Utilize este identificador para referenciar este registo: http://hdl.handle.net/10400.5/8294
Título: Automatizing photo interpretation of satellite imagery in the context of the Common Agriculture Policy subsidy control
Autor: Schmedtmann, Jonas
Orientador: Campagnolo, Manuel Lameiras Figueiredo
Palavras-chave: Common Agricultural Policy
subsidy control
Landsat
multitemporal analysis
operational crop discrimination
parcel-based classification
remote sensing
Data de Defesa: 2014
Editora: ISA/UL
Citação: Schmedtmann. J. - Automatizing photo interpretation of satellite imagery in the context of the Common Agriculture Policy subsidy control. Lisboa: ISA, 2014, 64 p.
Resumo: Computer Assisted Photo-Interpretation (CAPI) uses remotely sensed imagery to control farmers’ subsidy applications in the context of the EU’s Common Agriculture Policy. A simple and reproducible method to automatize CAPI in an operational context with the overreaching goal to reduce control costs and completion time was developed in this study. Validated control data provided by the Portuguese Control and Paying Agency for Agriculture (IFAP) and a multispectral atmospherically corrected Landsat ETM+ time series were used to calibrate and test the method. Taking advantage of the nature of subsidy declarations, object-based land cover classification for the 12 most controlled classes was carried out in the region of Ribatejo. The main feature of the presented method is that it allows choosing a confidence level on the automatic classification of farmers’ parcels. While higher confidence levels reduce the risk of misclassifications, lower levels increase the number of automatic control decisions. A confidence level of 80% is a good compromise. This confidence level leads to over 55% of automatically taken control decisions with an overall accuracy of 84%. Furthermore, over 85% of all parcels classified as maize, rice, wheat or vineyard can be controlled by the method with the optimal confidence level.
Descrição: Mestrado em Engenharia do Ambiente - Instituto Superior de Agronomia
URI: http://hdl.handle.net/10400.5/8294
Aparece nas colecções:BISA - Dissertações de Mestrado / Master Thesis

Ficheiros deste registo:
Ficheiro Descrição TamanhoFormato 
TESE v2.0 CD.pdf11,01 MBAdobe PDFVer/Abrir


FacebookTwitterDeliciousLinkedInDiggGoogle BookmarksMySpace
Formato BibTex MendeleyEndnote Degois 

Todos os registos no repositório estão protegidos por leis de copyright, com todos os direitos reservados.