Research On Reconstruction Method Of Two-dimensional Temperature Distribution Field Based On CCD Image

Temperature participation is one of the important parameter indicators in the fields of daily life,industrial,and defense.It accurately measures temperature for ensuring people’s lives,improving production and economic benefits.With the continuous advancement of science and technology,temperature parameters measuring from contact type to non-contact temperature measurement,temperature measurement technology has greatly developed from single-point measuring measurement of temperature field.However,due to the limitations of sensor techniques,conventional contact temperature mode is unable to measure the high temperature flame,and cannot react the distribution of the entire flame temperature field.In recent years,non-contact temperature measuring methods with CCD cameras and introducing radiant energy principles have been widely concerned.According to the flame image taken,this paper studies the calibration method of high temperature blackbody furnace and the measurement method of two-dimensional temperature field distribution of the flame image.The main tasks are as follows:1.This paper presents a high-temperature black body furnace temperature calibration method based on a random sample consensus.By eliminating the abnormal data point in the measurement,the method accurately calibrates the relationship between the image grayscale value and the measured temperature function of the high-temperature black body furnace.The RANSAC method can effectively reduce the effects of calibration errors on the distribution measurement of the temperature field.The experimental results show that this method can obtain better calibration accuracy than the least squares commonly used in the literature.2.An improved annealing algorithm is given to achieve the inversion calculation of the flame temperature field.The method can avoid the complication of traditional manual adjustment by introducing Bayesian optimizing parameters in an annealing algorithm.At the same time,Bayesian optimization utilizes the information calculated from the previous data point to improve the speed of solving the optimal value of parameters.The experimental results show that the improved method is better than the traditional simulated annealing algorithm.3.Designed and implemented a two-dimensional temperature field distribution measurement system based on a flame image.Its functions include preprocessing flame images,fitting the function relation between temperature and gray scale,and calculating the two-dimensional temperature distribution of flame.