Design And Test Of CCD Imaging System For Hyperspectral Total Ozone Unit

Ozone is an important trace gas in the atmosphere.Stratospheric ozone can absorb and weaken the ultraviolet and visible sunlight,protect life on earth from being exposed to the sun.Ozone in troposphere is an important greenhouse gas.In the past decades,stratospheric ozone hole and tropospheric ozone pollution have attracted great attention.At the same time,the spaceborne atmosphere detection technology has got further development and application with its all-weather,global coverage observation.Until now,more than ten atmospheric ozone detection devices have been lunched to orbit at home and abroad,and obtained the change information of ozone including total ozone,vertical distribution,providing valuable information for the study of global climate change and control of air pollution.The FY-3 Total Ozone Unit developed and launched in 2008 by China sucessfully obtained the global distribution of total ozone through measuring the atmospheric ultraviolet backscattering to the sun radiation,marking a new stage of China’s spaceborne atmospheric ozone detection.On this basis,China is also developing a new generation of hyperspectral total ozone unit,which takes the frame-transfer array detector CCD55-30 as the detector,develops an imaging system,tests and verifies its performance parameters.To this end,the research work and innovation points of this paper include the following aspects:(1)Debugging the hardware circuit of CCD55-30 imaging system,including timing driven conversion circuit,pretreatment and analog front-end circuit,bias voltage circuit,and FPGA control circuit,providing hardware conditions for the normal operation of CCD.Testing the functions of FPGA including CCD driving timing generation,correlated double sampling of signals,caching and transmission of gray value.The ground detection system of CCD imaging is established to test the CCD imaging function and verify the correctness of the working state of CCD55-30 imaging electronics system.(2)Establishing the signal model and noise model of CCD55-30 imaging system,and the imaging process parameters are combed,the meaning of CCD imaging performance parameters is summarized.The CCD test system is studied,including the performance parameter 1288 test system of photoelectric imaging device and the simple CCD test system built in the laboratory,their working principle,structure and composition,stability and uniformity of the light source are analyzed in detail,and their advantages and disadvantages are compared.(3)Using the method of adjusting the exposure time and optical power density,the linearity of CCD imaging system is tested respectively,the results show that the imaging system is linear,adjusting the exposure time is the best method of linearity test.Taking the gray value of the signal corresponding to the minimum exposure point as the lower limit of the detectable signal,the dynamic range of CCD is calculated with the test data of different exposure time,which better reflects the CCD’s detection ability to weak signal.The effect of reducing the readout noise on improving the dynamic range is analyzed,and the effect of reducing the CCD substrate voltage on improving the dynamic range is proposed and verified by experiments.The standard optical power meter is used to obtain the input optical power density,measure and calculate the quantum efficiency of CCD at different wavelengths.The temperature characteristics of the imaging system are tested,and the results show that cooling can effectively reduce the dark current of CCD and improve the image uniformity in dark field.(4)Analyzing the source of inhomogeneity of CCD55-30 gray value image,and the weakness of using DSNU(e-/pixel)and PRNU(%)to represent the pixel inhomogeneity.Putting forward the concept of DCNU(e-/pixel/s)and PCNU(%)innovatively,and designing the method that sets multiple exposure time,multi-frame images are collected under dark field(or light field)at each exposure time,then obtains dark current(or "dark current and photocurrent")by fitting,calculates DCNU and PCNU on the basis of dark current(or "dark current and photocurrent").The inhomogeneity of pixels obtained by the new and old methods is compared and analyzed.The results show that the test data obtained by the new method is more universal and convincing.(5)The signal-to-noise ratio of CCD imaging system is studied theoretically and tested experimentally.According to different proceduress of the imaging process,the theoretical curve of SNR is drawn,and then it is verified by experiment.The Dummy Output channel without dark signal is used innovatively to test the readout noise of the imaging system,which overcomes the shortcoming of traditional test method which is difficult to eliminate the dark current noise.The factors affecting SNR are analyzed,and the experiment verifies that pixel binning,superposition of multi-frame images and improving AD9814 gain can effectively improve the SNR of the imaging system,the driving timing of 2-line pixel binning and the programmable spectral channel gain is designed innovatively.