Navegando por Autor "Santos, Alyson de Jesus dos Santos"

Agora exibindo 1 - 6 de 6

Aplicação de controle PI e PID na construção para dispositivo de resfriamento: uma proposta prática
(2023-09-29) Souza, Matheus Afonso Lima de; Neves, Cleonor Crescencio das; http://lattes.cnpq.br/2176855362313017; Reis, Ailton Gonçalves; http://lattes.cnpq.br/7119484789256081; Neves, Cleonor Crescêncio das; http://lattes.cnpq.br/2176855362313017; Reis, Ailton Gonçalves; http://lattes.cnpq.br/7119484789256081; Santos, Eliton Smith dos; Santos, Alyson de Jesus dos Santos; http://lattes.cnpq.br/5998752909180697
This undergraduate thesis details the development of a prototype for temperature control in a closed environment using PID control techniques. The prototype, housed in an MDF box, regulates a 100-watt lamp via a dimmer to maintain the voltage at 33 Vac. DHT11 and DS18B20 sensors are employed to monitor internal and external temperatures. Real-time sensor data is collected and analyzed through MATLAB-generated graphs. Cooling is facilitated by a 12Vdc fan controlled through PWM signals from the ESP8266 and a BC458 transistor. The PID control implementation, using the Arduino's PID_v1 library, is fine-tuned based on open loop responses and COHEN & COON relationships. Results highlight the energy consumption for cooling concerning the external temperature, establishing an effective temperature control solution for controlled environments.
Desenvolvimento de Hardware e firmware para gerenciamento de energia elétrica
(2024-03-15) Nogueira, Warley Matheus Santos; Rodrigues, Marlos André Silva; http://lattes.cnpq.br/0650250324042531; Rodrigues, Marlos André Silva; http://lattes.cnpq.br/0650250324042531; Santos, Alyson de Jesus dos Santos; http://lattes.cnpq.br/5998752909180697; Nascimento, José Fábio de Lima; ttp://lattes.cnpq.br/0056522906128231
A indústria global, em meio às preocupações ambientais, busca reduzir sua pegada de carbono através de estratégias sustentáveis, com ênfase na eficiência energética. Esta que não apenas diminuem o consumo de eletricidade, crucial devido à indústria ser uma grande consumidora de energia, mas também mitigam as mudanças climáticas e elevam a competitividade ao reduzir custos e aumentar a produtividade. A eficiência energética, uma solução abrangente, encontra destaque nas tecnologias emergentes, como a utilização de sistema sem fio como o sistema LoRa (Long Range). Aplicada no Gerenciamento de Consumo de Energia Elétrica baseado em Internet das coisas (Internet of Thing ou IoT), a LoRa eficientemente monitora e controla o consumo em larga escala, destacando-se por transmitir dados em longas distâncias com baixo consumo de energia, vital em ambientes industriais extensos. Entretanto, a implantação da LoRa enfrenta desafios, como a integração de sistemas legados, garantia da segurança dos dados e otimização da rede para máximo desempenho com mínimo consumo energético. Baseado nestas premissas, este trabalho tem o objetivo de desenvolvimento de hardware e firmware para gerenciamento de energia em uma indústria de Manaus, possibilitando uma gestão eficiente em ambientes industriais, permitindo medidas corretivas para redução do consumo e otimização do uso da energia elétrica.
Prevenção contra deepfakes: desenvolvimento de um sistema de reconhecimento facial para diferenciar rostos humanos de rostos gerados por IA
(2024-03-19) Silva, Romão Charles Silva e; Santos, Alyson de Jesus dos Santos; http://lattes.cnpq.br/5998752909180697; Santos, Alyson de Jesus dos; http://lattes.cnpq.br/5998752909180697; Martins, Gilbert Breves; http://lattes.cnpq.br/4932200790121123; Rodrigues, Marlos André Silva; http://lattes.cnpq.br/0682962508867807
This work aims to develop a system based on machine learning that differentiates images of people generated by artificial intelligence from real people, a capability that can be very useful for identifying scams that use generated images. The development of the project was done in 3 main steps: data organization, training and testing. The system was entirely made in Google Colab, therefore, it used Python and the main development tool Tensorflow, two models were trained, one of which has almost twice as many images used for training, with the intention of observing the consequences of using a larger set. At the end of the project, the quantitative and qualitative results of the image classification are shown.
Simulação de um sistema automatizado de inscrições de rótulos de baterias webots
(2024-08-05) Feitosa, Paulo Rafael Rodrigues; Santos, Alyson de Jesus dos Santos; http://lattes.cnpq.br/5998752909180697; Santos, Alyson de Jesus dos; http://lattes.cnpq.br/5998752909180697; Guerreiro, Gabriel Rebello; Souza, Daniel Fonseca de; http://lattes.cnpq.br/4043793492782488
This work consists of a simulation in Webots of a visual inspection system for information printed on cell phone battery labels using pad printing. The environment includes an entry conveyor where the batteries are inserted. A camera attached to a UR5e robot captures images of the batteries and sends them to the inspection algorithm developed in Python. The inspection algorithm comprises an OCR (Optical Character Recognition) system to check if the textual information is present on the label and a classification model to verify if the symbols have been printed on the label. The UR5e robot then separates the approved and rejected batteries.
Uso de redes neurais para identificação de embarcações
(2024-09-23) Pedroso, Ytalo Ribeiro da Silva; Santos, Alyson de Jesus dos Santos; http://lattes.cnpq.br/5998752909180697; Santos, Alyson de Jesus dos; http://lattes.cnpq.br/5998752909180697; Souza, Daniel Fonseca de; http://lattes.cnpq.br/4043793492782488; Rodrigues, Marlos André Silva; http://lattes.cnpq.br/0650250324042531
Due toitsuniquegeography,theriversoftheAmazonplayasignificant roleinthedailylife of theregion,includingbeingtheonlyaccessinmoreremotecommunities,andthelackof infrastructure inmuchofthiswaterwaynetworkmakesitconducivetothetransportofillegal goods duetothelackofsupervision. When comparedtourbanareas,itispossibletonoticethepotentialforpositiveassistanceto local authoritiesthroughtheuseofrecognitiontechnologies,butthelackofcommunication structures hinderstheimplementationofthesetechnologiesinthesamewayitisdoneincities. This projectproposesthedevelopmentofaneuralnetworkapplicationaimedatassistingin public securityissuesthroughthedetectionofvesselsinportregionslocatedintheStateof Amazonas, combiningArtificial IntelligencetechniquesandEmbeddedSystemsdevices,without the useofnetworkdatatransmission.Themethodologyconsistsof:(a)Acquisitionofvessel images atMarinadoDaviandtheportofManausModerna-StateofAmazonas;(b)Preparation of thedatasetintheexpectedformatforthemodel;(c)Installation,configuration, anduseofthe Jetson NanoBoardintrainingtheNeuralNetwork;(d)UseofTransferLearning,leveraginga pre-trained model(YOLOv4Tiny)formodelgenerationonthenewdataset;(e)Validationofthe obtained results. The resultsshowthatwiththeuseofversionsoptimizedforembeddeddevices,itwaspossible to executetheYOLOv4Tinyalgorithm,performingtheclassification ofvesselsandcontributing with newvesselimages,whichprovidegreaterregionalitytothepresenteddata.
Uso de uma rede neural convolucional para detecção de covid-19 automática através de imagens de Raio-x
(2023-12-22) Cavalcante, Vinícius Loureiro; Santos, Alyson de Jesus dos Santos; http://lattes.cnpq.br/5998752909180697; Santos, Alyson de Jesus dos; http://lattes.cnpq.br/5998752909180697; Santos, Lucèlia Cunha da Rocha; http://lattes.cnpq.br/2242046166554146; Fialho, Michaella Socorro Bruce; http://lattes.cnpq.br/9348859124436505
This study aims to evaluate the effectiveness of using neural networks in the detection of COVID-19 through chest X-rays. Based on a literature review, the methodology for building the neural network will be defined, and it will be trained with data collected from reliable sources and analyzed to evaluate the accuracy of detection. The use of neural networks can be a promising and non-invasive alternative for the diagnosis of COVID-19, especially in regions where PCR tests are scarce or time-consuming. Additionally, the use of neural networks may offer advantages over other forms of diagnosis, such as computed tomography (CT), as chest radiographs are more widely available and less costly. However, it is important to consider the limitations and challenges encountered in using neural networks for this purpose, such as the lack of specificity in mild or asymptomatic cases and the need for quality equipment and trained professionals to interpret the images. This study aims to contribute to the advancement of COVID-19 diagnosis through non-invasive and effective methods, as well as to identify possible limitations and challenges in using neural networks for this purpose.