—This paper describes an approach to classification of noisy signals using a technique based on the fuzzy ARTMAP neural network (FAMNN). The proposed method is a modification of ...
Dimitrios Charalampidis, Michael Georgiopoulos, Ta...
Abstract. In this paper, we presen t an algorithm that provides adaptive plasticity in function approximation problems: the deformable (feature) map (DM) algorithm. The DM approach...
Cheap and versatile cameras make it possible to easily and quickly capture a wide variety of documents. However, low resolution cameras present a challenge to OCR because it is vi...
Charles E. Jacobs, Patrice Y. Simard, Paul A. Viol...
: Conventional hardware platforms are far from reaching real-time simulation requirements of complex spiking neural networks (SNN). Therefore we designed an accelerator board with ...
This paper presents a new approach for shape description and invariant recognition by geometric-normalization implemented by neural networks. The neural system consists of a shape...