Fingerprint Retrieval Using a Specialized Ensemble of Attractor Networks

Abstract

We tested the performance of the Ensemble of Attractor Neural Networks (EANN) model for fingerprint learning and retrieval. The EANN model has proved to increase the random patterns storage capacity, when compared to a single attractor of equal connectivity. In this work, we tested the EANN with real patterns, i.e. fingerprints dataset. The EANN improved the retrieval performance for real patterns more than tripling the capacity of the single attractor with the same number of connections. The EANN modules can also be specialized for different patterns sets according to their characteristics, i.e. pattern/network sparseness (activity). Three EANN modules were assigned with skeletonized fingerprints (low activity), binarized (original) fingerprints (medium activity), and dilated/thickened fingerprint (high activity), and their retrieval was checked. The more sparse the code the larger the storage capacity of the module. The EANN demonstrated to improve the retrieval capacity of the single network, and it can be very helpful for module specialization for different types of real patterns.