Virtual Sensors Determined Through Machine Learning

Yumi Iwashita, Adrian Stoica, Kazuto Nakashima, Ryo Kurazume, Jim Torresen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

We propose a method that increases the capability of a conventional sensor, transforming it into an enhanced virtual sensor. This paper focuses on a virtual thermal Infrared Radiation (IR) sensor based on a conventional visual (RGB) sensor. The estimation of thermal IR images can enhance the ability of terrain classification, which is crucial for autonomous navigation of rovers. The estimate in IR from visual band has inherent limitations, as these are different bands, yet correlations between visual RGB and thermal IR images exist, as different terrains, which visually may appear different, also have different thermal inertia. This paper describes the developed deep learning-based algorithm that estimates thermal IR images from RGB images of terrains, providing the feasibility of the idea with average 1.21 error [degree Celsius].

Original languageEnglish
Title of host publication2018 World Automation Congress, WAC 2018
PublisherIEEE Computer Society
Pages318-321
Number of pages4
Volume2018-June
ISBN (Print)9781532377914
DOIs
Publication statusPublished - Aug 8 2018
Event2018 World Automation Congress, WAC 2018 - Stevenson, United States
Duration: Jun 3 2018Jun 6 2018

Publication series

NameWorld Automation Congress Proceedings
Volume2018-June
ISSN (Print)2154-4824
ISSN (Electronic)2154-4832

Conference

Conference2018 World Automation Congress, WAC 2018
CountryUnited States
CityStevenson
Period6/3/186/6/18

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering

Fingerprint Dive into the research topics of 'Virtual Sensors Determined Through Machine Learning'. Together they form a unique fingerprint.

Cite this