Esta dissertação apresenta o desenvolvimento de um sistema em forma de armband para a captura de sinais de eletromiográficos de superfície para o reconhecimento de movimentos do braço. São apresentadas todas as suas etapas de projeto, desde a construção física, projeto de circuitos, sistema de aquisição, processamento e classificação por meio de Redes Neurais Artificiais. Foi construído um bracelete contendo oito canais para a captação do sinal de eletromiografia e um sistema auxiliar (giroscópio) de referência foi utilizado para indicar o instante em que o braço foi movimentado. Foram adquiridos dados dos grupos musculares do bíceps e do tríceps. Por meio da fusão de dados de sensores, os sinais foram processados por meio de rotinas no software LabVIEWTM. Após a extração de características do sinal, as amostras foram encaminhadas para uma Rede Neural Multi-Layer Perceptron para a classificação dos movimentos de flexão e extensão do braço. A mesma armband foi inserida na região do antebraço e os sinais de eletromiografia foram comparados com os sinais obtidos pelo dispositivo comercial MyoTM. O sistema apresentou como resultado altas taxas de classificação, acima de 95% e os sinais obtidos na região do antebraço apresentaram semelhanças com os obtidos pelo dispositivo comercial. / This master thesis presents the development of an armband system to capture of superficial electromyography signals to arm movement recognition. All project steps, since the physical building, project of the circuits, acquisition system, processing and classification by Artificial Neural Networks are presented. An armband with eight channel to capture the electromyography signal was constructed and an auxiliary system (gyroscope) was used to indicate the instant when the arm was moved. The muscle acquired groups were the biceps and triceps. By sensor data fusion, the signals were processed by LabVIEWTM routines. After the signal characteristic extraction, the samples were forwarded to a Multi-Layer Perceptron Neural Network to movement classification of arm flexion and extension. The same armband was inserted on the forearm and the electromyography signals were compared with the signals obtained by the commercial device MyoTM. The system presented, as results, high classification rates, above of 95% and the obtained signals on the region of forearm showed similarities with the obtained ones by commercial device.

CAPES / O reconhecimento facial está se tornando uma tarefa comum com a evolução da tecnologia da informação. Este artefato pode ser utilizado na área de segurança, controlando acesso a lugares restritos, identificando pessoas que tenham cometido atos ilícitos, entre outros. Executar o reconhecimento facial é uma tarefa complexa e, para completar este processo, são implementadas etapas que compreendem: a captura de imagens faciais, a detecção de regiões de interesse, a normalização facial, a extração de características e o reconhecimento em si. Dentre estas, as três primeiras são tratadas neste trabalho, que tem como objetivo principal a normalização automática de faces. Tanto para a captura de imagens quanto para a normalização frontal existem normas internacionais que padronizam o procedimento de execução destas tarefas e que foram utilizadas neste trabalho. Além disto, algumas normas foram adaptadas para a construção de uma base de imagens faciais com o objetivo de auxiliar o processo de reconhecimento facial. Também foi criada uma nova metodologia para normalização de imagens faciais laterais, baseando-se nas normas da normalização frontal. Foram implementadas normalização semiautomática frontal, semiautomática lateral e automática lateral. Para a execução da normalização facial automática são necessários dois pontos de controle, os dois olhos, o que torna indispensável a execução da etapa de detecção de regiões de interesse. Neste trabalho, foram comparadas duas metodologias semelhantes para detecção. Primeiramente foi detectada uma região contendo ambos os olhos e, em seguida, dentro desta região, foram detectados cada um dos olhos de forma mais precisa. Para as duas metodologias foram utilizadas técnicas de processamento de imagens e reconhecimento de padrões. A primeira metodologia utiliza como filtro o Haar-Like Features em conjunto com a técnica de reconhecimento de padrões Adaptative Boosting. Sendo que as técnicas equivalentes no segundo algoritmo foram o Local Binary Pattern e o Support Vector Machines, respectivamente. Na segunda metodologia também foi utilizado um algoritmo de otimização de busca baseado em vizinhança, o Variable Neighborhood Search. Os estudos resultaram em uma base com 3726 imagens, mais uma base normalizada frontal com 966 imagens e uma normalizada lateral com 276 imagens. A detecção de olhos resultou, nos melhores testes, em aproximadamente 99% de precisão para a primeira metodologia e 95% para a segunda, sendo que em todos os testes a primeira foi o mais rápida. Com o desenvolvimento de trabalhos futuros pretende-se: tornar públicas as bases de imagens, melhorar a porcentagem de acerto e velocidade de processamento para todos os testes e melhorar a normalização, implementando a normalização de plano de fundo e também de iluminação. / With the evolution of information technology Facial recognition is becoming a common task. This artifact can be used in security, controlling access to restricted places and identifying persons, for example. Facial recognition is a complex task, and it's divided into some process, comprising: facial images capture, detection of regions of interest, facial normalization, feature extraction and recognition itself. Among these, the first three are treated in this work, which has as its main objective the automatic normalization of faces. For the capture of images and for the image normalization there are international standards that standardize the procedure for implementing these tasks and which were used in this work. In addition to following these rules, other standardizations have been developed to build a database of facial images in order to assist the process of face recognition. A new methodology for normalization of profile faces, based on the rules of frontal normalization. Some ways of normalization were implemented: frontal semiautomatic, lateral semiautomatic and automatic frontal. For the execution of frontal automatic normalization we need two points of interest, the two eyes, which makes it a necessary step to execute the detection regions of interest. In this work, we compared two similar methods for detecting. Where was first detected a region containing both eyes and then, within this region were detected each eye more accurately. For the two methodologies were used techniques of image processing and pattern recognition. The first method based on the Viola and Jones algorithm, the filter uses as Haar-like Features with the technique of pattern recognition Adaptive Boosting. Where the second algorithm equivalent techniques were Local Binary Pattern and Support Vector Machines, respectively. In the second algorithm was also used an optimization algorithm based on neighborhood search, the Variable Neighborhood Search. This studies resulted in a database with 3726 images, a frontal normalized database with 966 images and a database with face's profile normalized with 276 images. The eye detection resulted in better tests, about 99 % accuracy for the first method and 95 % for the second, and in all tests the first algorithm was the fastest. With the development of future work we have: make public the images database, improve the percentage of accuracy and processing speed for all tests and improve the normalization by implementing the normalization of the background and also lighting.

Orientadores Eleri Cardozo, Diogo Coutinho Soriano / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Elétrica e de Computação / Made available in DSpace on 2018-08-24T05:15:26Z (GMT). No. of bitstreams: 1 SuarezUribe_LuisaFernanda_M.pdf: 6173860 bytes, checksum: 3a1e038c4b2f0b3d7f6c4981c268de63 (MD5) Previous issue date: 2013 / Resumo: Os potenciais evocados visuais de estado em regime permanente (SSVEP) são uma resposta cerebral medida ao capturar sinais cerebrais através de Eletroencefalograma (EEG) usando-se uma touca com eletrodos posicionados no escalpo mediante estimulação visual externa. Um sistema que detecta a resposta SSVEP gerada em um sujeito estimulado com fontes de luz (LEDs) piscando a frequências diferentes foi desenvolvido nesta dissertação com o intuito de criar uma interface homen-máquina. Para tanto, foram usados quatro estímulos visuais codificando quatro possíveis comandos, mais o estado de repouso, a serem classificados e identificados a partir da resposta observada no EEG de um sujeito e da devida análise espectral deste sinal. As características que determinam se a resposta SSVEP está presente foram estudadas através de diferentes heurísticas para a seleção de descritores (coeficientes associados a densidade espectral de potência), os quais foram posteriormente passados para um classificador linear para a determinação do comando associado ao estímulo. Para a seleção de características, o índice Davies Bouldin (DB) foi utilizado. No que concerne a análise de frequência realizado para os três sujeitos estudados foi possível observar a presença da resposta SSVEP nas frequências de estimulação, mas tal como esperado, com forte presença de ruído e com grande variabilidade entre os sujeitos. Apesar da variabilidade das características selecionadas pela estratégia adotada, obteve-se aqui resultados de classificação próximos a 90% de acerto para cada classe. Estes resultados de classificação indicaram que esta metodologia de processamento pode ser usada num sistema de BCI em tempo real, dado que o atual sistema foi projetado para operar somente com dados que foram gravados off-line / Abstract: Steady State Visual Evoked Potentials (SSVEP) is a brain response measured by capturing brain signals generated by external visual stimulation, through electroencephalogram (EEG) using a cap with electrodes placed on the scalp. A system for detection SSVEP response generated in a subject stimulated with light sources (LEDs) flashing at different frequencies has been developed in this dissertation in order to implement a human machine interface. Thus, we used four visual stimuli encoding four possible commands, besides idle state, to be classified and identified from the response observed in the EEG of a subject through the spectral analysis of this signal. The characteristics that determine if a SSVEP response is present were studied through different heuristics for selecting descriptors (coefficients associated with power spectral density), which were then passed to a linear classifier for the determination of the stimulus associated command. Davies Bouldin (DB) index was the method used for the selection of features. From the frequency analysis carried out for three subjects it was observed the presence of SSVEP response on the stimulation frequencies, but in some cases, as expected, was blurred by noise, which differ among individuals. Despite the variability of the features selected by the strategy adopted, it was obtained classification results with accuracy nearly 90\% for each class. These classification results indicated that this processing methodology could be used in a BCI system in real time, as the current system was designed to operate only with data that were recorded off-line / Mestrado / Engenharia de Computação / Mestra em Engenharia Elétrica

ARécemment, les caméras RGB-D ont été introduites sur le marché et ont permis l’exploration de nouvelles approches de reconnaissance d’actions par l’utilisation de deux modalités autres que les images RGB, à savoir, les images de profondeur et les séquences de squelette. Généralement, ces approches ont été évaluées en termes de taux de reconnaissance. Cette thèse s’intéresse principalement à la reconnaissance rapide d’actions à partir de caméras RGB-D. Le travail a été focalisé sur une amélioration conjointe de la rapidité de calcul et du taux de reconnaissance en vue d’une application temps-réel. Dans un premier temps, nous menons une étude comparative des méthodes existantes de reconnaissance d’actions basées sur des caméras RGB-D en utilisant les deux critères énoncés : le taux de reconnaissance et la rapidité de calcul. Suite aux conclusions résultant de cette étude, nous introduisons un nouveau descripteur de mouvement, à la fois précis et rapide, qui se base sur l’interpolation par splines cubiques de valeurs cinématiques du squelette, appelé Kinematic Spline Curves (KSC). De plus, afin de pallier les effets négatifs engendrés par la variabilité anthropométrique, la variation d’orientation et la variation de vitesse, des méthodes de normalisation spatiale et temporelle rapide ont été proposées. Les expérimentations menées sur quatre bases de données prouvent la précision et la rapidité de ce descripteur. Dans un second temps, un deuxième descripteur appelé Hiearchical Kinematic Coavarince(HKC) est introduit. Ce dernier est proposé dans l’optique de résoudre la question de reconnaissance rapide en ligne. Comme ce descripteur n’appartient pas à un espace euclidien, mais à l’espace des matrices Symétriques semi-Définies Positives (SsDP), nous adaptons les méthodes de classification à noyau par l’introduction d’une distance inspirée de la distance Log-Euclidienne, que nous appelons distance Log-Euclidienne modifiée. Cette extension nous permet d’utiliser des classifieurs adaptés à l’espace de caractéristiques (SPsD).Une étude expérimentale montre l’efficacité de cette méthode non seulement en termes de rapidité de calcul et de précision, mais également en termes de latence observationnelle. Ces conclusions prouvent que cette approche jointe à une méthode de segmentation d’actions pourrait s’avérer adaptée à la reconnaissance en ligne et ouvrent ainsi de nouvelles perspectives pour nos travaux futurs. / The recent availability of RGB-D cameras has renewed the interest of researchers in the topic of human action recognition. More precisely, several action recognition methods have been proposed based on the novel modalities provided by these cameras, namely, depth maps and skeleton sequences. These approaches have been mainly evaluated in terms of recognition accuracy. This thesis aims to study the issue of fast action recognition from RGB-D cameras. It focuses on proposing an action recognition method realizing a trade-off between accuracy and latency for the purpose of applying it in real-time scenarios. As a first step, we propose a comparative study of recent RGB-D based action recognition methods using the two cited criteria: accuracy of recognition and rapidity of execution. Then, oriented by the conclusions stated thanks to this comparative study, we introduce a novel, fast and accurate human action descriptor called Kinematic Spline Curves (KSC).This latter is based on the cubic spline interpolation of kinematic values. Moreover, fast spatialand temporal normalization are proposed in order to overcome anthropometric variability, orientation variation and rate variability. The experiments carried out on four different benchmarks show the effectiveness of this approach in terms of execution time and accuracy. As a second step, another descriptor is introduced, called Hierarchical Kinematic Covariance(HKC). This latter is proposed in order to solve the issue of fast online action recognition. Since this descriptor does not belong to a Euclidean space, but is an element of the space of Symmetric Positive semi-definite (SPsD) matrices, we adapt kernel classification methods by the introduction of a novel distance called Modified Log-Euclidean, which is inspiredfrom Log-Euclidean distance. This extension allows us to use suitable classifiers to the feature space SPsD of matrices. The experiments prove the efficiency of our method, not only in terms of rapidity of calculation and accuracy, but also in terms of observational latency. These conclusions show that this approach combined with an action segmentation method could be appropriate to online recognition, and consequently, opens up new prospects for future works.

This paper proposes an end-to-end, scale invariant, visual object recognition system, composed of computational components that mimic the cortex in the brain. The system uses a two stage process. The first stage is a filter that extracts scale invariant features from the visual field. The second stage uses inference based spacio-temporal analysis of these features to identify objects in the visual field. The proposed model combines Numenta's Hierarchical Temporal Memory (HTM), with HMAX developed by MIT's Brain and Cognitive Science Department. While these two biologically inspired paradigms are based on what is known about the visual cortex, HTM and HMAX tackle the overall object recognition problem from different directions. Image pyramid based methods like HMAX make explicit use of scale, but have no sense of time. HTM, on the other hand, only indirectly tackles scale, but makes explicit use of time. By combining HTM and HMAX, both scale and time are addressed. In this paper, I show that HTM and HMAX can be combined to make a com- plete cortex inspired object recognition model that explicitly uses both scale and time to recognize objects in temporal sequences of images. Additionally, through experimentation, I examine several variations of HMAX and its

Strongly inspired by an understanding of mammalian cortical structure and function, the Hierarchical Temporal Memory Cortical Learning Algorithm (HTM CLA) is a promising new approach to problems of recognition and inference in space and time. Only a subset of the theoretical framework of this algorithm has been studied, but it is already clear that there is a need for more information about the performance of HTM CLA with real data and the associated computational costs. For the work presented here, a complete implementation of Numenta's current algorithm was done in C++. In validating the implementation, first and higher order sequence learning was briefly examined, as was algorithm behavior with noisy data doing simple pattern recognition. A pattern recognition task was created using sequences of handwritten digits and performance analysis of the sequential implementation was performed. The analysis indicates that the resulting rapid increase in computing load may impact algorithm scalability, which may, in turn, be an obstacle to widespread adoption of the algorithm. Two critical hotspots in the sequential code were identified and a parallelized version was developed using OpenMP multi-threading. Scalability analysis of the parallel implementation was performed on a state of the art multi-core computing platform. Modest speedup was readily achieved with straightforward parallelization. Parallelization on multi-core systems is an attractive choice for moderate sized applications, but significantly larger ones are likely to remain infeasible without more specialized hardware acceleration accompanied by optimizations to the algorithm.

A significant number of action recognition research efforts use spatio-temporal interest point detectors for feature extraction. Although the extracted features provide useful information for recognizing actions, a significant number of them contain irrelevant motion and background clutter. In many cases, the extracted features are included as is in the classification pipeline, and sophisticated noise removal techniques are subsequently used to alleviate their effect on classification. We introduce a new action database, created from the Weizmann database, that reveals a significant weakness in systems based on popular cuboid descriptors. Experiments show that introducing complex backgrounds, stationary or dynamic, into the video causes a significant degradation in recognition performance. Moreover, this degradation cannot be fixed by fine-tuning the system or selecting better interest points. Instead, we show that the problem lies at the descriptor level and must be addressed by modifying descriptors.

This dissertation explores the intersection of information retrieval and universal source coding techniques and studies an optimal multidimensional source representation from an information theoretic point of view. Previous research on information retrieval particularly focus on learning probabilistic or deterministic source models based on primarily two different types of source representations, e.g., fixed-shape partitions or uniform regions. We study the limitations of the conventional source representations on capturing the semantics of the given multidimensional source sequences and propose a new type of primitive source representation generated by a universal source coding technique. We propose a multidimensional incremental parsing algorithm extended from the Lempel-Ziv incremental parsing and its three component schemes for multidimensional source coding. The properties of the proposed coding algorithm are exploited under two-dimensional lossless and lossy source coding. By the proposed coding algorithm, a given multidimensional source sequence is parsed into a number of variable-size patches. We call this methodology a parsed representation. Based on the source representation, we propose an information retrieval framework that analyzes a set of source sequences under a linguistic processing technique and implemented content-based image retrieval systems. We examine the relevance of the proposed source representation by comparing it with the conventional representation of visual information. To further extend the proposed framework, we apply a probabilistic linguistic processing technique to modeling the latent aspects of a set of documents. In addition, beyond the symbol-wise pattern matching paradigm employed in the source coding and the image retrieval systems, we devise a robust pattern matching that compares the first- and second-order statistics of source patches. Qualitative and quantitative analysis of the proposed framework justifies the superiority of the proposed information retrieval framework based on the parsed representation. The proposed source representation technique and the information retrieval frameworks encourage future work in exploiting a systematic way of understanding multidimensional sources that parallels a linguistic structure.

This study is an attempt to simulate aspects of human visual perception by automating the detection of specific types of objects in digital images. The success of the methods attempted here was measured by how well results of experiments corresponded to what a typical human’s assessment of the data might be. The subject of the study was images of live fish taken underwater by digital video or digital still cameras. It is desirable to be able to automate the processing of such data for efficient stock assessment for fisheries management. In this study some well known statistical pattern classification techniques were tested and new syntactical/ structural pattern recognition techniques were developed. For testing of statistical pattern classification, the pixels belonging to fish were separated from the background pixels and the EM algorithm for Gaussian mixture models was used to locate clusters of pixels. The means and the covariance matrices for the components of the model were used to indicate the location, size and shape of the clusters. Because the number of components in the mixture is unknown, the EM algorithm has to be run a number of times with different numbers of components and then the best model chosen using a model selection criterion. The AIC (Akaike Information Criterion) and the MDL (Minimum Description Length) were tested.The MDL was found to estimate the numbers of clusters of pixels more accurately than the AIC, which tended to overestimate cluster numbers. In order to reduce problems caused by initialisation of the EM algorithm (i.e. starting positions of mixtures and number of mixtures), the Dynamic Cluster Finding algorithm (DCF) was developed (based on the Dog-Rabbit strategy). This algorithm can produce an estimate of the locations and numbers of clusters of pixels. The Dog-Rabbit strategy is based on early studies of learning behaviour in neurons. The main difference between Dog-Rabbit and DCF is that DCF is based on a toroidal topology which removes the tendency of cluster locators to migrate to the centre of mass of the data set and miss clusters near the edges of the image. In the second approach to the problem, data was extracted from the image using an edge detector. The edges from a reference object were compared with the edges from a new image to determine if the object occurred in the new image. In order to compare edges, the edge pixels were first assembled into curves using an UpWrite procedure; then the curves were smoothed by fitting parametric cubic polynomials. Finally the curves were converted to arrays of numbers which represented the signed curvature of the curves at regular intervals. Sets of curves from different images can be compared by comparing the arrays of signed curvature values, as well as the relative orientations and locations of the curves. Discrepancy values were calculated to indicate how well curves and sets of curves matched the reference object. The total length of all matched curves was used to indicate what fraction of the reference object was found in the new image. The curve matching procedure gave results which corresponded well with what a human being being might observe.

Thesis (M.S.)--University of Missouri-Columbia, 2009. / The entire dissertation/thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file (which also appears in the research.pdf); a non-technical general description, or public abstract, appears in the public.pdf file. Thesis advisor: Yang Gong. "May 2009" Includes bibliographical references.