The detection of geometric anomalies on fractal images with the ALIAS image processing engine

Based on collective learning systems theory and a versatile general-purpose architecture for massively parallel networks of processors, a transputer-based parallel-processing image-processing system, known as ALIAS (Adaptive Learning Image Analysis System), has been applied to a difficult image processing task: the phase, translation, and scale invariant detection of anomalous features, in otherwise `normal´ images. ALIAS consists of a three-layer hierarchical network of 32 learning cells and 33 non-learning cells. After being trained with a set of `normal´ images, when presented with an anomalous input image, ALIAS automatically generates an isomorphic output image (Super Hypothesis) in which each pixel represents the `normality´ of the associated pixel in the input image. The paper presents an evaluation of the performance of ALIAS for the detection of anomalies which are square sections of a class of geometric anomalies (called Manhattan) placed on binary fractal images belonging to one of three different `normal´ equivalence classes (called Islands, Plateaus, and Wheatfields)