Shortwave Infrared Focal Plane Information Acquisition Technology Based On Lossless Exposure

The Short Wave infrared(SWIR)is an electromagnetic wave band with a wavelength of 0.9 μm to 2.5 μm and is widely present in various environments in nature.There are many sources of short-wave infrared,including reflection of environmental objects,spontaneous emission of high-temperature objects,and various artificial shortwave infrared light sources.According to Wien’s displacement law,objects with a temperature of around 1000K～3000K can spontaneously radiate shortwave infrared electromagnetic waves.And most of the night skylight is concentrated in the short-wave infrared region,so shortwave infrared is widely used.This degree topics based on “Mars Mineral Spectrum Analyzer” project.First,This paper introduces the development and status quo of short-wave infrared focal plane array(FPA)detectors,and then studies several key technologies of information acquisition systems including the short-wave infrared detector technology,the signal components and noise characteristics of the system,the influence of the integral time on the signal-to-noise ratio,etc.Then a set of electronic system for shortwave infrared focal plane information acquisition was designed,including hardware circuit and FPGA software program.Finally,the short-wave infrared focal plane detector,FPGA circuit board,and host computer were successfully used to build the information acquisition system.We know that the energy of the solar spectrum decreases with the increase of wavelength in the short-wave infrared region.At the same time,the atmospheric transmittance decreases with the increase of the wavelength in the short-wave infrared region,too.This phenomenon result that the energy FPA received decreases as the spectral dimension increases.In the end,these factors cause an inadequacy in the exposure mode of the short-wave infrared spectrometer: When the FPA integrate on the signal,if all the spectral dimension detectors in the entire FPA have the same integration time,the short-wavelength spectral dimension energy will be integrated into the saturation state earlier than long-wavelength spectral dimension.And since the signal-to-noise ratio becomes higher as the integration time increases,the signalto-noise ratio of the spectral dimension with long wavelength is far from reaching the limit.The signal-to-noise ratio is a very important parameter to measure the performance of the short-wave infrared imaging spectrometer.Therefore,in order to overcome this shortcoming,this paper proposes a method to improve the shortwave infrared spectrometer(which has no effect on ordinary imaging systems without spectrometers)called “lossless exposure” technology: that is,according to a specific application scenario,the short-wavelength infrared spectrum energy of the detected target has the same trend of the solar spectrum,then different spectral dimensions are given different integration times.It is expected that all spectral dimensions in the FPA will reach saturation at the same time to obtain the maximum signal-to-noise ratio of the spectral image.This topic also studied the simple image processing technology of infrared images,including blind element detection and compensation,non-uniformity correction,etc.The two-point correction method was used to correct short-wave infrared images.Finally,the noise and signal-to-noise ratio of the entire shortwave infrared focal plane information acquisition system were tested.And the lossless exposure experiment scheme was designed.The influence of the integration time on the SNR was verified through experiments,and then the effect of the lossless exposure technique on the SNR improvement of a large number of application scenarios was verified.From the experiment in Chapter5,we can see that when the integration time of the three different parts of the focal plane are 0.85 ms,3.31 ms,and5.76 ms,the SNR is 68.8% higher than the case when all integration time of focal plane are 0.85 ms.