Supervised classification is based on statistical identification function , according to typical sample
training methods. In other words, according to the sample that known training areas provide, by selecting
he paramenters,obtained characteristic paramenters as the decision rules. Developing the discriminant
function to classify images of the image classification is a method of pattern recognition. Unsupervised
classification is known as the cluster analysis. A general clustering algorithm firstly selects the model
number of points as the cluster center. Each center represents a category, according to some similarity
measure, makes each pattern attribute to the cluster centers represented by the type.