ABSTRACT
In this chapter, we adopt user-defined salient image features as training examples for a specially designed model-based neural network to perform feature detection. Specifically, we have investigated an alternative MBNN with hierarchical architecture [48, 62, 128] which is capable of learning user-specified features through an interactive process. It will be shown that this network is capable of generalizing from the set of specified features to identify similar features in images which have not been included in the training data.
