Data used for detection and tracking of dynamic objects for visually impaired people

Autor(es)

Cordeiro, Natal Henrique ^{[1, 2]} ; Pedrino, Emerson Carlos ^[1]

Afiliação do(s) autor(es)

^[1] Univ Fed Sao Carlos, Sao Carlos, SP - Brazil

^[2] Fed Inst Sao Paulo, Sao Paulo - Brazil

• Tipo de documento: Artigo Científico
• Fonte: DATA IN BRIEF; v. 26, OCT 2019.
• Citações Web of Science: 0

Resumo

This paper presents in detail the methodology for the detection and tracking of dynamic objects from the article in press (A new methodology applied to dynamic object detection and tracking systems for visually impaired people {[}1]). In order to validate this methodology, four different architectures have been designed in this paper. These architectures have implemented the techniques of pattern recognition, optical flow, background subtraction and color tracking to enable comparison and to see which is the most appropriate in a given environment. In this paper we also present a method created to quantify the effectiveness of each architecture implemented. (C) 2019 The Author(s). Published by Elsevier Inc. (AU)

• Processo FAPESP: 17/26421-3 - Investigação do uso de Sistemas Inteligentes para Mapeamento Eficiente de Aplicações em Arquiteturas Many-Core
• Beneficiário: Emerson Carlos Pedrino
• Modalidade de apoio: Bolsas no Exterior - Pesquisa