Filter design using radial basis function neural network and genetic algorithm for improved operational health monitoring

The problem of denoising damage indicator signals for improved operational health monitoring of systems is addressed by applying soft computing methods to design filters. Since measured data in operational settings is contaminated with noise and outliers, pattern recognition algorithms for fault detection and isolation can give false alarms. A direct approach to improving the fault detection and isolation is to remove noise and outliers from time series of measured data or damage indicators before performing fault detection and isolation. Many popular signal-processing approaches do not work well with damage indicator signals, which can contain sudden changes due to abrupt faults and non-Gaussian outliers. Signal-processing algorithms based on radial basis function (RBF) neural network and weighted recursive median (WRM) filters are explored for denoising simulated time series. The RBF neural network filter is developed using a K-means clustering algorithm and is much less computationally expensive to develop than feedforward neural networks trained using backpropagation. The nonlinear multimodal integer-programming problem of selecting optimal integer weights of the WRM filter is solved using genetic algorithm. Numerical results are obtained for helicopter rotor structural damage indicators based on simulated frequencies. Test signals consider low order polynomial growth of damage indicators with time to simulate gradual or incipient faults and step changes in the signal to simulate abrupt faults. Noise and outliers are added to the test signals. The WRM and RBF filters result in a noise reduction of 54 - 71 and 59 - 73% for the test signals considered in this study, respectively. Their performance is much better than the moving average FIR filter, which causes significant feature distortion and has poor outlier removal capabilities and shows the potential of soft computing methods for specific signal-processing applications. (C) 2005 Elsevier B. V. All rights reserved.

Hantavirus glycoprotein GN in virion assembly and regulation of interferon response

Hantaviruses have a tri-segmented negative-stranded RNA genome. The S segment encodes the nucleocapsid protein (N), M segment two glycoproteins, Gn and Gc, and the L segment the RNA polymerase. Gn and Gc are co-translationally cleaved from a precursor and targeted to the cis-Golgi compartment. The Gn glycoprotein consists of an external domain, a transmembrane domain and a C-terminal cytoplasmic domain. In addition, the S segment of some hantaviruses, including Tula and Puumala virus, have an open reading frame (ORF) encoding a nonstructural potein NSs that can function as a weak interferon antagonist. The mechanisms of hantavirus-induced pathogenesis are not fully understood but it is known that both hemorrhagic fever with renal syndrome (HFRS) and hantavirus (cardio) pulmonary syndrome (HCPS) share various features such as increased capillary permeability, thrombocytopenia and upregulation of TNF-. Several hantaviruses have been reported to induce programmed cell death (apoptosis), such as TULV-infected Vero E6 cells which is known to be defective in interferon signaling. Recently reports describing properties of the hantavirus Gn cytoplasmic tail (Gn-CT) have appeared. The Gn-CT of hantaviruses contains animmunoreceptor tyrosine-based activation motif (ITAM) which directs receptor signaling in immune and endothelial cells; and contain highly conserved classical zinc finger domains which may have a role in the interaction with N protein. More functions of Gn protein have been discovered, but much still remains unknown. Our aim was to study the functions of Gn protein from several aspects: synthesis, degradation and interaction with N protein. Gn protein was reported to inhibit interferon induction and amplication. For this reason, we also carried out projects studying the mechanisms of IFN induction and evasion by hantavirus. We first showed degradation and aggresome formation of the Gn-CT of the apathogenic TULV. It was reported earlier that the degradation of Gn-CT is related to the pathogenicity of hantavirus. We found that the Gn-CT of the apathogenic hantaviruses (TULV, Prospect Hill virus) was degraded through the ubiquitin-proteasome pathway, and TULV Gn-CT formed aggresomes upon treatment with proteasomal inhibitor. Thus the results suggest that degradation and aggregation of the Gn-CT may be a general property of most hantaviruses, unrelated to pathogenicity. Second, we investigated the interaction of TULV N protein and the TULV Gn-CT. The Gn protein is located on the Golgi membrane and its interaction with N protein has been thought to determine the cargo of the hantaviral ribonucleoprotein which is an important step in virus assembly, but direct evidence has not been reported. We found that TULV Gn-CT fused with GST tag expressed in bacteria can pull-down the N protein expressed in mammalian cells; a mutagenesis assay was carried out, in which we found that the zinc finger motif in Gn-CT and RNA-binding motif in N protein are indispensable for the interaction. For the study of mechanisms of IFN induction and evasion by Old World hantavirus, we found that Old World hantaviruses do not produce detectable amounts of dsRNA in infected cells and the 5 -termini of their genomic RNAs are monophosphorylated. DsRNA and tri-phosphorylated RNA are considered to be critical activators of innate immnity response by interacting with PRRs (pattern recognition receptors). We examined systematically the 5´-termini of hantavirus genomic RNAs and the dsRNA production by different species of hantaviruses. We found that no detectable dsRNA was produced in cells infected by the two groups of the old world hantaviruses: Seoul, Dobrava, Saaremaa, Puumala and Tula. We also found that the genomic RNAs of these Old World hantaviruses carry 5´-monophosphate and are unable to trigger interferon induction. The antiviral response is mainly mediated by alpha/beta interferon. Recently the glycoproteins of the pathogenic hantaviruses Sin Nombre and New York-1 viruses were reported to regulate cellular interferon. We found that Gn-CT can inhibit the induction of IFN activation through Toll-like receptor (TLR) and retinoic acid-inducible gene I-like RNA helicases (RLH) pathway and that the inhibition target lies at the level of TANK-binding kinase 1 (TBK-1)/ IKK epislon complex and myeloid differentiation primary response gene (88) (MyD88) / interferon regulatory factor 7 (IRF-7) complex.

Agglomerative clustering using the concept of mutual nearest neighbourhood

A method for determining the mutual nearest neighbours (MNN) and mutual neighbourhood value (mnv) of a sample point, using the conventional nearest neighbours, is suggested. A nonparametric, hierarchical, agglomerative clustering algorithm is developed using the above concepts. The algorithm is simple, deterministic, noniterative, requires low storage and is able to discern spherical and nonspherical clusters. The method is applicable to a wide class of data of arbitrary shape, large size and high dimensionality. The algorithm can discern mutually homogenous clusters. Strong or weak patterns can be discerned by properly choosing the neighbourhood width.

An algorithm to find distant repeats in a pair of protein sequences

Distant repeats between a pair of protein sequences can be exploited to study the various aspects of proteins such as structure-function relationship, disorders due to protein malfunction, evolutionary analysis, etc. An in-depth analysis of the distant repeats would facilitate to establish a stable evolutionary relation of the repeats with respect to their three-dimensional structure. To this effect, an algorithm has been devised to identify the distant repeats in a pair of protein sequences by essentially using the scores of PAM (Percent Accepted Mutation) matrices. The proposed algorithm will be of much use to researchers involved in the comparative study of various organisms based on the amino-acid repeats in protein sequences. (C) 2010 Elsevier B.V. All rights reserved.

A comparison between conceptual clustering and conventional clustering

Clustering is a process of partitioning a given set of patterns into meaningful groups. The clustering process can be viewed as consisting of the following three phases: (i) feature selection phase, (ii) classification phase, and (iii) description generation phase. Conventional clustering algorithms implicitly use knowledge about the clustering environment to a large extent in the feature selection phase. This reduces the need for the environmental knowledge in the remaining two phases, permitting the usage of simple numerical measure of similarity in the classification phase. Conceptual clustering algorithms proposed by Michalski and Stepp [IEEE Trans. PAMI, PAMI-5, 396–410 (1983)] and Stepp and Michalski [Artif. Intell., pp. 43–69 (1986)] make use of the knowledge about the clustering environment in the form of a set of predefined concepts to compute the conceptual cohesiveness during the classification phase. Michalski and Stepp [IEEE Trans. PAMI, PAMI-5, 396–410 (1983)] have argued that the results obtained with the conceptual clustering algorithms are superior to conventional methods of numerical classification. However, this claim was not supported by the experimental results obtained by Dale [IEEE Trans. PAMI, PAMI-7, 241–244 (1985)]. In this paper a theoretical framework, based on an intuitively appealing set of axioms, is developed to characterize the equivalence between the conceptual clustering and conventional clustering. In other words, it is shown that any classification obtained using conceptual clustering can also be obtained using conventional clustering and vice versa.

A divisive scheme for constructing minimal spanning trees in coordinate space

An algorithm to generate a minimal spanning tree is presented when the nodes with their coordinates in some m-dimensional Euclidean space and the corresponding metric are given. This algorithm is tested on manually generated data sets. The worst case time complexity of this algorithm is O(n log2n) for a collection of n data samples.

A near-optimal initial seed value selection in K-means means algorithm using a genetic algorithm

The K-means algorithm for clustering is very much dependent on the initial seed values. We use a genetic algorithm to find a near-optimal partitioning of the given data set by selecting proper initial seed values in the K-means algorithm. Results obtained are very encouraging and in most of the cases, on data sets having well separated clusters, the proposed scheme reached a global minimum.

Attitude estimation using moment invariants

The problem of estimating the three-dimensional rotational parameters of a rigid body from its monocular image data has been considered using the method of moment invariants. Second- and third-order moment invariants are used to construct the feature vector for the scale and orientation independent identification of the camera view axis direction in the body-fixed reference frame. The camera rotation angle about the view axis is derived from second-order central moments. The relative attitude of the rigid body is then expressed in terms of quaternion parameters to model the outputs of a video sensor in attitude control simulations. Experimental results and simulation outputs are presented using the mathematical model of a spacecraft.

Moments estimation in Radon space

In this paper, we show a method of obtaining general and orthogonal moments, specifically Legendre and Zernicke moments, from the Radon Transform data of a two-dimensional function. The regular or geometric moments are first evaluated directly from the projection data and the orthogonal moments are derived from these regular moments.

Fast computation of Legendre and Zernike moments

This paper presents recursive algorithms for fast computation of Legendre and Zernike moments of a grey-level image intensity distribution. For a binary image, a contour integration method is developed for the evaluation of Legendre moments using only the boundary information. A method for recursive calculation of Zernike polynomial coefficients is also given. A square-to-circular image transformation scheme is introduced to minimize the computation involved in Zernike moment functions. The recursive formulae can also be used in inverse moment transforms to reconstruct the original image from moments. The mathematical framework of the algorithms is given in detail, and illustrated with binary and grey-level images.

Learning the global maximum with parameterized learning automata

A feedforward network composed of units of teams of parameterized learning automata is considered as a model of a reinforcement teaming system. The internal state vector of each learning automaton is updated using an algorithm consisting of a gradient following term and a random perturbation term. It is shown that the algorithm weakly converges to a solution of the Langevin equation implying that the algorithm globally maximizes an appropriate function. The algorithm is decentralized, and the units do not have any information exchange during updating. Simulation results on common payoff games and pattern recognition problems show that reasonable rates of convergence can be obtained.

Off-line signature verification using genetically optimized weighted features

This paper is concerned with off-line signature verification. Four different types of pattern representation schemes have been implemented, viz., geometric features, moment-based representations, envelope characteristics and tree-structured Wavelet features. The individual feature components in a representation are weighed by their pattern characterization capability using Genetic Algorithms. The conclusions of the four subsystems teach depending on a representation scheme) are combined to form a final decision on the validity of signature. Threshold-based classifiers (including the traditional confidence-interval classifier), neighbourhood classifiers and their combinations were studied. Benefits of using forged signatures for training purposes have been assessed. Experimental results show that combination of the Feature-based classifiers increases verification accuracy. (C) 1999 Pattern Recognition Society. Published by Elsevier Science Ltd. All rights reserved.

Tree structure for efficient data mining using rough sets

In data mining, an important goal is to generate an abstraction of the data. Such an abstraction helps in reducing the space and search time requirements of the overall decision making process. Further, it is important that the abstraction is generated from the data with a small number of disk scans. We propose a novel data structure, pattern count tree (PC-tree), that can be built by scanning the database only once. PC-tree is a minimal size complete representation of the data and it can be used to represent dynamic databases with the help of knowledge that is either static or changing. We show that further compactness can be achieved by constructing the PC-tree on segmented patterns. We exploit the flexibility offered by rough sets to realize a rough PC-tree and use it for efficient and effective rough classification. To be consistent with the sizes of the branches of the PC-tree, we use upper and lower approximations of feature sets in a manner different from the conventional rough set theory. We conducted experiments using the proposed classification scheme on a large-scale hand-written digit data set. We use the experimental results to establish the efficacy of the proposed approach. (C) 2002 Elsevier Science B.V. All rights reserved.