ANDRE GOMES LAMAS OTERO

ANDRE GOMES LAMAS OTERO

Página do item completo

Resultados de Busca

Agora exibindo 1 - 8 de 8

A methodology for automated radioactive waste characterization
2023 - OTERO, ANDRE G.L.; MARUMO, JULIO T.; JUNIOR POTIENS, ADEMAR J.
Aplicação de redes neurais profundas na caracterização de rejeitos radioativos
2022 - OTERO, ANDRE G.L.
O desenvolvimento da tecnologia nuclear deve permitir a gestão segura dos rejeitos radioativos, provenientes das várias etapas do ciclo do combustível nuclear, da produção de radiofármacos e das aplicações de radioisótopos na medicina, indústria e centros de pesquisa. A caracterização destes rejeitos é uma tarefa complexa, devido à grande variedade de aplicações, materiais e composição. Neste trabalho foi desenvolvida uma metodologia de caracterização final de rejeitos radioativos utilizando redes neurais profundas. O método de Monte Carlo foi empregado para realizar a simulação de espectros gama, considerando o cenário de um tambor de rejeitos de 200 litros contendo até dez diferentes radionuclídeos: Am-241, Ba-133, Cd-109, Co-57, Co-60, Cs-137, Eu-152, Mn-54, Na-22, Pb-210. Os dados provenientes das simulações foram utilizados para treinar e avaliar o desempenho de diferentes arquiteturas de redes neurais profundas. A arquitetura selecionada foi VGG-19 a qual, após adaptações, apresentou o melhor desempenho na tarefa de classificação, sendo capaz de identificar quais radionuclídeos e qual a intensidade de cada radionuclídeos que compõe o espectro de radiação gama, emitido por um tambor de rejeito. Os resultados obtidos mostram que a metodologia desenvolvida pode atuar como uma importante ferramenta no processo de caracterização de rejeitos radioativos, realizada rotineiramente pelo Serviço de Gerência de Rejeitos Radioativos do IPEN, permitindo a diminuição à exposição ocupacional as radiações ionizantes.
A desktop application for automatic gamma spectroscopy analysis with deep learning
2022 - OTERO, A.G.L.; POTIENS, A.J.; MARUMO, J.T.
A desktop application for automatic gamma spectroscopy analysis with deep learning
2021 - OTERO, ANDRE G.L.; POTIENS JUNIOR, ADEMAR J.; LINO, JULIANA dos S.; MARUMO, JULIO T.
Comparing deep learning architectures on gamma-spectroscopy analysis for nuclear waste characterization
2021 - OTERO, A.G.L.; POTIENS JUNIOR, A.J.; MARUMO, J.T.
Neural networks, particularly deep neural networks, are used nowadays with great success in several tasks, such as image classification, image segmentation, translation, text to speech, speech to text, achieving super-human performance. In this study, the capabilities of deep learning are explored on a new field: gamma-spectroscopy analysis, comparing the classification performance of different deep neural network architectures. The following architectures where tested: VGG-16, VGG-19, Xception, ResNet, InceptionV3, and MobileNet, which are available through the Keras Deep Learning framework to identify several different radionuclides (Am-241, Ba133, Cd-109, Co-60, Cs-137, Eu-152, Mn-54, Na-24, and Pb-210). Using an HPGe detector to acquire several gamma spectra from different sealed sources to create a dataset used for the training and validation of the neural network's comparison. This study demonstrates the strengths and weaknesses of applying deep learning on gamma-spectroscopy analysis for nuclear waste characterization.
Um comparativo entre a utilização de redes neurais perceptron e redes neurais profundas na identificação de radionuclídeos em espectrometria gama
2020 - OTERO, A.G.L.; POTIENS JUNIOR, A.J.; MARUMO, J.T.
Apresentamos os resultados da comparação entre uma Rede Neural Profunda e uma Rede Neural Perceptron na classificação de espectros gama obtidos utilizando um detector de germânio hiper-puro. Utilizando dados de diversas fontes seladas (Am-241, Ba-133, Cd-109, Co-57, Co-60, Cs-137, Eu-152, Mn-54, Na-24, and Pb-210) foram gerados uma lista extensa de espectros para treino e validação contendo, respectivamente, 500 e 160 espectros, onde foram mesclados até três radionuclídeos em um único espectro. Depois de 250 épocas de treino foram validadas a exatidão de cada um dos modelos utilizando o conjunto de validação. O modelo de rede neural profunda obteve uma exatidão de classificação de 96,25% enquanto a rede neural perceptron obteve uma exatidão de 80,62%. Os resultados mostram um desempenho robusto e consistentemente melhor das redes neurais profundas, frente as redes neurais perceptron.
Applying deep-learning in gamma-spectroscopy for radionuclide identification
2019 - OTERO, ANDRE G.L.; MARUMO, JULIO T.; POTIENS JUNIOR, ADEMAR J.
Introduction Neural networks, particularly deep neural networks, are used nowadays with great success in several tasks, such as image classifi cation, image segmentation, translation, text to speech, speech to text, achieving super-human performance. In this study we explore the capabilities of deep learning on a new fi eld: gamma-spectroscopy analysis. Using a well-known deep neural network architecture with gamma spectroscopy data, we successfully identify the radionuclides (Am-241, Ba-133, Cd-109, Co-60, Cs-137, Eu-152, Mn- 54, Na-24 and Pb-210) contained in several experiments. This neural network is also capable to identify different mixed radionuclide in the same source, demonstrating that deep neural networks can be successfully applied on gamma-spectroscopy analysis. Methods Using a HPGe detector to acquire several gamma spectra, from different sealed sources, we created a dataset that was used for the training and validation of the neural network. We created our deep neural network using python as programing language, alongside with Keras, a deep learning framework. Applying the VGG19 network architecture, except by the last layer which using softmax as activation function, we used sigmoid in order to allow classifi cation of not mutually exclusive classes in the same instance. Results After 250 epochs of training the classifi cation error on the training and test datasets reached a minimum, the same occurred with accuracy. As a fi nal test we used a spectrum from a triple sealed source, containing Am-241, Cs-137 and Co-60. As this kind of data was never seen by the network before we expect that the network generalizes well and correctly classify the spectra as containing the three isotopes. When applying the new data, the model correctly classifi ed the spectra as containing the tree radionuclide. Conclusions The model successfully classifi es different spectra with different radionuclides and his performance is good on never seen before data (the triple source sealed) demonstrating that deep learning can be used on a new domain.
Comparing deep learning architectures on gamma-spectroscopy analysis for nuclear waste characterization
2019 - OTERO, ANDRE G.L.; POTIENS JUNIOR, ADEMAR J.; CALZETA, EDUARDO P.; MARUMO, JULIO T.
Neural networks, particularly deep neural networks, are used nowadays with great success in several tasks, such as image classification, image segmentation, translation, text to speech, speech to text, achieving super-human performance. In this study, we explore the capabilities of deep learning on a new field: gamma-spectroscopy analysis, comparing the classification performance of different deep neural networks architectures. We choose VGG-16, VGG-19, Xception, ResNet, InceptionV3 and MobileNet architectures which are available through the Keras Deep Learning framework to identify several different radionuclides (Am-241, Ba- 133, Cd-109, Co-60, Cs-137, Eu-152, Mn-54, Na-24, and Pb-210). Using an HPGe detector to acquire several gamma spectra, from different sealed sources to created a dataset that was used for the training and validation of the neural networks comparison. This study demonstrates the strengths and weakness of applying deep learning on gamma-spectroscopy analysis for nuclear waste characterization.