Structural health monitoring and building assessment are crucial to acquire structures’ states and maintain their conditions. Besides human-labor surveys that are subjective, time-consuming, and expensive, autonomous image and video analysis is a faster, more efficient, and non-destructive way. This thesis focuses on crack detection from videos, crack segmentation from images, and building assessment from street view images. For crack detection from videos, three approaches are proposed based on local binary pattern (LBP) and support vector machine (SVM), deep convolution neural network (DCNN), and fully-connected network (FCN). A parametric Naïve Bayes data fusion scheme is introduced that registers video frames in a spatiotemporal coordinate system and fuses information based on Bayesian probability to increase detection precision. For crack segmentation from images, the rotation-invariant property of crack is utilized to enhance the segmentation accuracy. The architectures of several approximately rotation-invariant DCNNs are discussed and compared using several crack datasets. For building assessment from street view images, a framework of multiple DCNNs is proposed to detect buildings and predict their attributes that are crucial for flood risk estimation, including founding heights, foundation types (pier, slab, mobile home, or others), building types (commercial, residential, or mobile home), and building stories. A feature fusion scheme is proposed that combines image feature with meta information to improve the predictions, and a task relation encoding network (TREncNet) is introduced that encodes task relations as network connections to enhance multi-task learning.

Various research has shown the potential and robustness of deep learning-based approaches to synthesise novel motions of 3D characters in virtual environments, such as video games and films. The models are trained with the motion data that is bound to the respective character skeleton (rig). It inflicts a limitation on the scalability and the applicability of the models since they can only learn motions from one particular rig (domain) and produce motions in that domain only. Transfer learning techniques can be used to overcome this issue and allow the models to better adapt to other domains with limited data. This work presents a study of three transfer learning techniques for the proposed Objective-driven motion generation model (OMG), which is a model for procedurally generating animations conditioned on positional and rotational objectives. Three transfer learning approaches for achieving rig-agnostic encoding (RAE) are proposed and experimented with: Feature encoding (FE), Feature clustering (FC) and Feature selection (FS), to improve the learning of the model on new domains with limited data. All three approaches demonstrate significant improvement in both the performance and the visual quality of the generated animations, when compared to the vanilla performance. The empirical results indicate that the FE and the FC approaches yield better transferring quality than the FS approach. It is inconclusive which of them performs better, but the FE approach is more computationally efficient, which makes it the more favourable choice for real-time applications. / Många studier har visat potentialen och robustheten av djupinlärningbaserade modeller för syntetisering av nya rörelse för 3D karaktärer i virtuell miljö, som datorspel och filmer. Modellerna är tränade med rörelse data som är bunden till de respektive karaktärskeletten (rig). Det begränsar skalbarheten och tillämpningsmöjligheten av modellerna, eftersom de bara kan lära sig av data från en specifik rig (domän) och därmed bara kan generera animationer i den domänen. Kunskapsöverföringsteknik (transfer learning techniques) kan användas för att överkomma denna begränsning och underlättar anpassningen av modeller på nya domäner med begränsade data. I denna avhandling presenteras en studie av tre kunskapsöverföringsmetoder för den föreslagna måldriven animationgenereringsnätverk (OMG), som är ett neural nätverk-baserad modell för att procedurellt generera animationer baserade på positionsmål och rotationsmål. Tre metoder för att uppnå rig-agnostisk kodning är presenterade och experimenterade: Feature encoding (FE), Feature clustering (FC) and Feature selection (FS), för att förbättra modellens lärande på nya domäner med begränsade data. All tre metoderna visar signifikant förbättring på både prestandan och den visuella kvaliteten av de skapade animationerna, i jämförelse med den vanilla prestandan. De empiriska resultaten indikerar att både FE och FC metoderna ger bättre överföringskvalitet än FS metoden. Det går inte att avgöra vilken av de presterar bättre, men FE metoden är mer beräkningseffektiv, vilket är fördelaktigt för real-time applikationer.

<p>Ni-based superalloys continue to exert themselves as the industry standards in high stress and highly corrosive/oxidizing environments, such as are present in a gas turbine engine, due to their excellent high temperature strengths, thermal and microstructural stabilities, and oxidation and creep resistances. Gas turbine engines are essential components for energy generation and propulsion in the modern age. However, Ni-based superalloys are reaching their limits in the operating conditions of these engines due to their melting onset temperatures, which is approximately 1300 °C. Therefore, a new class of materials must be formulated to surpass the capabilities Ni-based superalloys, as increasing the operating temperature leads to increased efficiency and reductions in fuel consumption and greenhouse gas emissions. One of the proposed classes of materials is termed refractory complex concentrated alloys, or RCCAs, which consist of 4 or more refractory elements (in this study, selected from: Ti, Zr, Hf, V, Nb, Ta, Cr, Mo, and W) in equimolar or near-equimolar proportions. So far, there have been highly promising results with these alloys, including far higher melting points than Ni-based superalloys and outstanding high-temperature strengths in non-oxidizing environments. However, improvements in room temperature ductility and high-temperature oxidation resistance are still needed for RCCAs. Also, given the millions of possible alloy compositions spanning various combinations and concentrations of refractory elements, more efficient methods than just serial experimental trials are needed for identifying RCCAs with desired properties. A coupled computational and experimental approach for exploring a wide range of alloy systems and compositions is crucial for accelerating the discovery of RCCAs that may be capable of replacing Ni-based superalloys. </p> <p>In this thesis, the CALPHAD method was utilized to generate basic thermodynamic properties of approximately 67,000 Al-bearing RCCAs. The alloys were then down-selected on the basis of certain criteria, including solidus temperature, volume percent BCC phase, and aluminum activity. Machine learning models with physics-based descriptors were used to select several BCC-based alloys for fabrication and characterization, and an active learning loop was employed to aid in rapid alloy discovery for high hardness and strength. This method resulted in rapid identification of 15 BCC-based, four component, Al-bearing RCCAs exhibiting room-temperature Vickers hardness from 1% to 35% above previously reported alloys. This work exemplifies the advantages of utilizing Integrated Computational Materials Engineering- and Materials Genome Initiative-driven approaches for the discovery and design of new materials with attractive properties.</p> <p> </p> <p><br></p>

Learning-based testing is a testing paradigm that combines model-based testing with machine learning algorithms to automate the modeling of the SUT, test case generation, test case execution and verdict construction. A tool that implements LBT been developed at the CSC school at KTH called LBTest. LBTest utilizes machine learning algorithms with off-the-shelf equivalence- and model-checkers, and the modeling of user requirements by propositional linear temporal logic. In this study, it is be investigated whether LBT may be suitable for testing a micro bus architecture within an embedded telecommunication device. Furthermore ideas to further automate the testing process by designing a data model to automate user requirement generation are explored. / Inlärningsbaserad testning är en testningsparadigm som kombinerar model-baserad testning med maskininlärningsalgoritmer för att automatisera systemmodellering, testfallsgenering, exekvering av tester och utfallsbedömning. Ett verktyg som är byggt på LBT är LBTest, utvecklat på CSC skolan på KTH. LBTest nyttjar maskininlärningsalgoritmer med färdiga ekvivalent- och model-checkers, och modellerar användarkrav med linjär temporal logik. I denna studie undersöks det om det är lämpat att använda LBT för att testa en mikrobus arkitektur inom inbyggda telekommunikationsenheter. Utöver det undersöks även hur testprocessen skulle kunna ytterligare automatiseras med hjälp av en data modell för att automatisera generering av användarkrav.

<p>  This paper explored the application of three machine learning models focused on predictive motor maintenance. Logistic Regression, Sequential Minimal Optimization (SMO), and NaïveBayes models. A comparative analysis of these models illustrated that while each had an accuracy greater than 95% in this study, the Logistic Regression Model exhibited the most reliable operation.</p>

This work implements a prototype of a music practice tool, and evaluates alignment methods in an audio matching scenario required for it. By two interviews with piano teachers, we investigated the user demands towards a music performance practice tool that incorporates an alignment technique between a shorter practice segment and a reference performance, from a jazz and classical music point of view. Regarding technical aspects, we studied how Deep Learning (DL) based signal representations compare to standard manually tailored features in the alignment task. Experiments were conducted using a well-known alignment algorithm on a piano dataset. The dataset had manually annotated beat positions which was used for evaluation. We found the traditional features to be superior compared with the DL based signal representations when used independently. We also found that the DL based signal representations, on their own, were insufficient for our test cases. However we found that the DL representations contained valuable information. Multiple test cases demonstrated that the combination of DL representations and traditional representations outperformed all other considered approaches. We also did experiments using deadpan midi renditions as references instead of actual performances, in which we got slight, but insignificant improvement in alignment performance. Finally, the prototype was implemented as a website, using a traditional signal representation as input to the alignment algorithm. / Detta arbete implementerar en prototyp av ett musikövningsverktyg och utvärderar ljudjusteringssmetoder som krävs för det. Användarkraven för verktyget undersöktes genom två intervjuer med pianolärare och fokuserade på ljudmatchning mellan en kort övningsinspelning och en referensinspelning, fokuserat på jazz och klassisk musik. De tekniska aspekterna inkluderade en jämförelse mellan djupinlärningsbaserade signalrepresentationer och traditionella manuellt anpassade funktioner i ljudmatchningsuppgiften. Experiment utfördes på ett pianodataset med en välkänd ljudjusterings algoritm, anpassad för ljudmatchning. Datasetet hade manuellt annoterade taktpositioner som användes för utvärdering. Vi fann att de traditionella funktionerna var överlägsna jämfört med djupinlärningsbaserade signalrepresentationer när de användes ensamma. Vi fann också att djupinlärningsbaserade-baserade signalrepresentationer, ensamma, var otillräckliga för våra testfall. Dock upptäckte vi att de djupinlärningsbaserade representationerna innehöll värdefull information. Flera testfall visade att kombinationen av djupinlärnings-representationer och traditionella representationer överträffade alla andra övervägda metoder. Test med midi-renderade inspelningar som referenser visade en svag, men insignifikant förbättring i prestanda. Slutligen implementerades en prototyp av övningsverktyget som en webbplats, med en traditionell signalrepresentation som inmatning till matchningsalgoritmen.

Burn injuries are a significant global health concern, causing high mortality and morbidity rates. Clinical assessment is the current standard for diagnosing burn injuries, but it suffers from interobserver variability and is not suitable for intermediate burn depths. To address these challenges, machine learning-based techniques were proposed to evaluate burn wounds in a thesis. The study utilized image-based networks to analyze two medical image databases of burn injuries from Caucasian and Black-African cohorts. The deep learning-based model, called BurnsNet, was developed and used for real-time processing, achieving high accuracy rates in discriminating between different burn depths and pressure ulcer wounds. The multiracial data representation approach was also used to address data representation bias in burn analysis, resulting in promising performance. The ML approach proved its objectivity and cost-effectiveness in assessing burn depths, providing an effective adjunct for clinical assessment. The study's findings suggest that the use of machine learning-based techniques can reduce the workflow burden for burn surgeons and significantly reduce errors in burn diagnosis. It also highlights the potential of automation to improve burn care and enhance patients' quality of life. / Petroleum Technology Development Fund (PTDF); Gombe State University study fellowship

[pt] O presente trabalho inicia com uma investigação sobre o uso de modelos de Aprendizado Profundo 3D para a detecção aprimorada de morangos em túneis de cultivo. Focou-se em duas tarefas principais: primeiramente, a detecção de frutas, comparando o modelo original MaskRCNN com uma versão adaptada que integra informações de profundidade (MaskRCNN-D). Ambos os modelos são capazes de classificar morangos baseados em sua maturidade (maduro, não maduro) e estado de saúde (afetados por doença ou fungo). Em segundo lugar, focou-se em identificar a região mais ampla dos morangos, cumprindo um requisito para um sistema de espectrômetro capaz de medir o conteúdo de açúcar das frutas. Nesta tarefa, comparouse um algoritmo baseado em contorno com uma versão aprimorada do modelo VGG-16. Os resultados demonstram que a integração de dados de profundidade no MaskRCNN-D resulta em até 13.7 por cento de melhoria no mAP através de diversos conjuntos de teste de morangos, incluindo os simulados, enfatizando a eficácia do modelo em cenários agrícolas reais e simulados. Além disso, nossa abordagem de solução ponta-a-ponta, que combina a detecção de frutas (MaskRCNN-D) e os modelos de identificação da região mais ampla (VGG-16 aprimorado), mostra um erro de localização notavelmente baixo, alcançando até 11.3 pixels de RMSE em uma imagem de morango cortada de 224 × 224. Finalmente, explorou-se o desafio de aprimorar a qualidade das leituras de dados do espectrômetro através do posicionamento automático do sensor. Para tal, projetou-se e treinou-se um modelo de Aprendizado Profundo com dados simulados, capaz de prever a acurácia do sensor com base em uma imagem dada de um morango e o deslocamento desejado da posição do sensor. Usando este modelo, calcula-se o gradiente da saída de acurácia em relação à entrada de deslocamento. Isso resulta em um vetor indicando a direção e magnitude com que o sensor deve ser movido para melhorar a acurácia do sinal do sensor. Propôs-se então uma solução de Servo Visão baseada neste vetor, obtendo um aumento significativo na acurácia média do sensor e melhoria na consistência em novas iterações simuladas. / [en] The present work begins with an investigation into the use of 3D Deep Learning models for enhanced strawberry detection in polytunnels. We focus on two main tasks: firstly, fruit detection, comparing the standard MaskRCNN with an adapted version that integrates depth information (MaskRCNN-D). Both models are capable of classifying strawberries based on their maturity (ripe, unripe) and health status (affected by disease or fungus). Secondly, we focus on identifying the widest region of strawberries, fulfilling a requirement for a spectrometer system capable of measuring their sugar content. In this task, we compare a contour-based algorithm with an enhanced version of the VGG-16 model. Our findings demonstrate that integrating depth data into the MaskRCNN-D results in up to a 13.7 percent improvement in mAP across various strawberry test sets, including simulated ones, emphasizing the model s effectiveness in both real-world and simulated agricultural scenarios. Furthermore, our end-to-end pipeline approach, which combines the fruit detection (MaskRCNN-D) and widest region identification models (enhanced VGG-16), shows a remarkably low localization error, achieving down to 11.3 pixels of RMSE in a 224 × 224 strawberry cropped image. Finally, we explore the challenge of enhancing the quality of the data readings from the spectrometer through automatic sensor positioning. To this end, we designed and trained a Deep Learning model with simulated data, capable of predicting the sensor accuracy based on a given image of the strawberry and the subsequent displacement of the sensor s position. Using this model, we calculate the gradient of the accuracy output with respect to the displacement input. This results in a vector indicating the direction and magnitude with which the sensor should be moved to improve the sensor signal accuracy. A Visual Servoing solution based on this vector provided a significant increase in the average sensor accuracy and improvement in consistency across new simulated iterations.

Submitted by Johnny Rodrigues (johnnyrodrigues@ufcg.edu.br) on 2018-05-07T15:29:04Z No. of bitstreams: 1 ADSON DIEGO DIONISIO DA SILVA - DISSERTAÇÃO PPGCC 2015..pdf: 2453030 bytes, checksum: 863c817f9714377b827d4d6fa0770c51 (MD5) / Made available in DSpace on 2018-05-07T15:29:04Z (GMT). No. of bitstreams: 1 ADSON DIEGO DIONISIO DA SILVA - DISSERTAÇÃO PPGCC 2015..pdf: 2453030 bytes, checksum: 863c817f9714377b827d4d6fa0770c51 (MD5) Previous issue date: 2015-08-31 / O reconhecimento automático de eventos de interesse em vídeos envolvendo conjuntos de ações ou de interações entre objetos. Pode agregar valor a sistemas de vigilância,aplicações de cidades inteligentes, monitoramento de pessoas com incapacidades físicas ou mentais, dentre outros. Entretanto, conceber um arcabouço que possa ser adaptado a diversas situações sem a necessidade de um especialista nas tecnologias envolvidas, continua sendo um desafio para a área. Neste contexto, a pesquisa realizada tem como base a criação de um arcabouço genérico para detecção de eventos em vídeo com base em regras. Para criação das regras, os usuários formam expressões lógicas utilizando Lógica de Primeira Ordem e relacionamos termos com a álgebra de intervalos de Allen, adicionando assim um contexto temporal às regras. Por ser um arcabouço, ele é extensível, podendo receber módulos adicionais para realização de novas detecções e inferências Foi realizada uma avaliação experimental utilizando vídeos de teste disponíveis no site Youtube envolvendo um cenário de trânsito, com eventos de ultrapassagem do sinal vermelho e vídeos obtidos de uma câmera ao vivo do site Camerite, contendo eventos de carros estacionando. O foco do trabalho não foi criar detectores de objetos (e.g. carros ou pessoas) melhores do que aqueles existentes no estado da arte, mas propor e desenvolver uma estrutura genérica e reutilizável que integra diferentes técnicas de visão computacional. A acurácia na detecção dos eventos ficou no intervalo de 83,82% a 90,08% com 95% de confiança. Obteve acurácia máxima (100%) na detecção dos eventos, quando substituído os detectores de objetos por rótulos atribuídos manualmente, o que indicou a eficácia do motor de inferência desenvolvido para o arcabouço. / Automatic recognition of relevant events in videos involving sets of actions or interactions between objects can improve surveillance systems, smart cities applications, monitoring of people with physical or mental disabilities, among others. However, designing a framework that can be adapted to several situations without an expert in the involved technologies remains a challenge. In this context, this work is based on the creation of a rule-based generic framework for event detection in video. To create the rules, users form logical expressions using firstorder logic (FOL) and relate the terms with the Allen’s interval algebra, adding a temporal context to the rules. Once it is a framework, it is extensible, and may receive additional modules for performing new detections and inferences. Experimental evaluation was performed using test videos available on Youtube, involving a scenario of traffic with red light crossing events and videos from Camerite website containing parking car events. The focus of the work was not to create object detectors (e.g. cars or people) better than those existing in the state-of-the-art, but, propose and develop a generic and reusable framework that integrates differents computer vision techniques. The accuracy in the detection of the events was within the range of 83.82% and 90.08% with 95% confidence. Obtained maximum accuracy (100 %) in the detection of the events, when replacing the objects detectors by labels manually assigned, what indicated the effectiveness of the inference engine developed for this framework.

Recent improvements in the area of Internet of Things (IoT) has led to extensive utilization of embedded devices and sensors. Hence, along with utilization the need for safety and security of these devices also increases proportionately. In the last two decades, the side-channel attack (SCA) has become a massive threat to the interrelated embedded devices. Moreover, extensive research has led to the development of many different forms of SCA for extracting the secret key by utilizing the various leakage information. Lately, machine learning (ML) based models have been more effective in breaking complex encryption systems than the other types of SCA models. However, these ML or DL models require a lot of data for training that cannot be collected while attacking a device in a real-world situation. Thus, in this thesis, we try to solve this issue by proposing the new technique of leakage conversion. In this technique, we try to convert the high signal to noise ratio (SNR) power traces to low SNR averaged electromagnetic traces. In addition to that, we also show how artificial neural networks (ANN) can learn various non-linear dependencies of features in leakage information, which cannot be done by adaptive digital signal processing (DSP) algorithms. Initially, we successfully convert traces in the time interval of 80 to 200 as the cryptographic operations occur in that time frame. Next, we show the successful conversion of traces lying in any time frame as well as having a random key and plain text values. Finally, to validate our leakage conversion technique and the generated traces we successfully implement correlation electromagnetic analysis (CEMA) with an approximate minimum traces to disclosure (MTD) of 480.