Parcel based classification for agricultural mapping and monitoring using multi-temporal satellite image sequence

In this paper, we propose a new approach to parcel-based classification of multi-temporal optical satellite imagery with missing data due to clouds and shadows based on vector and raster data fusion in different phase of classification methodology in Ukraine within the JECAM project. For obtaining pixel-based classification map, an ensemble of neural networks, in particular multilayer perceptron (MLPs), is used. The proposed approach is applied for regional scale crop classification using multi-temporal Landsat-8 images for the Kyivska oblast in Ukraine in 2013. The obtained results are also validated through comparison to official statisticsJRC.D.5-Food Securit