cm spatial resolution with low user intervention. These data were used in a second stage to identify terraces using a multi-scale object oriented classification method. The results showed the potential of this method even in high complexity agricultural areas, both regarding the DSM reconstruction and the classification stage. The UAV-derived DSM showed a RMSE lower than 0.5 m when assessing the heights of the terraces against field GPS data. The subsequent automated terrace classification yielded an overall accuracy of 90 % based exclusively on spectral and altitudinal information derived from the UAV imagery.
JRC Directorate:Sustainable Resources

Files in This Item:
There are no files associated with this item.


Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.