Publicação

Driver monitoring system based on eye tracking

Ver documento

Detalhes bibliográficos
Resumo:Recent statistics indicate that driver drowsiness is one of the major causes of road accidents and deaths behind the wheel. This reveals the need of reliable systems capable of predict when drivers are in this state and warn them in order to avoid crashes with other vehicles or stationary objects. Therefore, the purpose of this dissertation is to develop a driver’s monitoring system based on eye tracking that will be able to detect driver’s drowsiness level and actuate accordingly. The alert to the driver may vary from a message on the cluster to a vibration on the seat. The proposed algorithm to estimate driver’s state only requires one variable: eyelid opening. Through this variable the algorithm computes several eye parameters used to decide if the driver is drowsy or not, namely: PERCLOS, blink frequency and blink duration. Eyelid opening is obtained over a software and hardware platform called SmartEye Pro. This eye tracking system uses infrared cameras and computer vision software to gather eye’s state information. Additionally, since this dissertation is part of the project "INNOVATIVE CAR HMI", from Bosch and University of Minho partnership, the driver monitoring system will be integrated in the Bosch DSM (Driver Simulator Mockup).
Autores principais:Cruz, Diogo Manuel Ferreira da
Ano:2016
País:Portugal
Tipo de documento:dissertação de mestrado
Tipo de acesso:acesso aberto
Instituição associada:Universidade do Minho
Idioma:inglês
Origem:RepositóriUM - Universidade do Minho
Descrição
Resumo:Recent statistics indicate that driver drowsiness is one of the major causes of road accidents and deaths behind the wheel. This reveals the need of reliable systems capable of predict when drivers are in this state and warn them in order to avoid crashes with other vehicles or stationary objects. Therefore, the purpose of this dissertation is to develop a driver’s monitoring system based on eye tracking that will be able to detect driver’s drowsiness level and actuate accordingly. The alert to the driver may vary from a message on the cluster to a vibration on the seat. The proposed algorithm to estimate driver’s state only requires one variable: eyelid opening. Through this variable the algorithm computes several eye parameters used to decide if the driver is drowsy or not, namely: PERCLOS, blink frequency and blink duration. Eyelid opening is obtained over a software and hardware platform called SmartEye Pro. This eye tracking system uses infrared cameras and computer vision software to gather eye’s state information. Additionally, since this dissertation is part of the project "INNOVATIVE CAR HMI", from Bosch and University of Minho partnership, the driver monitoring system will be integrated in the Bosch DSM (Driver Simulator Mockup).