Automatic generation of high performance morphological filters to fix missing data in depth images on real-time embedded systems for visually impaired people

Author(s)

Batista Dourado, Antonio Miguel ^[1] ; Pedrino, Emerson Carlos ^{[2, 3]}

Affiliation

^[1] Fed Inst Sao Paulo, Av Zelia Lima Rosa 100, BR-18550000 Boituva, SP - Brazil

^[2] Univ Fed Sao Carlos, Sao Carlos, SP - Brazil

^[3] Univ Fed Sao Carlos, Dept Comp, Rod Washington Luis Km 235, BR-13565905 Sao Carlos, SP - Brazil

• Document type: Journal article
• Source: PRZEGLAD ELEKTROTECHNICZNY; v. 96, n. 1, p. 113-117, 2020.
• Web of Science Citations: 0

Abstract

The paper presents an evolutionary multi-objective approach to automatically generate morphological filters to solve unknown distances areas, found in depth images used by real-time embedded systems for visually impaired people, and to prevent accidents. It was used Cartesian Genetic Programming as base for the NSGAII multi-objective optimization algorithm proposed to optimize two objectives: low error rates for quality x low complexity for speed. Results showed this approach was able to deliver feasible solutions with good quality and speed to be used in real-time systems. (AU)

• FAPESP's process: 17/26421-3 - Investigation of the use of Intelligent Systems for Efficient Mapping of Applications in Many-Core Architectures
• Grantee: Emerson Carlos Pedrino
• Support Opportunities: Scholarships abroad - Research