Signature verification revisited: promoting practical exploitationof biometric technology

Despite research over a long period, biometric approaches to authenticating personal identity have not met with the degree of success in practical applications originally predicted. This paper discusses approaches to biometric testing, focusing particularly on automatic signature verification, and addresses some of the important issues which might help to promote the introduction of practical systems in the future. Examples associated with the adoption of an explicitly flexible approach to signature verification are used to illustrate the discussion, and it is argued that there is still considerable potential for practical exploitation of this type of technology     

Sequence recognition with an adaptive-logic network

Dynamic nets of random-access memory elements used in an adaptive-logic mode and whose output is fed back to be ORed with the input are used for recognising input sequences. Experimental results are presented, and the conditions required for success are discussed.     

Efficient detection of spurious inputs for improving the robustness of MLP networks in practical applications

The problem of the rejection of patterns not belonging to identified training classes is investigated with respect to Multilayer Perceptron Networks (MLP). The reason for the inherent unreliability of the standard MLP in this respect is explained, and some mechanisms for the enhancement of its rejection performance are considered. Two network configurations are presented as candidates for a more reliable structure, and are compared to the so-called negative training approach. The first configuration is an MLP which uses a Gaussian as its activation function, and the second is an MLP with direct connections from the input to the output layer of the network. The networks are examined and evaluated both through the technique of network inversion, and through practical experiments in a pattern classification application. Finally, the model of Radial Basis Function (RBF) networks is also considered in this respect, and its performance is compared to that obtained with the other networks described.     

Metal organic frameworks for hydrogen purification

High purity hydrogen is one of the key factors in determining the lifetime of proton exchange membrane (PEM) fuel cells. However, the current industrial processes for producing high purity hydrogen are not only expensive, but also come with low energy efficiencies and productivity. Finding more cost-effective methods of purifying hydrogen is essential for ensuring wider scale deployment of PEM fuel cells. Among various hydrogen purification methods, adsorption in porous materials and membrane technologies are seen as two of the most promising candidates for the current industrial hydrogen purification methods, with metal organic frameworks (MOF) being particularly popular in research over the last decade. Despite many available reviews on MOFs, most focus on synthesis and production, with few reports focused on performance for hydrogen purification. This review describes the working principle and performance parameters of adsorptive separations and membrane materials and identifies MOFs that have been reported for hydrogen purification. The MOFs are summarised and their performance in separating hydrogen from common impurities (CO₂, N₂, CH₄, CO) is compared systematically. The challenges of commercial application of MOFs for hydrogen purification are discussed.     

Application of the Convergence-Confinement method of tunnel design to rock masses that satisfy the Hoek-Brown failure criterion

This paper discusses the practical application of the Convergence-Confinement Method of tunnel design to rock-masses that satisfy the Hoek-Brown failure criterion. The strength of intact rock and jointed rock-masses, as defined by the Hoek-Brown criterion, and the basis of the Convergence-Confinement method are reviewed. Equations that allow the construction of the three basic components of the Convergence-Confinement method, i) the Longitudinal Deformation Profile (LDP), ii) the Ground Reaction Curve (GRC) and iii) the Support Characteristic Curve (SCC) are given. A practical case of support design for a circular tunnel is discussed and solved using the Convergence Confinement method. A spreadsheet summarizing the implementation of the method is also included. Reference values of typical rock properties and geometrical and mechanical properties for typical support systems are presented in tables and charts.     

A Niching Memetic Algorithm for Simultaneous Clustering and Feature Selection

Clustering is inherently a difficult task, and is made even more difficult when the selection of relevant features is also an issue. In this paper we propose an approach for simultaneous clustering and feature selection using a niching memetic algorithm. Our approach (which we call NMA_CFS) makes feature selection an integral part of the global clustering search procedure and attempts to overcome the problem of identifying less promising locally optimal solutions in both clustering and feature selection, without making any a priori assumption about the number of clusters. Within the NMA_CFS procedure, a variable composite representation is devised to encode both feature selection and cluster centers with different numbers of clusters. Further, local search operations are introduced to refine feature selection and cluster centers encoded in the chromosomes. Finally, a niching method is integrated to preserve the population diversity and prevent premature convergence. In an experimental evaluation we demonstrate the effectiveness of the proposed approach and compare it with other related approaches, using both synthetic and real data.     

Detection of circumscribed masses in digital mammograms usingbehaviour-knowledge space method

In classification applications where reliability and flexibility are priorities it is advantageous to adopt a processing configuration based on the combination of multiple classifiers. The application of classifier combination based on behaviour-knowledge space to the detection of circumscribed masses in mammographic X-ray images is discussed     

An Investigation of the Effects of Variable Vigilance within the RePART Neuro-Fuzzy Network

RePART is a variation of fuzzy ARTMAP to which a reward/punishment concept has been added. Previously, an improvement in performance of RePART had been noted compared with other ARTMAP-based models, such as fuzzy ARTMAP and ARTMAP-IC. In this paper, a wider investigation of RePART performance is described, in which RePART is analysed in relation to a multi-layer perceptron and a RAM-based network in a handwritten numeral recognition task. In the RePART network, a variable vigilance parameter is proposed in order to smooth the poor-generalisation problem of RePART. Firstly, the same vigilance is associated within every neuron – general variable vigilance. Secondly, an individual variable vigilance for each neuron – which takes into account its average and frequency of activation – is used. In a handwritten numeral recognition task using individual variable vigilance, RePART performance improved and demonstrated a performance comparable with alternative architectures such as fuzzy multi-layer perceptron and Radial RAM.     

ICA-based multi-colour space texture classification system

Presented is a novel method which uses independent component analysis (ICA) for systematically partitioning and combining textural features extracted from different colour spaces, in a multiple classifier based system, for colour texture classification. Results obtained illustrate that the proposed ICA-based feature-partitioning and classifier combination system produces more accurate results compared to a system that combines classifiers applied to features extracted from individual colour spaces