Assessment of Severe Apnoea through Voice Analysis, Automatic Speech, and Speaker Recognition Techniques

This study is part of an ongoing collaborative effort between the medical and the signal processing communities to promote research on applying standard Automatic Speech Recognition (ASR) techniques for the automatic diagnosis of patients with severe obstructive sleep apnoea (OSA). Early detection of severe apnoea cases is important so that patients can receive early treatment. Effective ASR-based detection could dramatically cut medical testing time. Working with a carefully designed speech database of healthy and apnoea subjects, we describe an acoustic search for distinctive apnoea voice characteristics. We also study abnormal nasalization in OSA patients by modelling vowels in nasal and nonnasal phonetic contexts using Gaussian Mixture Model (GMM) pattern recognition on speech spectra. Finally, we present experimental findings regarding the discriminative power of GMMs applied to severe apnoea detection. We have achieved an 81% correct classification rate, which is very promising and underpins the interest in this line of inquiry.

Unsupervised segmentation using Gabor wavelets and statistical features in LIDAR data analysis

In this paper, we address issues in segmentation Of remotely sensed LIDAR (LIght Detection And Ranging) data. The LIDAR data, which were captured by airborne laser scanner, contain 2.5 dimensional (2.5D) terrain surface height information, e.g. houses, vegetation, flat field, river, basin, etc. Our aim in this paper is to segment ground (flat field)from non-ground (houses and high vegetation) in hilly urban areas. By projecting the 2.5D data onto a surface, we obtain a texture map as a grey-level image. Based on the image, Gabor wavelet filters are applied to generate Gabor wavelet features. These features are then grouped into various windows. Among these windows, a combination of their first and second order of statistics is used as a measure to determine the surface properties. The test results have shown that ground areas can successfully be segmented from LIDAR data. Most buildings and high vegetation can be detected. In addition, Gabor wavelet transform can partially remove hill or slope effects in the original data by tuning Gabor parameters.

A statistical approach to the problem of restoring damaged and contaminated images

We address the problem of automatically identifying and restoring damaged and contaminated images. We suggest a novel approach based on a semi-parametric model. This has two components, a parametric component describing known physical characteristics and a more flexible non-parametric component. The latter avoids the need for a detailed model for the sensor, which is often costly to produce and lacking in robustness. We assess our approach using an analysis of electroencephalographic images contaminated by eye-blink artefacts and highly damaged photographs contaminated by non-uniform lighting. These experiments show that our approach provides an effective solution to problems of this type.

Anaerobic threshold estimation by statistical modelling.

Anaerobic threshold (AT) is usually estimated as a change point problem by visual analysis of the cardiorespiratory response to incremental dynamic exercise. In this study, two phase linear (TPL) models of the linear-linear and linear-quadratic type were used for the estimation of AT. The correlation coefficient between the classical and statistical approaches was 0.88, and 0.89 after outlier exclusion. The TPL models provide a simple method for estimating AT that can be easily implemented using a digital computer for the automatic pattern recognition of AT.

Improving face recognition with multispectral fusion and support vector machines

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Quantitative Chemical Profile and Multivariate Statistical Analysis of Alembic Distilled Sugarcane Spirit Fractions

Concentrations of 39 organic compounds were determined in three fractions (head, heart and tail) obtained from the pot still distillation of fermented sugarcane juice. The results were evaluated using analysis of variance (ANOVA), Tukey's test, principal component analysis (PCA), hierarchical cluster analysis (HCA) and linear discriminant analysis (LDA). According to PCA and HCA, the experimental data lead to the formation of three clusters. The head fractions give rise to a more defined group. The heart and tail fractions showed some overlap consistent with its acid composition. The predictive ability of calibration and validation of the model generated by LDA for the three fractions classification were 90.5 and 100%, respectively. This model recognized as the heart twelve of the thirteen commercial cachacas (92.3%) with good sensory characteristics, thus showing potential for guiding the process of cuts.

Multi-q pattern analysis: A case study in image classification

This paper compares the effectiveness of the Tsallis entropy over the classic Boltzmann-Gibbs-Shannon entropy for general pattern recognition, and proposes a multi-q approach to improve pattern analysis using entropy. A series of experiments were carried out for the problem of classifying image patterns. Given a dataset of 40 pattern classes, the goal of our image case study is to assess how well the different entropies can be used to determine the class of a newly given image sample. Our experiments show that the Tsallis entropy using the proposed multi-q approach has great advantages over the Boltzmann-Gibbs-Shannon entropy for pattern classification, boosting image recognition rates by a factor of 3. We discuss the reasons behind this success, shedding light on the usefulness of the Tsallis entropy and the multi-q approach. (C) 2012 Elsevier B.V. All rights reserved.

Deep learning for computer vision: a comparison between convolutional neural networks and hierarchical temporal memories on object recognition tasks

In recent years, Deep Learning techniques have shown to perform well on a large variety of problems both in Computer Vision and Natural Language Processing, reaching and often surpassing the state of the art on many tasks. The rise of deep learning is also revolutionizing the entire field of Machine Learning and Pattern Recognition pushing forward the concepts of automatic feature extraction and unsupervised learning in general. However, despite the strong success both in science and business, deep learning has its own limitations. It is often questioned if such techniques are only some kind of brute-force statistical approaches and if they can only work in the context of High Performance Computing with tons of data. Another important question is whether they are really biologically inspired, as claimed in certain cases, and if they can scale well in terms of "intelligence". The dissertation is focused on trying to answer these key questions in the context of Computer Vision and, in particular, Object Recognition, a task that has been heavily revolutionized by recent advances in the field. Practically speaking, these answers are based on an exhaustive comparison between two, very different, deep learning techniques on the aforementioned task: Convolutional Neural Network (CNN) and Hierarchical Temporal memory (HTM). They stand for two different approaches and points of view within the big hat of deep learning and are the best choices to understand and point out strengths and weaknesses of each of them. CNN is considered one of the most classic and powerful supervised methods used today in machine learning and pattern recognition, especially in object recognition. CNNs are well received and accepted by the scientific community and are already deployed in large corporation like Google and Facebook for solving face recognition and image auto-tagging problems. HTM, on the other hand, is known as a new emerging paradigm and a new meanly-unsupervised method, that is more biologically inspired. It tries to gain more insights from the computational neuroscience community in order to incorporate concepts like time, context and attention during the learning process which are typical of the human brain. In the end, the thesis is supposed to prove that in certain cases, with a lower quantity of data, HTM can outperform CNN.

Prion crystalization model and its application to recognition pattern

This paper introduces APA (?Artificial Prion Assembly?): a pattern recognition system based on artificial prion crystalization. Specifically, the system exhibits the capability to classify patterns according to the resulting prion self- assembly simulated with cellular automata. Our approach is inspired in the biological process of proteins aggregation, known as prions, which are assembled as amyloid fibers related with neurodegenerative disorders.

A Statistical Convergence Aplication for the Hopfield Networks

When Recurrent Neural Networks (RNN) are going to be used as Pattern Recognition systems, the problem to be considered is how to impose prescribed prototype vectors ξ^1,ξ^2,...,ξ^p as fixed points. The synaptic matrix W should be interpreted as a sort of sign correlation matrix of the prototypes, In the classical approach. The weak point in this approach, comes from the fact that it does not have the appropriate tools to deal efficiently with the correlation between the state vectors and the prototype vectors The capacity of the net is very poor because one can only know if one given vector is adequately correlated with the prototypes or not and we are not able to know what its exact correlation degree. The interest of our approach lies precisely in the fact that it provides these tools. In this paper, a geometrical vision of the dynamic of states is explained. A fixed point is viewed as a point in the Euclidean plane R2. The retrieving procedure is analyzed trough statistical frequency distribution of the prototypes. The capacity of the net is improved and the spurious states are reduced. In order to clarify and corroborate the theoretical results, together with the formal theory, an application is presented

Continuous statistical modelling for rapid detection of adulteration of extra virgin olive oil using mid infrared and Raman spectroscopic data

The main objective of this work was to develop a novel dimensionality reduction technique as a part of an integrated pattern recognition solution capable of identifying adulterants such as hazelnut oil in extra virgin olive oil at low percentages based on spectroscopic chemical fingerprints. A novel Continuous Locality Preserving Projections (CLPP) technique is proposed which allows the modelling of the continuous nature of the produced in-house admixtures as data series instead of discrete points. The maintenance of the continuous structure of the data manifold enables the better visualisation of this examined classification problem and facilitates the more accurate utilisation of the manifold for detecting the adulterants. The performance of the proposed technique is validated with two different spectroscopic techniques (Raman and Fourier transform infrared, FT-IR). In all cases studied, CLPP accompanied by k-Nearest Neighbors (kNN) algorithm was found to outperform any other state-of-the-art pattern recognition techniques.

Improvements on ICA mixture models for image pre-processing and segmentation

Today several different unsupervised classification algorithms are commonly used to cluster similar patterns in a data set based only on its statistical properties. Specially in image data applications, self-organizing methods for unsupervised classification have been successfully applied for clustering pixels or group of pixels in order to perform segmentation tasks. The first important contribution of this paper refers to the development of a self-organizing method for data classification, named Enhanced Independent Component Analysis Mixture Model (EICAMM), which was built by proposing some modifications in the Independent Component Analysis Mixture Model (ICAMM). Such improvements were proposed by considering some of the model limitations as well as by analyzing how it should be improved in order to become more efficient. Moreover, a pre-processing methodology was also proposed, which is based on combining the Sparse Code Shrinkage (SCS) for image denoising and the Sobel edge detector. In the experiments of this work, the EICAMM and other self-organizing models were applied for segmenting images in their original and pre-processed versions. A comparative analysis showed satisfactory and competitive image segmentation results obtained by the proposals presented herein. (C) 2008 Published by Elsevier B.V.

Recognition of noisy images by PLL networks

We use networks composed of three phase-locked loops (PLLs), where one of them is the master, for recognizing noisy images. The values of the coupling weights among the PLLs control the noise level which does not affect the successful identification of the input image. Analytical results and numerical tests are presented concerning the scheme performance. (c) 2008 Elsevier B.V. All rights reserved.

A chemometric study on the analgesic activity of cannabinoid compounds using SDA, KNN and SIMCA methods

The supervised pattern recognition methods K-Nearest Neighbors (KNN), stepwise discriminant analysis (SDA), and soft independent modelling of class analogy (SIMCA) were employed in this work with the aim to investigate the relationship between the molecular structure of 27 cannabinoid compounds and their analgesic activity. Previous analyses using two unsupervised pattern recognition methods (PCA-principal component analysis and HCA-hierarchical cluster analysis) were performed and five descriptors were selected as the most relevants for the analgesic activity of the compounds studied: R (3) (charge density on substituent at position C(3)), Q (1) (charge on atom C(1)), A (surface area), log P (logarithm of the partition coefficient) and MR (molecular refractivity). The supervised pattern recognition methods (SDA, KNN, and SIMCA) were employed in order to construct a reliable model that can be able to predict the analgesic activity of new cannabinoid compounds and to validate our previous study. The results obtained using the SDA, KNN, and SIMCA methods agree perfectly with our previous model. Comparing the SDA, KNN, and SIMCA results with the PCA and HCA ones we could notice that all multivariate statistical methods classified the cannabinoid compounds studied in three groups exactly in the same way: active, moderately active, and inactive.