3. Super-resolution in multispectral data using neural networks is a novel procedure, which aims in increasing the spatial resolution of multispectral data, and simultaneously creating a high quality RGB fused representation. For this purpose, neural networks are employed and a successive training procedure is applied in order to incorporate in the network structure knowledge about recovering lost frequencies. MERIS multispectral data are employed to demonstrate the performance of the proposed method.

The target image used in the training procedure is an RGB image resulting by fusion of MERIS bands Cross-validation of the neural network performance in different regions has proved the generalization efficiency of this approach