Bayes risk weighted vector quantization with CART estimated class posteriors

Two new methods for estimating the class posterior probabilities required for the Bayes risk computation can be estimated based on a labeled training sequence are introduced and two types of tree-structured estimators are constructed.

Abstract

A Bayes risk weighted vector quantizer (Bayes VQ) combines compression and low-level classification of images by incorporating a Bayes risk component into the distortion measure used to design the code. The class posterior probabilities required for the Bayes risk computation can be estimated based on a labeled training sequence. We introduce two new methods for estimating these posteriors. In particular, two types of tree-structured estimators are constructed by applying the classification and regression tree algorithm CART to eight features of the training sequence. We apply the resulting Bayes VQ systems to aerial photographs where the goal is to compress the images and classify man-made and natural regions. These systems provide classification superior to that of previous work with Bayes VQ while maintaining similar compression performance. The systems also provide moderate to substantial improvement in classification with only a small loss in compression to performance obtained with a modified version of Kohonen's (1988) "learning vector quantizer" and with an independent design of quantizer and classifier.