Remotely sensed data analysis using two neural networks and its application to land cover mapping

In recent works, the authors have proposed a hybrid system using a Kohonen's self-organization feature mapping preprocessor (SOM) and a multi-layered neural network processor (BPM) to analyze remotely sensed data, and demonstrated the applicability of SOM preprocessor by a principal component analysis (PCA). In the present paper, the authors empirically examine the significance of the principal components for the input pattern.