English   español  
Por favor, use este identificador para citar o enlazar a este item: http://hdl.handle.net/10261/121360
logo share SHARE   Add this article to your Mendeley library MendeleyBASE
Visualizar otros formatos: MARC | Dublin Core | RDF | ORE | MODS | METS | DIDL
Exportar a otros formatos:

CROPCLASS-2.0 software for census parcel cropping systems classification from multitemporal remote imagery

Otros títulosSoftware CROPCLASS-2.0 para la clasificación de cultivos en imágenes multitemporales remotas a nivel parcela del censo agrícola
AutorGarcía Torres, Luis ; Gómez-Candón, David ; Caballero Novella, Juan José ; Peña Barragán, José Manuel ; López Granados, Francisca ; Jurado-Expósito, Montserrat
Palabras claveCROPCLASS
Fecha de publicaciónabr-2014
EditorCSIC - Instituto de Agricultura Sostenible (IAS)
ResumenA research group of the Institute for Sustainable Agriculture (CSIC, Cordoba, Spain) has developed an original procedure to classified crops from multitemporal remote sensing images, named CROPCLASS, to be used in agricultural and forestry scenes. The procedure is governed by registered CROPCLASS-2.0 software (1), which executes it semi-automatically. CROPCLASS procedure was patented (2) and it is described in a recent publication (3). Before developing CROPCLASS, the methodological approaches for cropping systems classification from remote sensing images widely varied among authors and required tremendous effort in image processing.
For each geographical area CROPCLASS consists of: 1) a definition of census parcels through vector files in the images; 2) the extraction of spectral bands (SB) and key vegetation index (VI) average values for each parcel and image; 3) the conformation of a matrix data (MD) of the extracted information; 4) the classification of MD through decision trees (DT), which provide a Structured Query Language (SQL) crop predictive model. The procedure is also based on preliminary land-use ground-truth work in a reduced number of parcels at least the first year of study. Crop SQL predictive models can be used to classify unidentified parcels land uses from the same area where the images were taken to generate the model.
CROPCLASS procedure meets additional advantages as follows. First, the census parcel is the unit for most administrative actions and CROPCLASS provide record for each census parcel. Second, administrations require a defined crop classification method, almost fully relying on remote sensed images automatically or semi-automatically executed, consistently reducing the ground-visit work as much as possible. Third, the predictive models for each crop/cropping system are likely to be used for the same area in subsequent years if the images were taken on about the same dates. This use is based on the true assumption that in each geographical area, the diversity of the crops and the crop calendar remain about the same throughout the years. The phenology or crop growth stages will approximately coincide, as the images were taken at about the same time in different years; therefore, the predictive models that were determined for one year with similar timings could tentatively be used in subsequent years.
Implementing the CROPCLASS procedure through conventional image processing is time consuming and requires computer language skills. The software CROPCLASS-2.0® can be implemented for any agricultural region semi-automatically, in an economically feasible manner. CROPCLASS-2.0 software is available at the digital repository (http://dx.doi.org/10.5061/dryad.j958j) only for research and academic purposes; furthermore its authorship should be mentioned with bold characters.
DescripciónContiene 6 documentos (1. Objetivos, alcance y publicaciones. 2. Manual de código. 3. Instalación) y 3 archivos de software
ReferenciasGarcía Torres, Luis; Caballero Novella, Juan José; Gómez-Candón, David; López Granados, Francisca. Automatic image processing for agriculture through specific ENVI modules (add-on). http://hdl.handle.net/10261/121363
Caballero Novella, Juan José; García Torres, Luis; Gómez-Candón, David. Procedimiento CROPCLASS® de clasificación de cultivos en imágenes remotas a nivel parcela para su uso en el censo agrícola. http://hdl.handle.net/10261/121367
García Torres, Luis, Caballero Novella, Juan José, Gómez-Candón, David, Peña, José Manuel. Census Parcels Cropping System Classification from Multitemporal Remote Imagery: A Proposed Universal Methodology. 10.1371/journal.pone.0117551. http://hdl.handle.net/10261/121368
Aparece en las colecciones: (IAS) Programas informáticos
Ficheros en este ítem:
Fichero Descripción Tamaño Formato  
cropclass_1.savSoftware106,54 kBUnknownVisualizar/Abrir
cropclass_2.savSoftware106,54 kBUnknownVisualizar/Abrir
cropclass_3.savSoftware12,52 kBUnknownVisualizar/Abrir
1_CROPCLASS_Objetives_scope_publications.pdf20,84 kBAdobe PDFVista previa
2_CROPCLASS_IDL_code.pdf82,91 kBAdobe PDFVista previa
2b_CROPCLASS_software_instalation.pdf28,46 kBAdobe PDFVista previa
Mostrar el registro completo

NOTA: Los ítems de Digital.CSIC están protegidos por copyright, con todos los derechos reservados, a menos que se indique lo contrario.