Remotely Piloted Aircraft Systems and Object-oriented classification: an approach into the submetric analysis of a suburban area

Abstract

Peralta, C., Solórzano, J.V., Fernández, A., Gallardo-Cruz, J.A. 2016. Remotely Piloted Aircraft Systems and Object-oriented classification: an approach into the submetric analysis of a suburban area. Ecosistemas 25(1): 90-93. Doi.: 10.7818/ECOS.2016.25-1.13

RPAS (Remotely Piloted Aircraft Systems) have opened the possibility to obtain submetric multispectral images with a high quality at a relative low cost. The use of this type of imagery together with precise classification methods, such as the object-oriented approach, should derive excellent results. Thus, this study aimed at evaluating the potential of this type of classification using a submetric image obtained with a RPAS to obtain a thematic map of a suburban scene. The classification was successful (total reliability = 0.83), and thirteen focal classes and one without classification were distinguished. In the classification, urban areas with different heights were discriminated, as well as five types of vegetation and water bodies.