Research On Testing Technology Of Frequency Response Characteristics Of Thermal Detectors And Thermal Imaging System

The frequency response characteristic is a key indicator of the uncooled infrared detector based on the microbolometer and thermal imaging system,it is highly related to the effective frame rate of the detector.Therefore,it is necessary to measure frequency response characteristic for device design and application.Thermal detector is an infrared detector that converts heat signals into electrical signals.At present,in the performance test of thermal detectors and thermal imaging system,people are more concerned about static parameters such as noise equivalent temperature difference(NETD),and ignore dynamic parameters such as frequency response characteristics.By analyzing the relationship between NETD and frequency response characteristics,a comprehensive evaluation of the detector and the thermal imager is performed.Time constant is also called the response time,which represents the time required for the system to start from working to stable state.When the frequency of the input signal is higher than the response time,the detector cannot effectively process the input signal within one frame time.The next signal affects the distortion of the output signal.Therefore,by establishing the relationship between the time constant and the frequency response characteristics,the frequency response characteristics of the detector can be effectively studied.The response time can be divided into thermal response time and electrical response time.Photon detectors mainly consider electrical response time,and thermal detection mainly considers thermal response time.However,the typical design value or the measurement result based on mono-element is not easy to establish the quantitative relationship with the frequency-response characteristic of the detector.In this paper a method for measuring thermal time constant based on array devices was introduced.The method employed a chopper to obtain a chopping frequency that was less than 1/2 frame rate,after the fast Fourier transform calculation,fitting the frequency response curve by the effective voltage response signals at different frequency point,the thermal time constant could be quickly and effectively extracted.I order to mitigate the effects of spurious noise and harmonics on the final test results,a window function is introduced to alleviate the problem of signal spectrum leakage.Meanwhile,by calculating dead pixels to improve the accuracy of the final testBy the experiment proving and results analyzing,the method has the advantages of high precision accuracy,strong anti-interference ability,good stability and short test time.What's more,there is no special requirement for testing instruments or testing samples,and is worthy to be popularized.