Use of IFS codes for learning 2D isolated-object classification systems

Automatic recognition of complex images is a hard and computationally expen sive task, mainly because it is extremely difficult to capture in an automa tic way and with a few features the necessary discriminant information. If such features were available, a proper learning system could be trained to distinguish images of different kinds of objects, starting from a set of la beled examples. In this paper we show that fractal features obtained from I terated Function System encodings capture the kind of information that is n eeded by learning systems and, thus, allow the successful classification of 2-dimensional images of objects. We also present a fractal feature extract ion algorithm and report the classification results obtained on two very di fferent test-beds by applying Machine Learning techniques to sets of encode d images. (C) 2000 Academic Press.