E-viri
-
Lugonja, P.; Panic, M.; Minic, V.; Culibrk, D.; Crnojevic, V.
2012 20th Telecommunications Forum (TELFOR), 2012-Nov.Conference Proceeding
Multispectral remote sensing data are rich source of information for precision agriculture and earth observation that requires advanced methods for its interpretation. In this paper we addressed the problem of crop classification on multispectral images. The aim is to learn classifier to discriminate between 6 crop types. Different techniques in learning classifiers were employed in order to achieve better accuracy and generalization. We compared obtained results and selected those with potential practical usage. In the light of increasing demand for the extraction of information from remotely collected data, our work contributes to the development of remote sensing inagriculture.
