Publication: Desarrollo de una aplicación Android para la detección de señales de tráfico
Loading...
Identifiers
Publication date
2015-06
Defense date
2015-07-06
Authors
Advisors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Impact
Export
Abstract
Los sistemas de visión por computador están ganando protagonismo en nuestra vida
diaria. Durante los últimos 30 años, han pasado de ser una mera utopía a convertirse en
una realidad. Encontramos estos sistemas en aplicaciones militares, aplicaciones de
seguridad para detección de movimiento o intrusos; aplicaciones industriales como
controles de calidad, aplicaciones para la gestión, administración y seguridad en medios
de transporte, etc.
Por otro lado, los teléfonos inteligentes o ‘smartphones’ han adquirido también un
papel fundamental en nuestras vidas, y el uso que les damos a estos dispositivos los ha
transformado en auténticos asistentes personales, simplificando nuestras vidas de manera
considerable.
Por tanto, era de esperar que ambos campos se uniesen y que los ‘smartphones’
comenzasen a ofrecer aplicaciones con funcionalidades de visión por computador o
realidad aumentada, es decir, utilizar la cámara del teléfono para obtener información de
nuestro alrededor.
En ese contexto de aplicaciones emergentes es donde se desarrolla este proyecto,
un campo aun poco explorado, con mucho potencial pero también con mucho trabajo por
delante.
Este trabajo se centra en la detección de señales de tráfico mediante el uso de la
cámara de los teléfonos inteligentes de la plataforma Android. Para ello, se hará uso de
los algoritmos implementados por las librerías OpenCV permitiendo localizar las áreas
de la imagen con mayor probabilidad de contener una señal de tráfico primero, y
verificando después si, efectivamente, el objeto detectado corresponde con una señal de
tráfico.
Finalmente, se analizarán otras posibles alternativas para la detección de señales de
tráfico no implementadas en este proyecto.
Computer Vision systems are gaining prominence in our daily lives. During the last 30 years, they have gone from being a mere utopia to become a reality. We find these systems in military applications, security applications for motion detection or intrusion; industrial applications such as quality control, management applications, management and security in transport, etc. On the other hand, smartphones have also acquired an important role in our lives, and the use we gave them has transformed the smartphones into our personal assistants, simplifying our lives in a great deal. Therefore, it was to be expected that both fields unify and smartphones began to deliver applications with computer vision capabilities or augmented reality, which is using the phone’s camera to get information about our surroundings. In this context of emerging applications it is where this project is developed, a field still little explored, with lot of potential but also a lot of work ahead. This paper focuses on the detection of road signs using the Android device’s camera. To do this, we will make use of the algorithms implemented by the OpenCV libraries allowing the application to locate the areas of the image which are most likely to contain a road signal first, and then, verifying if indeed the detected object corresponds to a road sign. Finally, other alternatives for the detection of road signs not implemented in this project will be as well analyzed.
Computer Vision systems are gaining prominence in our daily lives. During the last 30 years, they have gone from being a mere utopia to become a reality. We find these systems in military applications, security applications for motion detection or intrusion; industrial applications such as quality control, management applications, management and security in transport, etc. On the other hand, smartphones have also acquired an important role in our lives, and the use we gave them has transformed the smartphones into our personal assistants, simplifying our lives in a great deal. Therefore, it was to be expected that both fields unify and smartphones began to deliver applications with computer vision capabilities or augmented reality, which is using the phone’s camera to get information about our surroundings. In this context of emerging applications it is where this project is developed, a field still little explored, with lot of potential but also a lot of work ahead. This paper focuses on the detection of road signs using the Android device’s camera. To do this, we will make use of the algorithms implemented by the OpenCV libraries allowing the application to locate the areas of the image which are most likely to contain a road signal first, and then, verifying if indeed the detected object corresponds to a road sign. Finally, other alternatives for the detection of road signs not implemented in this project will be as well analyzed.
Description
Keywords
Visión por computador, Proceso de imágenes, OpenCV, Señales de tráfico, Segmentación, Android (Sistema operativo), Detección de imágenes