Reconhecimento de padrões de imagens de traços nucleares produzidos por partículas alfa em detectores de estado sólido

• Ano de defesa: 2010
• Autor(a) principal: Trindade, Ananda Coaglia
• Orientador(a): Cruvinel, Paulo Estevão
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus São Carlos
• Programa de Pós-Graduação: Programa de Pós-Graduação em Ciência da Computação - PPGCC
• Palavras-chave em Português: Processamento digital de imagens; Transformada de Hough; Redes Neurais Artificiais; Partículas Alfa; Radônio
• Palavras-chave em Inglês: Digital Image Processing; Hough Transform; Artificial Neural Networks; Alpha Particle; Radon
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/10707

This work presents a new technique for image analysis generated by alpha particles in solid state nuclear track detectors using Hough Transform to identify complete standards and Artificial Neural Networks which act as classifiers of incomplete standards. Radon, during their decay process emits alpha particles that may be registered in detector of certain materials as inorganic crystals, plastic or glass. Through these detectors is possible to obtain images generated by alpha particles and therefore calculate the concentration of radon in an environment. Radon and its decay products contribute to the majority of ionizing radiation received by worldwide population, particularly in indoor environments. Studies carried out over the years confirmed the association between radon exposure and lung cancer. Considering the importance of studies related to radon, the need arises to explore the existing techniques to obtain the concentration of radon indoors and create new techniques that are better adapted to current needs.