一种水雾干扰下红外测温误差补偿方法OA

Aiming at the influence of water mist in the imaging optical path of infrared thermal imager on the accuracy of infrared temperature measurement,an error compensation model based on the principle of infrared temperature measurement is constructed.Then,the texture features of infrared image affected by water mist are extracted by using the temperature co-occurrence matrix.On this basis,a hybrid kernel function is constructed for the extreme learning machine by combining the radial basis kernel function and the inverse cosine kernel function,and a regression model of water mist transmittance based on the hybrid-kernel extreme learning machine is established with the extracted image features as input.After calculating the transmissivity of water mist,the error compensation model is solved to realize error compensation.Finally,in order to verify the effectiveness of the proposed infrared temperature measurement error compensation method under the interference of water mist,the measurement results before and after the compensation method are compared in the blackbody furnace experiment.In the temperature range of 30℃to 500℃,the average temperature measurement error drops from 47.570℃to 4.287℃,and the maximum temperature measurement error decreases from 99.680℃to 11.730℃.The experimental results show that the proposed infrared temperature measurement error compensation method can improve the temperature measurement accuracy of the infrared thermal imager in the water mist interference environment.