Color-based image processing to measure local temperature distributions by wide-band liquid crystal thermography

This study presents a color-image-processing procedure for non-intrusive local temperature measurements by thermochromic liquid crystals (TLCs). The image evaluation software is completely independent of the color detection and acquisition hardware. This allows to use a wide variety of hardware solutions. An easy reproducible calibration of camera and light source is presented. The dependence of the detected hue values on intensity is investigated and further the hue versus temperature relation is studied. Sprayable TLC formulations and TLC-coated polyester sheets are studied and compared with regard to their signal-to-noise ratio and the dependence of their hue values on illumination and viewing angle. Furthermore, a method to investigate the hue resolution is presented. The relation between the resolution of hue values and the illumination intensity and its influence on signal noise is discussed for the first time for TLC applications. Different techniques of signal noise reduction an implemented in the image processing system. Their effects on the signal noise level are discussed. As an example the two dimensional temperature distribution caused by wing-type vortex generators in a channel how is given.