Prototype Car UC3M: a system to detect pedestrians in low visibility
Ivvi new Scientific super car that can detect pedestrians at night time up to 40m. It is based on an infrared sensor and an algorithm analyzing infrared images and identifies pedestrians according to some ” congruent features the silhouette that do not vary with temperature .”
More and more vehicles are equipped with active technology to reduce the risk of collision with another vehicle or a pedestrian. They consist of an obstacle detection device associated with an emergency braking system.
To identify pedestrians , current technology uses a camera and pattern recognition software . Effective in daylight , it is more limited at night. A new developed at Carlos III University of Madrid using infrared camera system could be a solution for driving in poor visibility.
Suitable for night driving
The use of infrared thermal cameras , which record the heat emitted by the human body, can scan the environment of the vehicle in a radius of 40 meters, even at night.
“In this situation , the cameras that are currently included in some vehicles may only be used in the illuminated by the headlights of the car zones. But our system does not require any type of outdoor lighting ,” explains Daniel Olmeda , one of the participants the project.
” Ivvi ” The operation of the system has been described in an article published in the journal Integrated Computer -Aided Engineering. It is based on an infrared sensor and an algorithm that analyzes infrared images and identifies pedestrians according to some ” congruent features the silhouette that do not vary with temperature ” explains Daniel Olmeda .
The device can be easily integrated vehicle researchers say. They have themselves tested on a ivvi 2.0 (intelligent vehicle based on visual information ) prototype developed by the University of Madrid. This new generation of detectors may also have applications in the field of robotics, according to scientists.