Vision artificielle pour les non-voyants : une approche bio-inspirée pour la reconnaissance de formes

Abstract : More than 315 million people worldwide suffer from visual impairments, with several studies suggesting that this number will double by 2030 due to the ageing of the population. To compensate for the loss of sight the current approaches consist of either specific aids designed to answer particular needs or generic systems such as neuroprostheses and sensory substitution devices. These holistic approaches, which try to restore vision as a whole, have been shown to be very inefficient in real life situations given the low resolution of output interfaces. To overcome these obstacles we propose the use of artificial vision in order to pre-process visual scenes and provide the user with relevant information. We have validated this approach through the development of a novel assistive device for the blind called Navig. Through shape recognition and spatialized sounds synthesis, this system allows users to locate and grab objects of interest. It also features navigational aids based on a new positioning method combining GPS, inertial sensors and the visual detection of geolocalized landmarks. To enhance the performance of the visual module we further developed, as part of this thesis, a bio-inspired pattern recognition algorithm which uses latency-based coding of visual information, oriented edge representations and a cascaded architecture combining detection at different resolutions.
Complete list of metadatas

https://tel.archives-ouvertes.fr/tel-01127709
Contributor : Adrien Brilhault <>
Submitted on : Saturday, March 7, 2015 - 8:31:42 PM
Last modification on : Thursday, June 27, 2019 - 4:27:48 PM
Long-term archiving on : Monday, June 8, 2015 - 5:32:05 PM

Licence


Distributed under a Creative Commons Attribution - NonCommercial - NoDerivatives 4.0 International License

Identifiers

  • HAL Id : tel-01127709, version 1

Collections

Citation

Adrien Brilhault. Vision artificielle pour les non-voyants : une approche bio-inspirée pour la reconnaissance de formes. Intelligence artificielle [cs.AI]. Université Toulouse III Paul Sabatier, 2014. Français. ⟨tel-01127709⟩

Share

Metrics

Record views

499

Files downloads

4440