No description available.

Signalų technologijų magistro darbo tema yra aktuali, nes nebepakanka įprastų informacijos įvedimo priemonių. Todėl ieškoti ir apdoroti informaciją, valdyti sudėtingus įrenginius ir programas daug patogiau būtų jei kompiuteriai ir įvairūs įrenginiai suprastų žmogaus kalbą. Pasaulyje panašios sistemos kuriamos jau daugelį metų. Tačiau šiuo metu lietuvių kalbos atpažinimo sistemos yra dar tik kūrimo stadijoje. Darbe nagrinėjamas Lietuvių kalbos žodžių atpažinimas skirstant juos į kategorijas ir naudojant paslėptuosius markovo modelius. Šio tyrimo tikslas – ištirti lietuvių kalbos žodžių skirstymo į kategorijas įtaką atpažinimo tikslumui.Taip pat tiriamas žodžių grupių bei pavienių žodžių atpažinimas. Akustinis modelis sukurtas su HTK paketu, kuris naudojasi paslėptųjų Markovo modelių metodika. Žodžių skirstymas į kategorijas aprašytas Backus-Naur formatu. Eksperimentai bus atliekami ir rezultatai gaunami naudojant, Julius programinės įrangos įrankius bei šio paketo, žodžių kategorijų pagrindu veikiančią, Julian kalbos atpažinimo sistemą. Geriausi rezultatai gauti bandant atpažinti pavienius žodžius suskirstytus į kategorijas. Atpažinimo tikslumas siekia 91 %. Bandant atpažinti žodžių sekas, nesuskirstytas į kategorijas, gautas atpažinimo tikslumas tesiekia 51 %. Microsoft Office Word 2003 meniu valdymo atpažinimo tikslumas siekia 82 %. / The theme of Master project of signal technology is actual, because not enough usual information introduction ways. Therefore information search and processing, complicated devices and programs control would be more handily if computers and devices understood human speech. Similar systems are designing for many years in the world. However Lithuanian speech recognition systems are still developing in nowadays. The thesis treats of isolated Lithuanian words recognition dividing them into category and using Hidden Markov Models. The idea of research is to explore categorization of Lithuanian words influence on the accuracy of recognition. The recognition of single words and word groups is under research too. Acoustic model is constructed by using HTK toolkit which is based on Hidden Markov Models. Categorization of words is described with Backus-Naur form. Experiments are made with Julius software speech recognition system Julian witch performs words category based recognition. Best results are got trying to recognize single words set into categories. The accuracy rate of recognition reaches 91 %. While trying to recognize uncategorized word sequences – the accuracy rate of recognition reaches only 51 %. The accuracy rate of Microsoft Office Word 2003 control menu recognition reaches 82 %.

This thesis, which is organized in two independent parts, presents work on distributional semantics and on variable selection. In the first part, we introduce a new method for learning good word representations using large quantities of unlabeled sentences. The method is based on a probabilistic model of sentence, using a hidden Markov model and a syntactic dependency tree. The latent variables, which correspond to the nodes of the dependency tree, aim at capturing the meanings of the words. We develop an efficient algorithm to perform inference and learning in those models, based on online EM and approximate message passing. We then evaluate our models on intrinsic tasks such as predicting human similarity judgements or word categorization, and on two extrinsic tasks: named entity recognition and supersense tagging. In the second part, we introduce, in the context of linear models, a new penalty function to perform variable selection in the case of highly correlated predictors. This penalty, called the trace Lasso, uses the trace norm of the selected predictors, which is a convex surrogate of their rank, as the criterion of model complexity. The trace Lasso interpolates between the $\ell_1$-norm and $\ell_2$-norm. In particular, it is equal to the $\ell_1$-norm if all predictors are orthogonal and to the $\ell_2$-norm if all predictors are equal. We propose two algorithms to compute the solution of least-squares regression regularized by the trace Lasso, and perform experiments on synthetic datasets to illustrate the behavior of the trace Lasso.

Reducing interference noise in a noisy speech recording has been a challenging task for many years yet has a variety of applications, for example, in handsfree mobile communications, in speech recognition, and in hearing aids. Traditional single-channel noise reduction schemes, such as Wiener filtering, do not work satisfactorily in the presence of non-stationary background noise. Alternatively, supervised approaches, where the noise type is known in advance, lead to higher-quality enhanced speech signals. This dissertation proposes supervised and unsupervised single-channel noise reduction algorithms. We consider two classes of methods for this purpose: approaches based on nonnegative matrix factorization (NMF) and methods based on hidden Markov models (HMM).  The contributions of this dissertation can be divided into three main (overlapping) parts. First, we propose NMF-based enhancement approaches that use temporal dependencies of the speech signals. In a standard NMF, the important temporal correlations between consecutive short-time frames are ignored. We propose both continuous and discrete state-space nonnegative dynamical models. These approaches are used to describe the dynamics of the NMF coefficients or activations. We derive optimal minimum mean squared error (MMSE) or linear MMSE estimates of the speech signal using the probabilistic formulations of NMF. Our experiments show that using temporal dynamics in the NMF-based denoising systems improves the performance greatly. Additionally, this dissertation proposes an approach to learn the noise basis matrix online from the noisy observations. This relaxes the assumption of an a-priori specified noise type and enables us to use the NMF-based denoising method in an unsupervised manner. Our experiments show that the proposed approach with online noise basis learning considerably outperforms state-of-the-art methods in different noise conditions.  Second, this thesis proposes two methods for NMF-based separation of sources with similar dictionaries. We suggest a nonnegative HMM (NHMM) for babble noise that is derived from a speech HMM. In this approach, speech and babble signals share the same basis vectors, whereas the activation of the basis vectors are different for the two signals over time. We derive an MMSE estimator for the clean speech signal using the proposed NHMM. The objective evaluations and performed subjective listening test show that the proposed babble model and the final noise reduction algorithm outperform the conventional methods noticeably. Moreover, the dissertation proposes another solution to separate a desired source from a mixture with arbitrarily low artifacts.  Third, an HMM-based algorithm to enhance the speech spectra using super-Gaussian priors is proposed. Our experiments show that speech discrete Fourier transform (DFT) coefficients have super-Gaussian rather than Gaussian distributions even if we limit the speech data to come from a specific phoneme. We derive a new MMSE estimator for the speech spectra that uses super-Gaussian priors. The results of our evaluations using the developed noise reduction algorithm support the super-Gaussianity hypothesis. / <p>QC 20130916</p>

In this study we look at the problem of analysing survival data in the presence of longitudinally collected covariates. New methodology for analysing such data has been developed through the use of hidden Markov modeling. Special attention has been given to the case of large information volume, where a preliminary data reduction is necessary. Novel graphical diagnostics have been proposed to assess goodness of fit and significance of covariates. The methodology developed has been applied to the data collected on behaviors of Mexican fruit flies, which were monitored throughout their lives. It has been found that certain patterns in eating behavior may serve as an aging marker. In particular it has been established that the frequency of eating is positively correlated with survival times. / Thesis (Ph.D, Mathematics & Statistics) -- Queen's University, 2010-04-26 18:34:01.131

Stochastic control problems that arise in sequential decision making applications typically assume that information used for decision-making is obtained according to a predetermined sampling schedule. In many real applications however, there is a high sampling cost associated with collecting such data. It is therefore of equal importance to determine when information should be collected as it is to decide how this information should be utilized for optimal decision-making. This type of joint optimization has been a long-standing problem in the operations research literature, and very few results regarding the structure of the optimal sampling and control policy have been published. In this thesis, the joint optimization of sampling and control is studied in the context of maintenance optimization. New theoretical results characterizing the structure of the optimal policy are established, which have practical interpretation and give new insight into the value of condition-based maintenance programs in life-cycle asset management. Applications in other areas such as healthcare decision-making and statistical process control are discussed. Statistical parameter estimation results are also developed with illustrative real-world numerical examples.

Stochastic control problems that arise in sequential decision making applications typically assume that information used for decision-making is obtained according to a predetermined sampling schedule. In many real applications however, there is a high sampling cost associated with collecting such data. It is therefore of equal importance to determine when information should be collected as it is to decide how this information should be utilized for optimal decision-making. This type of joint optimization has been a long-standing problem in the operations research literature, and very few results regarding the structure of the optimal sampling and control policy have been published. In this thesis, the joint optimization of sampling and control is studied in the context of maintenance optimization. New theoretical results characterizing the structure of the optimal policy are established, which have practical interpretation and give new insight into the value of condition-based maintenance programs in life-cycle asset management. Applications in other areas such as healthcare decision-making and statistical process control are discussed. Statistical parameter estimation results are also developed with illustrative real-world numerical examples.

Tracking and understanding human gait is an important step towards improving elderly mobility and safety. This thesis presents a vision-based tracking system that estimates the 3D pose of a wheeled walker user's lower limbs with cameras mounted on the moving walker. The tracker estimates 3D poses from images of the lower limbs in the coronal plane in a dynamic, uncontrolled environment. It employs a probabilistic approach based on particle filtering with three different camera setups: a monocular RGB camera, binocular RGB cameras, and a depth camera. For the RGB cameras, observation likelihoods are designed to compare the colors and gradients of each frame with initial templates that are manually extracted. Two strategies are also investigated for handling appearance change of tracking target: increasing number of templates and using different representations of colors. For the depth camera, two observation likelihoods are developed: the first one works directly in the 3D space, while the second one works in the projected image space. Experiments are conducted to evaluate the performance of the tracking system with different users for all three camera setups. It is demonstrated that the trackers with the RGB cameras produce results with higher error as compared to the depth camera, and the strategies for handling appearance change improve tracking accuracy in general. On the other hand, the tracker with the depth sensor successfully tracks the 3D poses of users over the entire video sequence and is robust against unfavorable conditions such as partial occlusion, missing observations, and deformable tracking target.

In this thesis two different problems in distributed sensor networks are considered. Part I involves optimal quantiser design for decentralised estimation of a two-state hidden Markov model with dual sensors. The notion of optimality for quantiser design is based on minimising the probability of error in estimating the hidden Markov state. Equations for the filter error are derived for the continuous (unquantised) sensor outputs (signals), which are used to benchmark the performance of the quantisers. Minimising the probability of filter error to obtain the quantiser breakpoints is a difficult problem therefore an alternative method is employed. The quantiser breakpoints are obtained by maximising the mutual information between the quantised signals and the hidden Markov state. This method is known to work well for the single sensor case. Cases with independent and correlated noise across the signals are considered. The method is then applied to Markov processes with Gaussian signal noise, and further investigated through simulation studies. Simulations involving both independent and correlated noise across the sensors are performed and a number of interesting new theoretical results are obtained, particularly in the case of correlated noise. In Part II, the focus shifts to the detection of faults in helicopter transmission systems. The aim of the investigation is to determine whether the acoustic signature can be used for fault detection and diagnosis. To investigate this, statistical change detection algorithms are applied to acoustic vibration data obtained from the main rotor gearbox of a Bell 206 helicopter, which is run at high load under test conditions.

Les séries temporelles d'images satellitaires sont une source d'information importante pour le suivi des changements spatio-temporels des surfaces terrestres. En outre, le nombre d’images est en augmentation constante. Pour les exploiter pleinement, des outils dédiés au traitement automatique du contenu informationnel sont développés. Néanmoins ces techniques ne satisfont pas complètement les géographes qui exploitent pourtant, de plus en plus couramment, les données extraites des images dans leurs études afin de prédire le futur. Nous proposons dans cette thèse, une méthodologie générique à base d’un modèle de Markov caché pour l’analyse et la prédiction des changements sur une séquence d’images satellitaires. Cette méthodologie présente deux modules : un module de traitement intégrant les descripteurs et les algorithmes classiquement utilisés en interprétation d'images, et un module d’apprentissage basé sur les modèles de Markov cachés. La performance de notre approche est évaluée par des essais d’interprétations des évènements spatio-temporels effectués sur plusieurs sites d’études. Les résultats obtenus permettront d’analyser et de prédire les changements issus des différentes séries temporelles d’images SPOT et LANDSAT pour l’observation des évènements spatio-temporels telle que l'expansion urbaine et la déforestation. / The time series of satellite images are an important source of information for monitoring spatiotemporal changes of land surfaces. Furthermore, the number of satellite images is increasing constantly, for taking full advantage, tools dedicated to the automatic processing of information content is developed. However these techniques do not completely satisfy the geographers who exploit more currently, the data extracted from the images in their studies to predict the future. In this research we propose a generic methodology based on a hidden Markov model for analyzing and predicting changes in a sequence of satellite images. The methodology that is proposed presents two modules : a processing module which incorporating descriptors and algorithms conventionally used in image interpretation and a learning module based on hidden Markov models. The performance of the approach is evaluated by trials of interpretation of spatiotemporal events conducted in several study sites. Results obtained allow us to analyze and to predict changes from various time series of SPOT and LANDSAT images for observation of spatiotemporal events such as urban development and deforestation.