A avaliação perceptivo-auditiva tem papel fundamental no estudo e na avaliação da voz, no entanto, por ser subjetiva está sujeita a imprecisões e variações. Por outro lado, a análise acústica permite a reprodutibilidade de resultados, porém precisa ser aprimorada, pois não analisa com precisão vozes com disfonias mais intensas e com ondas caóticas. Assim, elaborar medidas que proporcionem conhecimentos confiáveis em relação à função vocal resulta de uma necessidade antiga dentro desta linha de pesquisa e atuação clínica. Neste contexto, o uso da inteligência artificial, como as redes neurais artificiais, indica ser uma abordagem promissora. Objetivo: Validar um sistema automático utilizando redes neurais artificiais para a avaliação de vozes rugosas e soprosas. Materiais e métodos: Foram selecionadas 150 vozes, desde neutras até com presença em grau intenso de rugosidade e/ou soprosidade, do banco de dados da Clínica de Fonoaudiologia da Faculdade de Odontologia de Bauru (FOB/USP). Dessas vozes, 23 foram excluídas por não responderem aos critérios de inclusão na amostra, assim utilizaram-se 123 vozes. Procedimentos: avaliação perceptivo-auditiva pela escala visual analógica de 100 mm e pela escala numérica de quatro pontos; extração de características do sinal de voz por meio da Transformada Wavelet Packet e dos parâmetros acústicos: jitter, shimmer, amplitude da derivada e amplitude do pitch; e validação do classificador por meio da parametrização, treino, teste e avaliação das redes neurais artificiais. Resultados: Na avaliação perceptivo-auditiva encontrou-se, por meio do teste Coeficiente de Correlação Intraclasse (CCI), concordâncias inter e intrajuiz excelentes, com p = 0,85 na concordância interjuízes e p variando de 0,87 a 0,93 nas concordâncias intrajuiz. Em relação ao desempenho da rede neural artificial, na discriminação da soprosidade e da rugosidade e dos seus respectivos graus, encontrou-se o melhor desempenho para a soprosidade no subconjunto composto pelo jitter, amplitude do pitch e frequência fundamental, no qual obteve-se taxa de acerto de 74%, concordância excelente com a avaliação perceptivo-auditiva da escala visual analógica (0,80 no CCI) e erro médio de 9 mm. Para a rugosidade, o melhor subconjunto foi composto pela Transformada Wavelet Packet com 1 nível de decomposição, jitter, shimmer, amplitude do pitch e frequência fundamental, no qual obteve-se 73% de acerto, concordância excelente (0,84 no CCI), e erro médio de 10 mm. Conclusão: O uso da inteligência artificial baseado em redes neurais artificiais na identificação, e graduação da rugosidade e da soprosidade, apresentou confiabilidade excelente (CCI > 0,80), com resultados semelhantes a concordância interjuízes. Dessa forma, a rede neural artificial revela-se como uma metodologia promissora de avaliação vocal, tendo sua maior vantagem a objetividade na avaliação. / The auditory-perceptual evaluation is fundamental in the study and analysis of voice. This evaluation, however, is subjective and tends to be imprecise and variable. On the other hand, acoustic analysis allows reproducing results, although these results must be refined since the analysis is not precise enough for intense dysphonia or chaotic waves. Therefore, the will to develop measurements allowing reliable knowledge related to vocal function is not new on this research and clinical actuation field. In this context, the use of artificial intelligence such as neural networks seems to be a promising research field. Objective: to validate an automatic system using artificial neural networks for evaluation of vocal roughness and breathiness. Methods: One hundred fifty (150) voices were selected from from Clínica de Fonoaudiologia da Faculdade de Odontologia de Bauru (FOB/USP) database. These voices presented variation from neutral to intense roughness and/or breathiness. Twenty-three of them were excluded since they did not match inclusion criteria. Thus, 123 voices were used for analysis. The procedures include use of auditoryperception based on two scales: visual analog scale of 100 mm and four points numerical scale. Additionally, the characteristics of voice signals were extracted by Wavelet Packet Transform and by analysis of acoustic parameters: jitter, shimmer, derivative amplitude and pitch amplitude. Validation of classifying system was carried out by parameterization, training, test and evaluation of artificial neural networks. Results: In the auditory-perceptual evaluation, excellent interrater (p=0.85) and intrarater (0.87<p<0.93) agreement were obtained by means of Intraclass Correlation Coefficient (ICC) testing. The artificial neural network performance has achieved the best results for breathiness in the subset composed by parameters jitter, pitch amplitude and fundamental frequency. In this case, the neural network obtained a rate of 74%, demonstrating excellent concordance with auditory-perceptual evaluation for visual analog scale (0.80 ICC) and mean error of 9 mm. As for roughness evaluation, the best subset is composed by Wavelet Packet Transform with 1 resolution level, jitter, shimmer, pitch amplitude and fundamental frequency. For this case, a 73% rate was achieved (0.84 ICC) and mean error of 10 mm was obtained. Conclusion: The use of artificial neural networks for roughness and breathiness evaluation present high reliability (ICC&gt;0.80), with results similar to interrater agreement. Thus, the artificial neural network reveals a promising method for vocal evaluation, bringing objective analysis as a strong advantage.

Indoor climate control is responsible for a substantial amount of the world's total energy expenditure. In a time of climate crisis where a reduction of energy consumption is crucial to avoid climate disaster, indoor climate control is a ripe target for eliminating energy waste. The conventional method of adjusting the indoor climate with the use of setpoint curves, based solely on outdoor temperature, may lead to notable inefficiencies. This project evaluates the possibility to replace this method of regulation with a system based on model predictive control (MPC) in one of Uppsala University Hospitals office buildings. A prototype of an MPC controller using Artificial Neural Networks (ANN) as its system model was developed. The system takes several data sources into account, including indoor and outdoor temperatures, radiator flowline and return temperatures, current solar radiation as well as forecast for both solar radiation and outdoor temperature. The system was not set in production but the controller's predicted values correspond well to the buildings current thermal behaviour and weather data. These theoretical results attest to the viability of using the method to regulate the indoor climate in buildings in place of setpoint curves. / Bibehållande av inomhusklimat står för en avsevärd del av världens totala energikonsumtion. Med dagens klimatförändringar där minskad energikonsumtion är viktigt för den hållbara utvecklingen så är inomhusklimat ett lämpligt mål för att förhindra slösad energi. Konventionell styrning av inomhusklimat använder sig av börvärdeskurvor, baserade enbart på utomhustemperatur, vilket kan leda till anmärkningsvärt energispill. Detta projekt utvärderar möjligheten att ersätta denna styrmetod med ett system baserat på model predictive control (MPC) och använda detta i en av Akademiska sjukhusets lokaler i Uppsala. En MPC styrenhet som använder Artificiella Neurala Nätverk (ANN) som sin modell utvecklades. Systemet använder sig av flera datakällor däribland inomhus- och utomhustemperatur, radiatorslingornas framlednings- och returtemperatur, rådande solinstrålning såväl som prognoser för solinstrålning och utomhustemperatur. Systemet sattes inte i produktion men dess prognos stämmer väl överens med tillgänglig väderdata och husets termiska beteende. De presenterade resultaten påvisar metoden vara ett lämpligt substitut för styrning med börvärdeskurvor.

Este trabalho consiste em apresentar o desenvolvimento de um Smart meter, integrado a um analisador de qualidade de energia, para análise de distorções harmônicas, utilizando método de redes neurais artificiais embarcado em hardware. Tal Smart meter está incluído dentro dos conceitos de Smart Grid, que serão apresentados também neste trabalho. O intuito do desenvolvimento do Smart meter para análise de distorções harmônicas é auxiliar concessionárias de energia elétrica a identificar que tipo de carga o consumidor utiliza em sua residência, a fim de contribuir para a tomada de decisões apropriadas, tais como a diminuição da emissão de correntes harmônicas, demanda de energia, detecção de falhas no fornecimento de energia elétrica e faturas diferenciadas de acordo com a quantidade de harmônicas injetadas na rede elétrica. Adicionalmente, observou-se que o Smart meter desenvolvido pode ser ainda utilizado para detectar fenômenos de VTCD, como elevação, afundamento e interrupção de energia. Todo o processo de desenvolvimento do Smart meter é apresentado no decorrer desta tese de doutorado. / This thesis consists to present the development of a Smart Meter integrated to power quality analyzer for the analysis of harmonic distortion, using methods based on artificial neural networks in embedded hardware. This Smart Meter is included within the concepts of Smart Grid, which will be also presented in this work. The intention of the development of the Smart Meter for analysis of harmonic distortion is to assist utilities companies to identify what loads type the consumer uses at your residence in order to contribute for supporting decisions, such as reducing the emission of the harmonic currents, power demand and faults detection in electric energy supply and distinct bills according to the amount of harmonics injected into the power grid. In addition, it was observed that this developed Smart Meter can be even used to detect the VTCD phenomena, such as swell, sag and interruption of the energy supply. All development steps of this Smart Meter is presented in this doctoral thesis.

<div>U-net is an image segmentation technique developed primarily for medical image analysis that can precisely segment images using a scarce amount of training data. These traits provide U-net with a high utility within the medical imaging community and have resulted in extensive adoption of U-net as the primary tool for segmentation tasks in medical imaging. The success of U-net is evident in its widespread use in nearly all major image modalities from CT scans and MRI to X-rays and microscopy. Furthermore, while U-net is largely a segmentation tool, there have been instances of the use of U-net in other applications. Given that U-net's potential is still increasing, this review examines the numerous developments and breakthroughs in the U-net architecture and provides observations on recent trends. We also discuss the many innovations that have advanced in deep learning and discuss how these tools facilitate U-net. In addition, we review the different image modalities and application areas that have been enhanced by U-net.</div><div>In recent years, deep learning for health care is rapidly infiltrating and transforming medical fields thanks to the advances in computing power, data availability, and algorithm development. In particular, U-Net, a deep learning technique, has achieved remarkable success in medical image segmentation and has become one of the premier tools in this area. While the accomplishments of U-Net and other deep learning algorithms are evident, there still exist many challenges in medical image processing to achieve human-like performance. In this thesis, we propose a U-net architecture that integrates a residual skip connections and recurrent feedback with EfficientNet as a pretrained encoder. Residual connections help feature propagation in deep neural networks and significantly improve performance against networks with a similar number of parameters while recurrent connections ameliorate gradient learning. We also propose a second model that utilizes densely connected layers aiding deeper neural networks. And the proposed third model that incorporates fractal expansions to bypass diminishing gradients. EfficientNet is a family of powerful pretrained encoders that streamline neural network design. The use of EfficientNet as an encoder provides the network with robust feature extraction that can be used by the U-Net decoder to create highly accurate segmentation maps. The proposed networks are evaluated against state-of-the-art deep learning based segmentation techniques to demonstrate their superior performance.</div>

This thesis deals with the development of a complex methodology for probabilistic analysis of reliability and durability of reinforced concrete structures, especially road bridges. An estimation of load-bearing capacity is carried out using probabilistic methods in combination with the nonlinear finite element method analysis and also with respect to reliability requirements according to standards. The current state of the structure is taken into account based on the mathematical modeling of the degradation processes, such as concrete carbonation, chloride ingress and subsequent corrosion of reinforcement. Then the load-bearing capacity and reliability is predicted over time. Attention is focused on the upgrade of the analytical models for modeling of degradation processes with regard to the effect of mechanical load on the structure or its components, the development of an artificial neural network-based response surface method for efficient approximation of the original limit state function with respect to the accuracy and a minimization the number of simulations, and an interconnection of individual processes and advanced methods in an automatic process.

This master's thesis is focused on algorithmic trading on the forex market using artificial neural networks. In the introduction, there are generally described terms concerning the trading. Subsequently, in the following chapters, the thesis describes the theory of neural networks and their possible use. The practical part contains designed business strategies with neural networks. Inputs used in the network are indicators of technical analysis or directly price level. Business strategies have been implemented and tested. In the conclusion, there are summarized findings of individual business models.

Disertační práce je zaměřena na oblast inerciálních navigačních systémů a systémů, které pro odhad polohy používají pouze výpočty. Důležitým faktem v dané problematice je vysoká nepřesnost určení polohy při střednědobém a dlouhodobém využívání takového systému díky kumulativní chybě za předpokladu, že inerciální systém není podpořen žádným dalším přídavným systémem. V disertační práci jsou uvedeny možné přístupy k této problematice a návrh na zvýšení přesnosti určování polohy pouze na základě inerciálních senzorů. Základem inerciální měřicí jednotky je systém s 9 stupni volnosti, který umožňuje snímat celkové zrychlení, rychlost rotace a sílu magnetického pole, jednotlivě ve třech osách. Klíčovou myšlenkou je zařazení umělých neuronových sítí do navigačního systému tak, že jsou schopny rozpoznat charakteristické rysy pohybů, a tím zvýšit přesnost určení polohy. Popis navrhovaných metod zahrnuje analytický postup jejich vývoje a tam, kde je to možné, i analytické hodnocení jejich chování. Neuronové sítě jsou navrhovány v prostředí MATLABTM a jsou používány k určení stavu inerciální jednotky. Díky implementaci neuronových sítí lze určit pozici jednotky s řádově vyšší přesností. Aby byl inerciální polohovací systém s možností využití neuronových sítí demonstrativní, byla vyvinuta aplikace v prostředí Qt. Navržený systém a neuronové sítě byly použity při vyhodnocování reálných dat měřených senzory.

This work focuses on creating mobile application, that helps visitors of galleries and museums to find, in a more easier way, interesting information about visual art objects.

No description available.

Sectors of security, safety, and defence require methods for identifying people on the individual level. Automation of these tasks has the potential of outperforming manual labor, as well as relieving workloads. The ever-extending surveillance camera networks, advances in human pose estimation from monocular cameras, together with the progress of deep learning techniques, pave the way for automated walking gait analysis as an identification method. This thesis investigates the use of 2D kinematic pose sequences to represent gait, monocularly extracted from a limited dataset containing walking individuals captured from five camera views. The sequential information of the gait is captured using recurrent neural networks. Techniques in deep metric learning are applied to evaluate two network models, with contrasting output dimensionalities, against deep-metric-, and non-deep-metric-based embedding spaces. The results indicate that the gait representation, network designs, and network learning structure show promise when identifying individuals, scaling particularly well to unseen individuals. However, with the limited dataset, the network models performed best when the dataset included the labels from both the individuals and the camera views simultaneously, contrary to when the data only contained the labels from the individuals without the information of the camera views. For further investigations, an extension of the data would be required to evaluate the accuracy and effectiveness of these methods, for the re-identification task of each individual. / <p>Examensarbetet är utfört vid Institutionen för teknik och naturvetenskap (ITN) vid Tekniska fakulteten, Linköpings universitet</p>