Pierre Lorrentz

A Novel Weightless Artificial Neural Based Multi-Classifier for Complex Classifications

Artificial neural systems in general and weightless systems in particular, have traditionally struggled in performance terms when confronted with problem domains such as possessing a large number of independent pattern classes and pattern classes with non-standard distributions. A multi-classifier is proposed which explores problem domains with a large number of independent pattern classes typically found in forensic and security databases. Specifically, the multi-classifier system is demonstrated on the exemplar of fingerprint identification problem typical to forensic, biometric, and security. Furthermore, the multi-classifier is able to provide a reasonable solution to benchmark problems from medicinal and physical (science) fields, which are determining the health, state of thyroid glands and determining whether or not there is a structure in the ionosphere, respectively.

An FPGA based Adaptive Weightless Neural Network Hardware

This paper explores the significant practical difficulties inherent in mapping large artificial neural structures onto digital hardware. Specifically, a class of weightless neural architecture called the enhanced probabilistic convergent network is examined due to the inherent simplicity of the control algorithms associated with the architecture. The advantages for such an approach follow from the observation that, for many situations for which an intelligent machine requires very fast, unmanned, and uninterrupted responses, a PC-based system is unsuitable especially in electronically harsh and isolated conditions, The target architecture for the design is an FPGA, the Virtex-II pro which is statically and dynamically reconfigurable, enhancing its suitability for an adaptive weightless neural networks. This hardware is tested on a benchmark of unconstrained handwritten numbers from the National Institute Of Standards And Technology (NIST), USA.

A Fingerprint Identification System Using Adaptive FPGA-Based Enhanced Probabilistic Convergent Network

This paper explores the biometric identification and verification of human subjects via fingerprints utilising an adaptive FPGA-based weightless neural networks. The exploration espoused here is a hardware-based system motivated by the need for accurate and rapid response to identification of fingerprints which may be lacking in other alternative systems such as software based neural networks. The fingerprints are pre-processed and binarized, and the binarized fingerprints are partitioned into train- and test-set for the FPGA-based neural network. The neural network employed in this exploration is known as Enhanced Convergent Network (EPCN). The results obtained are compared to other alternative systems. They demonstrate the suitability of the FPGA-based EPCN for such tasks.

The FPGA-based multi-classifier

To achieve intelligent real-time processing, the paper presents a weightless neural-based cognitive system which is capable of classifying, analysing, and prediction from sight or sound. The proposed cognitive system fused recursive-least-square (RLS) filters in parallel with an enhanced probabilistic convergent network (EPCN) serially—implemented on field programmable gate array. The novelty is that EPCN does not require an optimum result from RLS to achieve good responses from the RLS–EPCN fusion, thereby further offering two main distinguishing features: compactness and speed. Test results demonstrate RLS–EPCN’s suitability to exploration/exploitation of hostile surroundings such as sea exploration.

An advanced combination strategy for multi-classifiers employed in large multi-class problem domains

Traditional artificial neural architectures possess limited ability to address the scale problem exhibited by a large number of distinct pattern classes and limited training data. To address these problems, this paper explores a novel advanced encoding scheme, which reduces both memory demand and execution time, and provides improved performance. The novel advanced encoding scheme known as the engine encoding, have been implemented in a multi-classifier, which combines the scaled probabilities, configuration information, and the discriminating abilities of the associated component classifiers. The problems of overloading and saturation experienced by traditional networks are solved by training the base classifiers on differing sub-sets of the required pattern classes and allowing the combiner classifier to derive a solution. Current Multi-classifier Systems are easily biased when trained on one class more often than another class, when patterns representing a class are very large compared to the rest, or when the multi-classifier depends on a certain fixed order of arrangement of pattern classes. A unique statistical arrangement method is hereby presented which aims to solve the bias problem. This statistical arrangement method also enhances independence of component classifiers. The system is demonstrated on the exemplar of fingerprint identification and utilizes a Weightless Neural System called the Enhanced Probabilistic Convergent Neural Network (EPCN) in a Multi-classifier System.