31 |
Autonomous Identification and Tracking of ThermoclinesHugo Miguel Gomes Antunes 23 August 2019 (has links)
All data acquired from oceanic water features is hard and crucial work. It's hard due to the difficulty to obtain the same data given the unfavourable conditions.It requires, therefore, equipment that are reliable in the measurements of the desired characteristics and robust equipment, that is to say, equipment that are capable to withstand unfavorable and variable conditions in spatial and temporal terms. Due to these same spatial and temporal changes, the traditional methods do not prove to be the most adequate, because these methods do not have sufficient capacity to sample measurements of the dynamic characteristics of oceanographic processes.Thus, to obtain such measurements the use of the autonomous robotic systems proves to be important. With these systems, it is ensured a faster, more efficient and systematic sampling and is not subject to human error. The data acquisition is then a crucial work to understand how oceanographic process happens and varies in time and space. This work proposes an implementation of an algorithm to perform the tracking of the thermocline, from the stratification model of the oceanic water.This model is a parametric model. This work will also take into account the capacity to perform measurements with a sampling capable of adapting the depth control of the underwater vehicle.The stratification of the oceanic water happens when exists different features between different layers. One of these layers is the thermocline. At this layer, the water temperature decreases rapidly with increasing depth. The characterization of the thermocline is so important to marine biology, given the high concentration of phytoplankton in this level, as for acoustic communications equipments or military services, given the special characteristics of speed sound in this level.The model of this stratification will be used to aid in the thermocline's tracking process. This model will serve as a basis for the algorithm to adapt the control in order to carry out the tracking with the greatest success, in real time. This algorithm will focus on the variations in the vertical temperature gradient.The algorithm responsible detect and track of the thermocline will be run on a profiler. The profiler is a vehicle that moves along the vertical axis. However, when subject to tides, the natural process in aquatic environments drifts along the horizontal axis. A set of sensors capable of measuring the water temperature and the depth at which the vehicle is below water shall be placed in this vehicle. These sensors will be important to calculate the vertical gradient.
|
32 |
Robotic simulator for the Tactode tangible block programming systemMárcia Sofia dos Santos Alves 22 October 2019 (has links)
No description available.
|
33 |
A machine learning approach to the optimization of inventory management policiesÁlvaro Silva de Melo 22 October 2019 (has links)
No description available.
|
34 |
Simulação e Melhoramento do PiTank com Sistema de Inteligência ArtificialSérgio Daniel Marinho de Lima Teixeira 22 October 2019 (has links)
No description available.
|
35 |
Real-Time Location Systems and Internet of Things SensorsFilipe Manuel Ferreira Cordeiro 22 October 2019 (has links)
No description available.
|
36 |
3D Lung Nodule Classification in Computed Tomography ImagesAna Rita Felgueiras Carvalho 31 October 2019 (has links)
Lung cancer is the leading cause of cancer death worldwide. One of the reasons is the absence of symptoms at an early stage, which means that it is only discovered at a later stage, where the treatment is more difficult [1]. Furthermore, when making a diagnosis, frequently done by reading computed tomographies (CT's), it is regularly allied with errors. One of the reasons is the variation of the opinion of the doctors regarding the diagnosis of the same nodule [2,3].The use of CADx, Computer-Aided Diagnosis, systems can be a great help for this problem by assisting doctors in diagnosis with a second opinion. Although its efficiency has already been proven [4], it often ends up not being used because doctors can not understand the "how and why" of CADx diagnostic results, and ultimately do not trust the system [5]. To increase the radiologists' confidence in the CADx system it is proposed that along with the results of malignancy prediction, there are also results with evidence that explains those malignancy results.There are some visible features in lung nodules that are correlated with malignancy. Since humans are able to visually identify these characteristics and correlate them with nodule malignancy, one way to present those evidence is to make predictions of those characteristics. To have these predictions it is proposed to use deep learning approaches. Convolutional neural networks had shown to outperform the state of the art results in medical image analysis [6]. To predict the characteristics and malignancy in CADx system, the architecture HSCNN, a deep hierarchical semantic convolutional neural network, proposed by Shen et al. [7], will be used.The Lung Image Database Consortium image collection (LIDC-IDRI) public dataset is frequently used as input for lung cancer CADx systems. The LIDC-IDRI consists of thoracic CT scans, presenting a lot of data's quantity and variability. In most of the nodules, this dataset has doctor's evaluations for 9 different characteristics. A recurrent problem in those evaluations is the subjectivity of the doctors' interpretation in what each characteristic is. In some characteristics, it can result in a great divergence in evaluations regarding the same nodule, which makes the inclusion of those evaluations as an input in CADx systems not useful as it could be. To reduce this subjectivity, it is proposed the creation of a metric that makes the characteristics classification more objective. For this, it is planned bibliographic and LIDC-IDRI dataset reviews. With that, taking into account this new metric, validated after by doctors from Hospital de São João, will be made a reclassification in LIDC-IDRI dataset. This way it could be possible to use as input all the relevant characteristics. The principal objective of this dissertation is to develop a lung nodule CADx system methodology which promotes the confidence of specialists in its use. This will be made classifying lung nodules according to relevant characteristics to diagnosis and malignancy. The reclassified LIDC-IDRI dataset will be used as an input for CADx system and the architecture used for predicting the characteristics and malignancy results will be the HSCNN. To measure the classification evaluation will be used sensitivity, sensibility, and area under the Receiver Operating Characteristic (ROC), curve. The proposed solution may be used for improving a CADx system, LNDetector, currently in development by the Center for Biomedical Engineering Research (C-BER) group from INESC-TEC in which this work will be developed.[1] - S. Sone M. Hasegawa and S. Takashima. Growth rate of small lung cancels detected on mass ct screening. Tire British Journal of Radiology, pages 1252-1259[2] - D. J. Bell S. E. Marley P. Guo H. Mann M. L. Scott L. H. Schwartz D. C. Ghiorghiu B. Zhao, Y. Tan. Exploring intra-and inter-reader variability in uni-dimensional, bi-dimensional, and volumetric measurements of solid tumors on ct scans reconstructed at different slice intervals. European journal of radiology 82, page 959-968, 2013[3] - H.T Winer-Muram. The solitary pulmonary nodule 1. Radiology, 239, pages 39-49, 2006.[4] - R. Yan J. Lee L. C. Chu C. T. Lin A. Hussien J. Rathmell B. Thomas C. Chen et al. P. Huang, S. Park. Added value of computer-aided ct image features for early lung cancer diagnosis with small pulmonary nodules: A matched case-control study. Radiology 286, page 286-295, 2017[5] - W Jorritsma, Fokie Cnossen, and Peter Van Ooijen. Improving the radiologist-cad interaction: Designing for appropriate trust. Clinical Radiology, 70, 10 2014.[6] - Tom Brosch, Youngjin Yoo, David Li, Anthony Traboulsee, and Roger Tam. Modeling the variability in brain morphology and lesion distribution in multiple sclerosis by deep learning. Volume 17, 09 2014.[7] - Simon Aberle Deni A. T. Bui Alex Hsu Willliam Shen, Shiwen X. Han. An interpretable deep hierarchical semantic convolutional neural network for lung nodule malignancy classification. june 2018
|
37 |
Framework for Developing Interactive 360-Degree Video Adventure GamesFrancisco José Rodrigues de Pinho 11 October 2019 (has links)
No description available.
|
38 |
Desempenho do Motor de Indução Trifásico Alimentado por Conversor de Frequência (PWM)Diogo Mendes de Vasconcelos Pinto Rodrigues 01 October 2019 (has links)
No description available.
|
39 |
Automatic Conversion of Cooking Recipes to Grocery ListsMarcelo Diocleciano Rodrigues Ferreira 11 October 2019 (has links)
No description available.
|
40 |
Growing Artificial Societies to Support Demand Modelling in Mobility-as-a-Service SolutionsManuel António Gonçalves Gomes 11 October 2019 (has links)
Tráfego intenso, congestionamentos e tempos de deslocamento mais longos são consequência doaumento da população, da continuação da posse de carro próprio e do fim do transporte público derota fixa. Embora esta situação tenha criado alguma pressão sobre as autoridades governamentaispara lidar com as questões acima mencionadas, isso também pode provar ser uma oportunidadepara numa nova abordagem ao conceito de mobilidade.Uma possível solução passaria pela Mobilidade como serviço (MaaS), um conceito relativa-mente novo no paradigma de mobilidade, que promete mudar em termos do que é mobilidade ecomo ela é entregue aos usuários finais. Fazendo uso das atuais infraestruturas físicas e meios detransporte, e combinando-as com tecnologias da informação e comunicação (TICs), o MaaS temcomo principal objetivo entregar a mobilidade aos usuários finais como um serviço que é consum-ido através de uma plataforma. Essas plataformas são baseadas em modelos de mercado, onde umregulador é responsável pelo equilíbrio entre oferta e demanda.As Sociedades Artificiais (AS) pretendem ser uma forma de simular sociedades reais, atravésde um modelo artificial de agentes proativos e dinâmicos, capazes de interagir entre eles. Essesagentes são capazes de se comunicar entre eles através de uma rede social, onde várias regras sãousadas para disciplinar e normalizar os agentes e o ambiente onde eles estão contidos.A modelação da demanda (DM) é um conceito que permite prever com precisão a demandapor algum mercado, dependendo do equilíbrio entre oferta e demanda. Além disso, tendo em contaa presença do regulador, responsável pela manutenção e implementação de políticas de regulação,o DM facilita a modelação de toda essa dinâmica.A análise dos melhores modelos de serviços, pode ser muito benéfica para o MaaS, uma vezque a modelação de metodologias novas e mais precisas poderia melhorar os processos de decisãopresentes nos vários modelos de mercado do MaaS.Este trabalho tem como objetivo desenvolver um metamodelo cognitivo de sistema multi-agente capaz de descrever a dinâmica do conceito de MaaS. O metamodelo desenvolvido deveser capaz de suportar diferentes estratégias deliberativas e de tomada de decisão em ambientes demercado de serviços abertos, com aplicações de mobilidade em Cidades Inteligentes. O objetivo édesenvolver uma plataforma de apoio à decisão para apoiar a análise e implementação de políticasde incentivo que promovam o desenvolvimento do conceito de MaaS. Esta plataforma fará uso detécnicas de modelagem e simulação de sistemas complexos recorrendo às metáforas de sociedadesartificiais e sistemas multiagentes. / Huge traffic, congestion, longer commute times, are a consequence of the increase in population,continuation of universal car ownership and demise of fixed route public transport. While thissituation have been creating some pressure on the governmental authorities to tackle the afore-mentioned issues, this could also prove to be an opportunity to try a different approach regardingthe concept of mobility.One particular solution could be Mobility-as-a-Service (MaaS), a relatively new concept in amobility paradigm that promises a big shift in terms of what is mobility and how it is deliveredto the end-users. Making use the current physical infrastructures and transport means, and com-bining them with information and communications technologies (ICTs), MaaS has the main goalto delivery the mobility to the end-users as a service that is consumed through a platform. Theseplatforms are based on market models, where there a regulator that is responsible for the balancethe balance between supply and demand.Artificial Societies (AS) aims to be a way to simulate real societies, through an artificial modelof proactive and dynamic agents, able to interact between them. These agents are able to commu-nicate between them through a social network, where several rules are used to discipline and normboth agents and the environment where they are contained.Demand modelling (DM) is a concept that allows accurately to forecast the demand regardingsome market, depending of the balance between supply and demand. Moreover, taken into accountthe presence of the regulator, which is responsible for the maintenance and implementation ofpolicies, DM facilitates the modelling of all this dynamic.The analysis of the best service models, could prove greatly beneficial for MaaS, as modelingnew and more accurate methodologies could better the decision processes present in the variousmarket models of MaaS.This work aims to develop a cognitive multi-agent system meta-model able to describe thedynamic of MaaS concept. The developed meta-model should be able to support different de-liberative and decision making strategies in open service market environments, with mobility ap-plications in Smart Cities. The purpose is to develop a decision support platform to support theanalysis and implementation of incentive policies that promote the development of the concept ofMaaS. This platform will make use of techniques of modeling and simulation of complex systemsresorting to the metaphors of artificial societies and multi-agent systems.
|
Page generated in 0.0799 seconds