Kaiwei Wang
Zhejiang University, China
Title: 5 years of implementing light to help the visually impaired at the National Optical Instrument Engineering Technology Research Center
Biography
Biography: Kaiwei Wang
Abstract
According to the World Health Organization, 285 million people around the world are estimated to be visually impaired and 39 million of them are blind. It is rather challenging for visually impaired people (VIP) to navigate through obstacles and avoid various hazards such as water puddle and approaching vehicles in unknown environments. At the National Optical Instrument Engineering Technology Research Center, a number of revolutionary techniques and apparatus have been explored to address the above problems. An eff ective approach is studied to expand the detection of traversable area based on an RGB-D sensor, which is compatible with both indoor and outdoor environments. Th e depth image of is enhanced with IR image largescale matching and RGB image-guided fi ltering. A polarized technique is implemented in order to detect traversable areas and water hazards by adequately considering polarization eff ects. A real-time crosswalk detection algorithm, adaptive and consistency aextraction analysis (AECA), is proposed to detect and remind the position of crosswalks at urban intersections. Compared with existing algorithms, which detect crosswalks in ideal scenarios, the algorithm performs better in challenging scenarios, such as crosswalks at far distances, low contrast crosswalks, pedestrian occlusion, various illuminances and the limited resources of portable PCs. A real-time Pedestrian Crossing Lights (PCL) detection algorithm for the visually impaired is also proposed. Diff erent from previous works which utilize analytic image processing to detect the PCL in ideal scenarios, the proposed algorithm detects PCL using machine learning scheme in the challenging scenarios, where PCL have arbitrary sizes and locations in the acquired image and suff er from the shake and movement of the camera. Up to date, tens of thousands visually impaired individuals are benefi ting from these technologies.