Implementación de un prototipo para la detección de signos de fatiga del conductor aplicando visión artificial en un vehículo liviano en la noche
In Ecuador, there have been multiple types of accidents, caused by side collisions, running over, crash and loss of track caused by symptoms of fatigue and / or drowsiness in drivers at night. Being these those that present a high degree of incidence in the present year (2017). The objective of this...
| Autor Principal: | España Tarira, Luis Gabriel |
|---|---|
| Otros Autores: | Oña Paredes, Edwin Moisés |
| Formato: | bachelorThesis |
| Idioma: | spa |
| Publicado: |
2018
|
| Materias: | |
| Acceso en línea: |
http://dspace.ups.edu.ec/handle/123456789/15181 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: |
In Ecuador, there have been multiple types of accidents, caused by side collisions, running over, crash and loss of track caused by symptoms of fatigue and / or drowsiness in drivers at night. Being these those that present a high degree of incidence in the present year (2017). The objective of this work is the implementation of a prototype for the detection of signs of fatigue of the driver applying artificial vision in a light vehicle at night. The work was carried out establishing in the first instance the study of the parameters that includes the artificial vision; and its way of operating when exposed to a person's face. In addition, it presents the Haar-Cascade method for the recognition of faces through the processing of images, and also other libraries used to meet the objectives set out in this work. The state of the art allowed to have a clear focus on the parameters to be analyzed in this work, as is the case of the behavior of the human eye when the person is presenting signs of fatigue; and the level of opening and closing of the eye of each human being. The latter is crucial to verify the status of the driver against signs of fatigue using as a medium a Raspberry Pi3 minicomputer with its respective night vision camera. The prototype obtained an 82.5% of correct operation by means of tests carried out to drivers in night conditions and after this a survey was conducted aimed at the test subjects. |
|---|