Authors:
Björn Zeise
and
Bernardo Wagner
Affiliation:
Leibniz Universität Hannover, Germany
Keyword(s):
Infrared Thermography, 3D Temperature Mapping, Emissivity Estimation, Temperature Correction, Thermal Reflection Removal, Mobile Service Robotics.
Related
Ontology
Subjects/Areas/Topics:
Image Processing
;
Informatics in Control, Automation and Robotics
;
Perception and Awareness
;
Robotics and Automation
;
Vision, Recognition and Reconstruction
Abstract:
Many mobile robots nowadays use thermal imaging cameras (TICs) in order to enhance the environment model that is created during exploration tasks. In conventional thermography, thermal images always have to be carefully revised by human operators, which is not practicable in autonomous applications. Unknown surface emissivities are the main source of misinterpretations in thermal images. In this work, we present two methods dealing with these misinterpretations by exploiting the TIC’s changing point of view. While the first approach classifies the regarded material in order to estimate improved surface temperature values, the second one is capable of detecting and removing thermal reflections. The spatial relationship between the thermal images and the regarded surface is made by using a rigidly mounted sensor stack consisting of a TIC and a 3D laser range finder, whose extrinsic calibration is described. During evaluation, we demonstrate the
functionality of both approaches.