Listar por tema "Convolutional neural network"
Mostrando ítems 1-6 de 6
-
Comet assay image analysis using convolutional neural networks
(2018) [Tesinas de grado]The comet assay is a technique used to detect DNA lesions. It involves analyzing images from individual cells, which can later be sorted into categories according to damage level. In this work we propose performing the ... -
Detecção e classificação de sinalização vertical de trânsito em cenários complexos
(2017) [Tesis de maestría]A mobilidade é uma marca da nossa civilização. Tanto o transporte de carga quanto o de passageiros compartilham de uma enorme infra-estrutura de conexões operados com o apoio de um sofisticado sistema logístico. Simbiose ... -
Detection and classification of ultrasonic vocalizations from neonatal mice using machine learning
(2019) [Tesis de maestría]The study of animal behavior has fascinated scientists for hundreds of years. An important source of behavior information can go unnoticed to heedless ears, which is the emission of ultrasonic vocalization (USV) by certain ... -
Inteligência artificial aplicada aos exames de imagem odontológicos
(2021) [Tesinas de especialización]A presente revisão da literatura teve por objetivo investigar a literatura relacionada à aplicação da Inteligência Artificial (IA) na análise de exames de imagem nas diversas especialidades odontológicas, seu desempenho ... -
Investigating the use of approximate computing on a case-study neural network implemented Into FPGA by using HLS
(2022) [Tesinas de grado]Neural networks have been used for all types of applications, ranging from stock market predictions to image recognition. They can be trained and synthesized into FPGAs using engines or entirely in parallel. However, ... -
Most influential feature form for supervised learning in voltage sag source localization
(2024) [Artículo de periódico]The paper investigates the application of machine learning (ML) for voltage sag source localization (VSSL) in electrical power systems. To overcome feature-selection challenges for traditional ML methods and provide more ...
