Advanced search
Start date
Betweenand


A Data Fusion Architecture Proposal for Visually Impaired People

Full text
Author(s):
Cordeiro, Natal Henrique ; Batista Dourado, Antonio Miguel ; Quirino, Gustavo da Silva ; Pedrino, Emerson Carlos ; IEEE
Total Authors: 5
Document type: Journal article
Source: 2016 29TH SIBGRAPI CONFERENCE ON GRAPHICS, PATTERNS AND IMAGES (SIBGRAPI); v. N/A, p. 8-pg., 2016-01-01.
Abstract

Based on the difficulties that a Visually Impaired Person (VIP) has in understanding certain contexts, this article proposes a new architecture that aims to contextualize elements of that environment and, thus, help the visually disabled person to move around safely. The main objective of this project is to extract environmental features and to subsequently perform the processes of perception, comprehension and projection of the Situation Awareness (SAW) model. The proposed architecture is composed of Computer Vision techniques (CV) and Data Fusion (DF). The CV techniques were used in order to obtain the necessary characteristics for the DF to perform the search for consistent relationships for pattern recognition and allow for the projection of possible collisions of the VIP with environmental elements. For the collision projection, the specific characteristics of each context were extracted and trained, thus providing the VIP with inferences made in real time. (AU)

FAPESP's process: 15/23297-4 - System for automatic hardware generation in FPGAs by Cartesian Genetic Programming
Grantee:Emerson Carlos Pedrino
Support Opportunities: Regular Research Grants