Research On Performance Testing Technology Of Infrared Thermal Radiation Imaging System

Infrared thermal imager has been widely used in civil and military fields.The more sensitive and precise components of infrared thermal imager put forward higher requirements for the corresponding test technology.Infrared test technology must adapt to the development of infrared thermal imager,so it is particularly important to test the performance parameters of infrared thermal imager.This paper mainly focuses on the application of infrared thermal imager in the field of thermal imaging sight,and analyzes several performance parameters related to it,which mainly include: installation reference axis deviation,optical axis consistency,zero walking momentum.Through research,we will establish a performance parameter test system for dual-field infrared thermal imager with higher test accuracy and efficiency.This paper first studies and discusses several conventional testing methods and shortcomings of these performance parameters,and combines with the sub-pixel centroid recognition and positioning algorithm to improve the testing techniques of these three performance parameters.Regarding the traditional barycenter method of seeking centroid and its exponential deformation,when faced with complex video background noise and the differential image gain and non-uniformity of the thermal imager,the barycenter method has weak anti-interference ability and high accuracy.and the Gauss surface fitting method with high accuracy involves a large amount of calculation,which will affect the real-time positioning of the target image points.In this subject,we conduct continuous experiments on the sub-pixel centroid localization algorithm.In this subject,we conduct continuous experiments on the sub-pixel centroid positioning algorithm.In view of the special situation of processing the gray spot of the circular hole target image point,we combine the image adaptive threshold segmentation algorithm and the image contour tracking technology,and propose an algorithm with high repeat accuracy can stably and accurately locate the spot centroid.In this paper,the hardware platform is divided into three modules,including optical collimation module,infrared temperature difference target module and orientation calibration module.Specifically,the optical collimation module is researched from type selection,structural design and calibration technology respectively;the infrared temperature difference target module is researched from platinum thermal resistance temperature measurement technology,digital PID temperature control algorithm and target scheme design,and the equipment composition and parameter selection of orientation calibration module are introduced to meet the test needs.At the same time,we use the API provided by DALSA to complete image acquisition,display,etc.,and the OpenCV computer vision library is used to analyze and process the video data of the thermal imager,Finally,we use MFC interface integration framework to complete the software design of the performance test system of this subject.It has been experimentally verified that the maximum difference of the centroid positioning algorithm in this paper is 0.010 pixels horizontally and 0.011 pixels vertically.Converted to the maximum horizontal difference of 0.00136 mrad,the maximum vertical difference is 0.00150mrad;under the condition of not performing the shock test,the maximum difference between the test value and the accurate value of the zero-travel amount is 0.002mrad;the installation reference axis The uncertainty of the test value is 0.01629mrad;the test accuracy of the optical axis coincidence degree depends on the centroid algorithm.