Colorimetric Temperature Measurement Algorithm Based on Multi-Sensor Data Fusion
Wavelength and other factors have a deep influence on temperature measurement system based on CCD image sensor. Deviations between measurement values sometimes are large. In order to reduce error, the temperature outputs in different wavelength combinations do have multi-sensor correlation properties in point of colorimetric temperature measurement system view. The colorimetric temperature measurement algorithm based on real-time adaptive weighted is put forward by such performance index that minimum of standard deviation. According to measured value of each sensor, we will find out the corresponding weights in adaptive manner. And discuss the statistical properties of the estimated standard deviation. Estimation unbiasedness is proved. The algorithm has the following advantages .A much smaller amount of calculation, without any a priori knowledge, only relying on the output of each sensor. All of these make temperature estimation be optimal, measurement accuracy and the real-time performance of the system are improved. Theoretical analysis and experimental results show that the colorimetric temperature measurement algorithm based on the real-time adaptive weighted improves obviously when comparing with the traditional means algorithm in measurement accuracy.