2D shape recognition using recursive determination of landmark and fuzzy ART network learning

In this paper, 2-D shape recognition is done using a combination of recursive search of landmarks, landmark-based invariant features, and a Fuzzy-ART neural-network classifier. To make this novel combination work well, a upper limit is imposed on the number of total landmarks allowed, and this maximum size is then translated into fixed dimensions of invariant features and into the neural processing of the features. It is shown that the recursive landmark search approximates very well any smooth 2-D shape contour, that the shape features used are independent of perspective transformation, and that, when combined with a Fuzzy-ART classifier, unknown features can be efficiently learned on-line to identify multiple distinct objects. An illustrative example is used to demonstrate effectiveness of the proposed algorithm.