Temperature Measurement Using Liquid Crystals Based on Color Image Processing

Improvement of Color-to-Temperature Calibration Line by Neural Networks

Abstract

A quantitative thermal flow visualization using thermo-sensive liquid crystals has been very popular to analyze heat transfer phenomena. In this method, a color-to-temperature calibration line is needed to obtain the temperature distribution. However, it is difficult to obtain the exact caliblation line because of its strong non-linear characteristic. In this study, a new algorithm for the calibration of color to temperature using a neural network is presented to improve the temperature measurement range.