Purchasing new hardware? Read our latest product comparisons
ADVERTISEMENT

Honda Develops Intelligent Night Vision for automobiles

By

January 20, 2004

but with the system, the infrared technology clearly shows humans

but with the system, the infrared technology clearly shows humans

Image Gallery (2 images)

Honda has announced an Intelligent Night Vision System for automobiles, which uses "far infrared" cameras to detect pedestrians in or approaching the vehicle's path. The system provides the driver with visual and audio cautions to help prevent accidents involving pedestrians. The new system will be available on the Honda Legend in Japan later this year.

The Intelligent Night Vision System uses images obtained from two far infrared cameras positioned in the lower section of the front bumper to detect the position and movement of infrared heat-emitting objects and determine whether they are in or approaching the vehicle's path.

Based on size and shape, the system also determines if the detected object is a pedestrian. In addition to the conventional night vision function of giving the driver an enhanced view of the road ahead, the system is the world's first to provide cautions that inform the driver of the presence of pedestrians that are on the road or about to cross the vehicle's path.

Pedestrian fatalities make up approximately 30 per cent of all traffic accident fatalities in Japan.

In its ongoing efforts to reach a better understanding of the kinematics of pedestrian accidents and to develop technologies to protect pedestrians, Honda created the world's first pedestrian dummy, POLAR I, in 1988.

This was followed by POLAR II in 2000, which features an even more human-like structure and more points of measurement. Honda places a high priority on the development of new pedestrian safety technologies.

Given that 70 per cent of all pedestrian fatalities are said to occur at night, the independently developed Intelligent Night Vision System addresses a vital problem. Honda is committed to the further development of technologies designed to help prevent accidents involving pedestrians.

Far infrared camera

The camera obtains a visual image based on the "far infrared" radiation emitted by humans and other objects. Because it uses far infrared radiation, it is capable of obtaining a viable image without the use of a light source, as is required by visible-light or "near infrared" cameras.

Heads-up display

The image is reflected in a mirror positioned on top of the dashboard, which retracts into the dashboard for daytime stowage.

Pedestrian detecting ECU

The ECU determines pedestrian position and motion based on the image from the cameras, along with vehicle speed and other vehicle information. The system detects pedestrians in or approaching the vehicle's path, and provides caution to the driver via a visual enhancement frame around the pedestrian image and an audio caution.

Sensors

Headlight on/off information, windshield wiper setting, yaw rate, vehicle speed, ambient temperature.

ADVERTISEMENT
About the Author
Mike Hanlon Mike grew up thinking he would become a mathematician, accidentally started motorcycle racing, got a job writing road tests for a motorcycle magazine while at university, and became a writer. As a travelling photojournalist during his early career, his work was published in a dozen languages across 20+ countries. He went on to edit or manage over 50 print publications, with target audiences ranging from pensioners to plumbers, many different sports, many car and motorcycle magazines, with many more in the fields of communication - narrow subject magazines on topics such as advertising, marketing, visual communications, design, presentation and direct marketing. Then came the internet and Mike managed internet projects for Australia's largest multimedia company, Telstra.com.au (Australia's largest Telco), Seek.com.au (Australia's largest employment site), top100.com.au, hitwise.com, and a dozen other internet start-ups before founding Gizmag in 2002. Now he writes and thinks. All articles by Mike Hanlon
Tags
Post a Comment

Login with your Gizmag account:

Related Articles
Looking for something? Search our articles
ADVERTISEMENT
ADVERTISEMENT