Anzeige
Erschienen in:
2012 | OriginalPaper | Buchkapitel
A Segmentation-Based Stereovision Approach for Assisting Visually Impaired People
verfasst von : Hao Tang, Zhigang Zhu
Erschienen in: Computers Helping People with Special Needs
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
An accurate 3D map, automatically generated in real-time from a camera-based stereovision system, is able to assist blind or visually impaired people to obtain correct perception and recognition of the surrounding objects and environment so that they can move safely. In this paper, a segmentation-based stereovision approach is proposed to rapidly obtain accurate 3D estimations of man-made scenes, both indoor and outdoor, with largely textureless areas and sharp depth changes. The new approach takes advantage of the fact that many man-made objects in an urban environment consist of planar surfaces. The final outcome of the system is not just an array of individual 3D points. Instead, the 3D model is built in a geometric representation of plane parameters, with geometric relations among different planar surfaces. Based on this 3D model, algorithms can be developed for traversable path planning, obstacle detection and object recognition for assisting the blind in urban navigation.