Spanish researchers have developed a pedestrian detection system based on infrared technology capable of seeing in the dark and through fog.
The driving-aid system relies on body-heat detection to determine the presence of people or animals in the field of vision of the infrared cameras. The system should not only be able to alert the driver but also, if necessary, to take immediate action to avoid an accident.
“With the model being used in our research, pedestrians up to 40 meters away can be detected, although this distance could be extended if we substitute the lens with one that has greater focus range,” explained Daniel Olmeda, a researcher at the Intelligent Systems Laboratory of the University of Carlos III in Madrid.
Using advanced image pattern recognition techniques, the system, described in a recent issue of the Integrated Computer-Aided Engineering journal, would be particularly helpful in situations with limited visibility.
“In this situation, sensitive cameras using the visible spectrum, which are already incorporated into some vehicles, can only be utilised in regions illuminated by the car’s headlights,” said Olmeda. “Our system does not require any type of external lighting.”
Cars equipped with pedestrian detection systems based on visible spectrum already exist and the researchers believe integrating their device into a broader system would be fairly simple.
The team has used the so called non-refrigerated microbolometer – an infrared sensor believed to be cheaper than other similar equipment.
“Generalised implementation of this type of sensors is viable and its mass production would lower production costs,” Olmeda said.
The system was tested using the university’s research vehicle, together with other existing artificial vision systems. The team believes such a technology could also find use in robotics.