Object Classification

The different objects in an image have different characteristics represented by their features. Given a set of features of an object, comparing that with those in the data base and assigning it to its proper class is classification. There are two main types of classification: (i) supervised classification and (ii) unsupervised classification. In supervised classification, features are specified apriori and objects are classified using them. Typical methods used are minimum distance, k-nearest neighbors, decision trees, and statistical (based on probability distribution models). The decision is prior. In unsupervised classification, we classify the objects by the constraints imposed by the features. The decision is posterior.