This algorithm uses less nodes as the number of training samples. The centers are determined in two steps using the K-means and EM algorithms. The output node does not have transfer function its role is only the pure summation.
The other way is to think of neuronal weights as pointers to the input space. They form a discrete approximation of the distribution of training samples.
Maximum likelihood classifier was trained by ground truth information collected from DFW of the area and the training samples were validated using a minimum distance classifier.
 See also: Class, Analysis, Map, Mapping, Image
  
|
RSS
