Filtering and Temperature Correction Algorithms for Smog Interference in Temperature Measurement Based on CCD Image Sensor

The visible radiation would be absorbed and scattered by the smog diffusing around the radiator and on the path of the radiation, which distorts the results of temperature measurement by two-color thermometry based on CCD image sensor. In this paper, an approach for filtering smog interference of high-temperature radiator and correcting temperature is proposed. Firstly, by the image target recognition method based on color information of radiator, the smog noises around a high-temperature radiator are eliminated. And then, the region interfered by smog is segmented. Finally, the temperature of the region calculated by the two-color thermometry formula is corrected using the nonlinear correction function. The experimental results show that the method can effectively reduce the influence in measurement results by smog interference and increase the accuracy and practicability in temperature measurement based on CCD.