Computer Vision Assists the Blind
Date: Tuesday, September 20 @ 04:08:10 BST
Topic: Applications

edestrian crossings and particularly the challenges these impose on the blind, are investigated at the Kyoto Institute of Technology. Canes, laser and ultrasonic waves have been used in the past to detect obstacles, but the new approach is somewhat more sophisticated. There is, as the article suggests, a cunning and principled approach, which helps detect the location, colour and orientation of various objects.
"Using images from a single camera, the device has a simple structure and does not need camera calibration, unlike sophisticated stereo camera systems, as the information is obtained using what is known as a "camera coordinate system". This means that separate images do not need to be taken to calibrate the device. The length of a pedestrian crossing is measured by projective geometry, where the camera makes an image of the white lines painted on the road. Using the properties of geometric shapes as seen in the image, the actual distances are determined."

Methods such as this may also be used for autonomous vehicles where understanding of a complex environment is essential.

Full article from

This article comes from The Computer Vision Digest

The URL for this story is: