Publication: Fusión sensorial para detección de adelantamientos de vehículos
Loading...
Identifiers
Publication date
2012-12
Defense date
2012-09-28
Authors
Advisors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Impact
Export
Abstract
En el presente proyecto, se pretende construir un sistema de fusión sensorial, que permita determinar cuándo un coche está siendo adelantado. Para ello, se utilizan dos algoritmos: el algoritmo de Lucas y Kanade de forma piramidal para determinar la cantidad, dirección y sentido del movimiento, y el algoritmo de Haar-like en cascada, para determinar la presencia o no de coche. Como herramientas, se dispone del IVVI 2.0, el vehículo inteligente basado en información visual, desarrollado por el Laboratorio de Sistemas Inteligentes de la Universidad Carlos III. Dicho vehículo está equipado con una cámara de vídeo en la parte delantera, que captura en tiempo real una panorámica de la vía por la que se circula, un ordenador en el maletero, que analiza la imágenes captadas por la cámara, y una pantalla en el salpicadero, que muestra las imágenes capturadas por la cámara y la información analizada por el ordenador. El objetivo del proyecto es reducir los accidentes de tráfico por salida de vía y colisión lateral, que causan la mayor parte de las muertes en carretera en España, además de proporcionar a las compañías automovilísticas un valor añadido a su producto, y servir de base en investigaciones futuras sobre sistemas inteligentes de ayuda a la conducción (ADAS). ______________________________________________________________________________________________________________________
In this project, a sensory fusion system is expected to be built. It should detect when a car is overtaking the test vehicle. This is done by using two algorithms, the pyramidal Lucas and Kanade algorithm, to determine the quantity and direction of the movement, and the Haar-like cascade algorithm, to determine the presence or not of a car. The fusion of the information of both algorithms is a key-point for provide a trustable and reliable road safety application. The project was developed in the frame of the project IVVI 2.0 (Intelligent Vehicle based on Visual Information), an intelligent vehicle developed by the Intelligent Systems Laboratory (LSI) of the University Carlos III of Madrid. This vehicle is equipped with different devices that intend to use the latest advances in computer vision and other sensing technologies to help and warn drivers in case of hazardous situations. Among the different devices available, the used for the present project was a stereo-camera mounted in the windshield, connected to one of the several computers located in the trunk of the vehicle. The information provided by the computer is displayed in a monitor situated in the dashboard. The objective of the project is to reduce the traffic accidents produced by road departures and side collisions, which cause the majority road deaths in Spain. Besides it can be used in further automotive applications and researches on intelligent systems for Advanced Driving Assistant System (ADAS).
In this project, a sensory fusion system is expected to be built. It should detect when a car is overtaking the test vehicle. This is done by using two algorithms, the pyramidal Lucas and Kanade algorithm, to determine the quantity and direction of the movement, and the Haar-like cascade algorithm, to determine the presence or not of a car. The fusion of the information of both algorithms is a key-point for provide a trustable and reliable road safety application. The project was developed in the frame of the project IVVI 2.0 (Intelligent Vehicle based on Visual Information), an intelligent vehicle developed by the Intelligent Systems Laboratory (LSI) of the University Carlos III of Madrid. This vehicle is equipped with different devices that intend to use the latest advances in computer vision and other sensing technologies to help and warn drivers in case of hazardous situations. Among the different devices available, the used for the present project was a stereo-camera mounted in the windshield, connected to one of the several computers located in the trunk of the vehicle. The information provided by the computer is displayed in a monitor situated in the dashboard. The objective of the project is to reduce the traffic accidents produced by road departures and side collisions, which cause the majority road deaths in Spain. Besides it can be used in further automotive applications and researches on intelligent systems for Advanced Driving Assistant System (ADAS).
Description
Keywords
Fusión sensorial, Vehículos inteligentes, Proceso de imágenes, Tecnología automovilística, Conducción