Abstract
A system to improve mobility of blind people based on computer vision is presented. It consists of a portable PC endowed with a digital camera and a pair of auriculars. The system is able to detect and to track green and red traffic lights used by the pedestrians without this disability. Colour histogram analysis whereas structural data have been used to recognize these traffic symbols. In order to handle with illumination changes a simple adaptation scheme is proposed. By means of the auriculars, the system inform to the user about semaphore state and also about its pan position in real time. © Springer-Verlag 2004.
Cite
CITATION STYLE
Aranda, J., & Mares, P. (2004). Visual system to help blind people to cross the street. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 3118, 454–461. https://doi.org/10.1007/978-3-540-27817-7_67
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
