Subspace Based Speech Enhancement Using Gaussian Mixture Model

Traditional subspace based speech enhancement (SSE)methods use linear minimum mean square error (LMMSE) estimation that is optimal if the Karhunen Loeve transform (KLT) coefficients of speech and noise are Gaussian distributed. In this paper, we investigate the use of Gaussian mixture (GM) density for modeling the non-Gaussian statistics of the clean speech KLT coefficients. Using Gaussian mixture model (GMM), the optimum minimum mean square error (MMSE) estimator is found to be nonlinear and the traditional LMMSE estimator is shown to be a special case. Experimental results show that the proposed method provides better enhancement performance than the traditional subspace based methods.Index Terms: Subspace based speech enhancement, Gaussian mixture density, MMSE estimation.

Non Identical Duplicate Video Detection using SIFT Method

Non-Identical Duplicate video detection is a challenging research problem. Non-Identical Duplicate video are a pair of videos that are not exactly identical but are almost similar.In this paper, we evaluate two methods - Keyframe -based and Tomography-based methods to determine the Non-Identical Duplicate videos. These two methods make use of the existing scale based shift invariant (SIFT) method to find the match between the key frames in first method, and the cross-sections through the temporal axis of the videos in second method.We provide extensive experimental results and the analysis of accuracy and efficiency of the above two methods on a data set of Non- Identical Duplicate video-pair.

Subspace-based DOA estimation using Fractional Lower Order statistics

Direction Of Arrival (DOA) estimation, using a sensor array, in the presence of non-Gaussian noise using Fractional Lower-Order Moments (FLOM)matrices is studied. In this paper, a new FLOM based technique using the Fractional Lower Order Infinity Norm based Covariance (FLIC) Matrix is proposed. The bounded property and the low-rank subspace structure of the FLIC matrix is derived. Performance of FLIC based DOA estimation using MUSIC, ESPRIT, is shown to be better than other FLOM based methods.

Study of molecular structure and vibrational spectra of poly (beta,-1-malic acid) PMLA] using DFT approach

Poly (beta-L-malic acid) (PMLA) is a biodegradable polymer and it has various important applications in the biomedical field. In the present work the structural and spectral characteristics of PMLA have been studied by methods of infrared. Raman spectroscopy and quantum chemistry. Electrostatic potential surface, optimized geometry, harmonic vibrational frequencies, infrared intensities and activities of Raman scattering were calculated by density functional theory (DFT) using oligomeric approach employing B3LYP with complete relaxation in the potential energy surface using 6-311++G (d, p) basis set. Based on results, we have discussed the correlation between the vibrational modes and the structure of the PMLA. A complete analysis of the experimental infrared and Raman spectra has been reported on the basis of wavenumber of the vibrational bands and potential energy distribution. The calculated HOMO and LUMO energies shows that charge transfer occur within the molecule. The calculated infrared and the Raman spectra of the polymer based on DFT calculations show reasonable agreement with the experimental results. (c) 2012 Elsevier Ltd. All rights reserved.

Intermolecular charge transfer and vibrational analysis of hydrogen bonding in acetazolamide

In the present work the structural and spectral characteristics of acetazolamide have been studied by methods of infrared, Raman spectroscopy and quantum chemistry. Electrostatic potential surface, optimized geometry, harmonic vibrational frequencies, infrared intensities and activities of Raman scattering were calculated by density functional theory (DFT) employing B3LYP with complete relaxation in the potential energy surface using 6-311++G(d,p) basis set. Based on these results, we have discussed the correlation between the vibrational modes and the structure of the dimers of acetazolamide. The calculated vibrational spectra of three dimers of acetazolamide have been compared with observed spectra, and the assignment of observed bands was carried out using potential energy distribution. The observed spectra agree well with the values computed from the OFT. A comparison of observed and calculated vibrational spectra clearly shows the effect of hydrogen bonding. The frequency shifts observed for the different dimers are in accord with the hydrogen bonding in acetazolamide. Natural bond orbital (NBO) analyses reflect the charge transfer interaction in the individual hydrogen bond units and the stability of different dimers of acetazolamide. (C) 2012 Elsevier B.V. All rights reserved.

Process Development for Functionalization of Cotton with Silver Nanoparticles Synthesized by Bio-based Approaches

Cotton is a widely used raw material for textiles but drawbacks regarding their poor mechanical properties often limit their applications as functional materials. The present investigation involved process development for one step coating of cotton with silver nanoparticles (SNP) synthesized using Azadirachta indica and Citrus limon extract to develop functional textiles. Addition of starch to functional textiles led to efficient binding of nanoparticles to fabric and led to drastic decrease in release of silver from fabricated textiles after ten washing cycles enhancing their environment friendliness. Differential scanning calorimetry, scanning electron microscopy, FT-IR analysis and mechanical studies demonstrated efficient binding of nanoparticles to fabric through bio-based processes. The functionalized textiles developed by the bio-based methods showed significant antibacterial activity against E. coli and S. aureus (with 99% microbial reduction). Present work offers a simple procedure for coating SNP using bio-based approaches with promising applications in specialized functions.

Pattern-growth based frequent serial episode discovery

Frequent episode discovery is a popular framework for pattern discovery from sequential data. It has found many applications in domains like alarm management in telecommunication networks, fault analysis in the manufacturing plants, predicting user behavior in web click streams and so on. In this paper, we address the discovery of serial episodes. In the episodes context, there have been multiple ways to quantify the frequency of an episode. Most of the current algorithms for episode discovery under various frequencies are apriori-based level-wise methods. These methods essentially perform a breadth-first search of the pattern space. However currently there are no depth-first based methods of pattern discovery in the frequent episode framework under many of the frequency definitions. In this paper, we try to bridge this gap. We provide new depth-first based algorithms for serial episode discovery under non-overlapped and total frequencies. Under non-overlapped frequency, we present algorithms that can take care of span constraint and gap constraint on episode occurrences. Under total frequency we present an algorithm that can handle span constraint. We provide proofs of correctness for the proposed algorithms. We demonstrate the effectiveness of the proposed algorithms by extensive simulations. We also give detailed run-time comparisons with the existing apriori-based methods and illustrate scenarios under which the proposed pattern-growth algorithms perform better than their apriori counterparts. (C) 2013 Elsevier B.V. All rights reserved.

Theory and Application of the Transient Injection Well Test

Based on the theory of the pumping well test, the transient injection well test was suggested in this paper. The design method and the scope of application are discussed in detail. The mathematical models are developed for the short-time and long-time transient injection test respectively. A double logarithm type curve matching method was introduced for analyzing the field transient injection test data. A set of methods for the transient injection test design, experiment performance and data analysis were established. Some field tests were analyzed, and the results show that the test model and method are suitable for the transient injection test and can be used to deal with the real engineering problems.

Growth, mortality and yield-per-recruit of the poor cod, Trisopterus minutus capelanus, from the Strait of Sicily

Length-based methods (LBMs) were used to study the growth of Trisopterus minutus capelanus in the Strait of Sicily (Messina Strait). A total of 16,304 'merluzzetto' or poor cod collected by experimental trawling off the southern coast of Sicily during spring, summer, autumn 1986 and winter 1987 were measured in order to estimate the length structure of the population. Length-frequency distribution were analyzed and normal components were discriminated. Von Bertalanffy growth parameters were derived from the mean length of the normal components. The growth parameters obtained by weighted non-linear regression were: K=0.462 (yr super(1)), L sub( infinity )=222.3 (TL,mm) and t sub(o)=-0.679 yr. The resulting growth performance index ( Phi ') was 4.36, a value slightly lower than those derived for Western Mediterranean (mean Phi '=4.45) and Adriatic ( Phi '=4.58) populations and slightly higher than that derived for Hellenic waters ( Phi '=4.27). On the basis of the von Bertalanffy parameters estimated, an array of age-specific instantaneous natural mortality rate (M sub(t)=0.5-1.1) and an average value of total natural mortality rate (Z=2.1 yr super(1)) were estimated and used in the Thompson and Bell yield per recruit (Y/R) analysis in order to evaluate the status of the fishery and forecast the effects of changes in the fishing pattern. Results indicate that this resource is overexploited and that Y/R could be increased by postponing the age at first capture from 0.5 to 1.0 yr. Even a slight reduction in fishing mortality could improve the performance of the fishery. At the present level of exploitation, and assuming a constant recruitment, the spawning stock biomass per recruit (SPR) is well below the conservative threshold of 30% of the pristine or unexploited SPR.

Localização de Terminais Móveis utilizando Correlação de Assinaturas de Rádio-Frequência.

Nesta dissertação são analisados métodos de localização baseados na rede, com destaque para os métodos de correlação de assinaturas de rádio-frequência (DCM - Database Correlation Methods). Métodos baseados na rede não requerem modificações nos terminais móveis (MS - Mobile Stations), sendo portanto capazes de estimar a localização de MS legados, i.e., sem suporte específico a posicionamento. Esta característica, associada a alta disponibilidade e precisão dos métodos DCM, torna-os candidatos viáveis para diversas aplicações baseadas em posição, e em particular para a localização de chamadas para números de emergência - polícia, defesa civil, corpo de bombeiros, etc. - originadas de telefones móveis celulares. Duas técnicas para diminuição do tempo médio para produção de uma estimativa de posição são formuladas: a filtragem determinística e a busca otimizada utilizando algoritmos genéticos. Uma modificação é realizada nas funções de avaliação utilizadas em métodos DCM, inserindo um fator representando a inacurácia intrínseca às medidas de nível de sinal realizadas pelos MS. As modificações propostas são avaliadas experimentalmente em redes de telefonia móvel celular de segunda e terceira gerações em ambientes urbanos e suburbanos, assim como em redes locais sem fio em ambiente indoor. A viabilidade da utilização de bancos de dados de correlação (CDB - Correlation Database) construídos a partir de modelagem de propagação é analisada, bem como o efeito da calibração de modelos de propagação empíricos na precisão de métodos DCM. Um dos métodos DCM propostos, utilizando um CDB calibrado, teve um desempenho superior ao de vários outros métodos DCM publicados na literatura, atingindo em área urbana a precisão exigida dos métodos baseados na rede pela regulamentação FCC (Federal Communications Commission) para o serviço E911 (Enhanced 911 ).

PCR-SSCP and Isoelectric Focusing as Screening Methods for Identification of (Sub-) Tropical Fish Species

To ensure the authentication of fishery products lacking biological characters, rapid species identification methods are required. Two DNA- and protein-based methods, PCR-SSCP (polymerase chain reaction - single strand conformation polymorphism) of a 464 bp segment of the cytochrome b – gene and isoelectric focusing (IEF) of water-soluble proteins from fish fillets, were applied to identify fillets of (sub-) tropical fish species available on the European market. Among the samples analysed were two taxonomically identified species from the family Sciaenidae and one from Sphyraenidae. By comparison of DNA- and protein patterns of different samples, information about intra-species variability of patterns, and homogeneity of batches (e.g. fillet blocks or bags) can be obtained. PCR-SSCP and IEF may be useful for pre-checking of a large number of samples by food control laboratories. Zusammenfassung Zur Sicherstellung der Authentizität von Fischerei-Erzeugnissen ohne biologische Merkmale sind schnelle Verfahren zur Speziesidentifizierung hilfreich. Zwei Methoden der DNA- bzw. Protein-Analyse wurden eingesetzt, um Filets (sub-) tropischer Fischarten, die auf dem europäischen Markt angeboten werden, zu identifizieren. Bei diesen Methoden handelt es sich um die PCR-SSCP (Polymerase-Kettenreaktion – Einzelstrang-Konformationspolymorphismus) – Analyse der PCR-Produkte und die IEF (isoelektrische Fokussierung) der wasserlöslichen Fischmuskelproteine. Unter den untersuchten Proben waren zwei taxonomisch bestimmte Arten aus der Familie Sciaenidae und eine Spezies aus der Familie Sphyraenidae. Durch Vergleich der DNA- bzw. Proteinmuster lassen sich Informationen über die intra-spezifische Variabilität solcher Muster und die Einheitlichkeit von Partien (beispielsweise Filetblöcke oder Filetbeutel) gewinnen. PCR-SSCP und IEF können in Laboratorien der Lebensmittelüberwachung als Vortest gerade bei hohen Probenzahlen sinnvoll eingesetzt werden.

Demographic assessment of the blue crab (Callinectes sapidus) in Chesapeake Bay using extractable lipofuscins as age markers

The blue crab (Callinectes sapidus) plays an important economic and ecological role in estuaries and coastal habitats from the Gulf of Mexico to the east coast of North America, but demographic assessments are limited by length-based methods. We applied an alternative aging method using biochemical measures of metabolic byproducts (lipofuscins) sequestered in the neural tissue of eyestalks to examine population age structure. From Chesapeake Bay, subsamples of animals collected from the 1998–99 (n=769) and 1999–2000 (n=367) winter dredge surveys were collected and lipofuscin was measured. Modal analysis of the lipofuscin index provided separation into three modes, whereas carapace-width data collected among the same individuals showed two broad modes. Lipofuscin modal analysis indicated that most adults (carapace width >120 mm) were <2 years old. The results indicate that use of extractable lipofuscin can provide a more accurate and better resolved estimation of demographic structure of blue crab populations in the field than size alone.

Physiological modelling for improved reliability in silhouette-driven gradient-based hand tracking

We present a gradient-based motion capture system that robustly tracks a human hand, based on abstracted visual information - silhouettes. Despite the ambiguity in the visual data and despite the vulnerability of gradient-based methods in the face of such ambiguity, we minimise problems related to misfit by using a model of the hand's physiology, which is entirely non-visual, subject-invariant, and assumed to be known a priori. By modelling seven distinct aspects of the hand's physiology we derive prior densities which are incorporated into the tracking system within a Bayesian framework. We demonstrate how the posterior is formed, and how our formulation leads to the extraction of the maximum a posteriori estimate using a gradient-based search. Our results demonstrate an enormous improvement in tracking precision and reliability, while also achieving near real-time performance. © 2009 IEEE.

Optimal design of Functionally Graded Materials using a procedural model and Particle Swarm Optimization

A new method for the optimal design of Functionally Graded Materials (FGM) is proposed in this paper. Instead of using the widely used explicit functional models, a feature tree based procedural model is proposed to represent generic material heterogeneities. A procedural model of this sort allows more than one explicit function to be incorporated to describe versatile material gradations and the material composition at a given location is no longer computed by simple evaluation of an analytic function, but obtained by execution of customizable procedures. This enables generic and diverse types of material variations to be represented, and most importantly, by a reasonably small number of design variables. The descriptive flexibility in the material heterogeneity formulation as well as the low dimensionality of the design vectors help facilitate the optimal design of functionally graded materials. Using the nature-inspired Particle Swarm Optimization (PSO) method, functionally graded materials with generic distributions can be efficiently optimized. We demonstrate, for the first time, that a PSO based optimizer outperforms classical mathematical programming based methods, such as active set and trust region algorithms, in the optimal design of functionally graded materials. The underlying reason for this performance boost is also elucidated with the help of benchmarked examples. © 2011 Elsevier Ltd. All rights reserved.

Mixed source model and its adapted vocal tract filter estimate for voice transformation and synthesis

In current methods for voice transformation and speech synthesis, the vocal tract filter is usually assumed to be excited by a flat amplitude spectrum. In this article, we present a method using a mixed source model defined as a mixture of the Liljencrants-Fant (LF) model and Gaussian noise. Using the LF model, the base approach used in this presented work is therefore close to a vocoder using exogenous input like ARX-based methods or the Glottal Spectral Separation (GSS) method. Such approaches are therefore dedicated to voice processing promising an improved naturalness compared to generic signal models. To estimate the Vocal Tract Filter (VTF), using spectral division like in GSS, we show that a glottal source model can be used with any envelope estimation method conversely to ARX approach where a least square AR solution is used. We therefore derive a VTF estimate which takes into account the amplitude spectra of both deterministic and random components of the glottal source. The proposed mixed source model is controlled by a small set of intuitive and independent parameters. The relevance of this voice production model is evaluated, through listening tests, in the context of resynthesis, HMM-based speech synthesis, breathiness modification and pitch transposition. © 2012 Elsevier B.V. All rights reserved.