基于红外图像的飞机主体合金温度检测

Temperature detection of aircraft body alloy based on infrared image

为提高飞机主体合金温度检测精度,提出基于红外图像的检测方法。以38框BT20钛合金为样品,用瞬态气动热试验模拟系统采集图像,用改进的中值滤波算法预处理,去噪;选彩色种子区域生长法检测。结果表明:本方法可有效处理亮度过饱和,改善图像对比均衡性,增强检测质量;测温距离为7~8 m时,温度误差最低,为0.7℃;高温时检测精度最高为98%;最高去噪值为13.6 dB,高斯噪声去除好。

In order to improve the temperature detection accuracy of aircraft body alloy,a temperature detection method of aircraft body alloy was proposed based on infrared image. The 38 BT20 titanium alloy frames were selected as the test samples. The images were collected by the transient aerothermal experiment simulation system,and the infrared images of the samples were preprocessed by the improved straight filter algorithm to remove the noise interference. Color image seed region growth method was selected to detect high temperature region. Experimental results show that the proposed method can effectively deal with the problem of image brightness supersaturation,improve the balance of image contrast and enhance the quality of image detection.When the temperature measurement distance is 7-8 m,the error of temperature value is the lowest,only 0.7 ℃. And the highest detection accuracy reaches 98% at high temperature. The highest de-noising value is 13.6 dB which has a good ability to remove Gaussian noise.