A method for processing data of SMILE satellite ultraviolet aurora imager

Abstract

The present invention discloses a method for processing data of a SMILE satellite ultraviolet aurora imager, including starting software, setting an operation mode, setting data monitoring, data processing, data display and result viewing. The present invention uses image processing software for SMILE satellite UVI observation data, including image background removal, image display, image geometry correction, image radiation correction, image projection, airglow background removal, aurora boundary recognition and other functions, to finally generate image data for scientific research.

Description

Technical Field

[0001] The invention relates to the technical field of ultraviolet aurora imaging data processing, and in particular to a method for processing data of an ultraviolet aurora imager of a SMILE satellite. Background Art

[0002] The SMILE satellite operates in a highly elliptical orbit around the Earth's polar region. The SMILE satellite carries a far-ultraviolet aurora camera (UVI) with an operating band of 160-180nm, which is used to obtain auroral images of energy deposition in the solar wind-magnetosphere coupling system. The main function of UVI is to image the entire auroral ellipse belt at a characteristic wavelength that overcomes sunlight pollution, with a spatial resolution of ~100km or better. The circular field of view of the UVI optical system is 9.9°, and when the SMILE satellite is more than 50,000km away from the center of the earth, it can cover the entire polar region. UVI is sensitive to the long-wavelength part of the N2 Lyman-Birge-Hopfield (LBH) band of 160-180nm. Multilayer coatings are applied on the surface of the reflector to achieve the required spectral response and effectively suppress out-of-band stray light. The detector is an intensified CCD. Photons emitted from the fluorescent screen are coupled to the CCD through a relay lens. By default, the CCD reads one frame (512×512) of image every two seconds. Then 30 frames are stacked in one minute to generate a UVI image. The stacking can be done in orbit or on the ground to compensate for satellite jitter. This system is used to receive, process and analyze the raw aurora data obtained by the UVI device. By adopting professional algorithms, it can generate quick-view images and level 1-3 data about the aurora, providing accurate and reliable data support for scientific research.

[0003] The present invention provides a SMILE satellite ultraviolet aurora imager data processing method for SMILE satellite UVI observation data. Summary of the invention

[0004] The present invention provides a SMILE satellite ultraviolet aurora imager data processing method for SMILE satellite UVI observation data.

[0005] The object of the present invention is to provide a method for processing data of a SMILE satellite ultraviolet aurora imager, comprising an image display area and a function button area, the method comprising the following steps:

[0006] Step 1: Start the software, open the UV1 data processing software, and enter the main interface;

[0007] Step 2: Set the operation mode. Click "Auto" to set it to automatic, and click "Manual" to set it to manual;

[0008] Step 3: Set the relevant file paths: Use the "Task List Path" button to select the UVI data distribution list folder that needs to be monitored; use the "L0 Level Data Path" button to select the top-level folder that stores UVI L0d data and auxiliary data; use the "Data Output Path" button to select the top-level folder that stores all levels of data products produced by UVI;

[0009] Step 4: Select automatic mode or manual processing mode according to your needs and start the processing.

[0010] Step 5. Users can view the generated Aurora QL, L1-L3 level data in the corresponding folder;

[0011] Step 6. After the task is completed, click the "Stop" button to stop monitoring the task path and exit the software.

[0012] Furthermore, the specific processing steps of the automatic mode in step 1 are as follows:

[0013] Step 1. Click the "Auto" button to set it to automatic mode;

[0014] Step 2. For the automatic mode, after completing the above settings, click the "Start" button, and the software will start monitoring the folder selected by the "Task List Path" button, automatically identify the newly added data processing task list file *.json in the folder, and automatically grab the file for full-process automatic processing. The processing starts from L0 data by default, including reading the original aurora data in the file, generating aurora quick-view images, correcting the original data and generating L1 data, performing physical dimension conversion, pixel-by-pixel projection positioning, and coordinate system conversion based on the L1 data to generate L2 data products;

[0015] Step 3: After the data processing task is completed, if there is a new task, repeat the above process. If there is no new task, the software is in monitoring and waiting state. The user can also click the "Stop" button to stop monitoring and exit the program.

[0016] Furthermore, the specific processing steps of the manual mode in step 1 are as follows:

[0017] Step 1. Click the "Manual" button to set it to manual mode;

[0018] Step 2: For manual mode, the user must first send a data processing request to the SMILE satellite science data processing subsystem integrated data processing software as needed. The data level that can be requested is L0, L1, or L2.

[0019] Step 3. Set the time period for requesting data. Click the "Start Time" and "End Time" buttons to set the start and end time. After clicking the "Start Time" button, the start time selection interface will pop up. The user can set the time by clicking the corresponding drop-down menu or button. After setting, click the "OK" button. The selected start time will be displayed in the text box after "Start Time". After clicking the "End Time" button, the end time selection interface will pop up. Similar to the start time selection interface, the user can set the time by clicking the corresponding drop-down menu or button. After setting, click the "OK" button. The selected start time will be displayed in the text box after "End Time".

[0020] Step 4. Select the data level and version you need to request. Select the drop-down menus behind "Data Level" and "Data Version". The default values ​​are L0 and V01 respectively.

[0021] Step 5. Click the "Request Science Data" button to send the data processing task list file to the SMILE satellite science data processing subsystem integrated data processing software, and then click the "Start" button to start monitoring the UVI data distribution list folder to see if there is a corresponding data list file sent by the system;

[0022] Step 6: Wait for the SMILE satellite science data processing subsystem to integrate the data processing software to respond to the list file, generate the data requested by the user based on it, and send the data processing task file to the software distribution list folder;

[0023] Step 7. After the file monitoring module responds, a pop-up window will display the status: "Request for scientific data completed!", and then the user can click the relevant button in the lower right corner to process the data;

[0024] Step 8. If the data requested by the user is of level L0, the user should click the "Process L0 data" button to process the L0 data into L2 data. The functions of this button include L0 quick view generation, L1 data processing and L2 data processing;

[0025] Step 9. If the data requested by the user is of level L1, the user should click the "Process L1 data" button to process the L1 data into L2 data. The functions included in this button include physical dimension conversion, pixel-by-pixel projection positioning, coordinate system conversion, etc. based on the L1 data to generate L2 data products;

[0026] Step 10. If the data requested by the user is of level L2, the user should click the "Process L2 Data" button to process the L2 data into L3 data. The functions of this button include airglow background removal, aurora boundary identification, and conjugate crossing analysis to generate L3 data products.

[0027] Step 11: After the data processing task is completed, if there is a new task, repeat the above process; if there is no new task, the user clicks the "Stop" button to stop monitoring and exit the program.

[0028] Furthermore, the specific steps of L1 data processing in step eight of the manual processing mode are as follows:

[0029] Step 1: Read in L0 level scientific data, auxiliary data and engineering telemetry data;

[0030] Step 2: Correct the received L0 aurora image data, including dark current correction, image bright spot and dark spot removal, image plane non-uniformity correction, and image plane geometric distortion correction to generate L1 level data products.

[0031] Furthermore, the specific steps of L2 data processing in step nine of the manual processing mode are as follows:

[0032] Step 1: Read in L1 scientific data, auxiliary data and engineering telemetry data;

[0033] Step 2: L2 data processing: Perform physical dimension conversion, pixel-by-pixel projection positioning, and coordinate system conversion on the received L1 aurora image data to generate L2 data products.

[0034] Furthermore, the specific steps of L3 data processing in step 10 of the manual processing mode are as follows:

[0035] Step 1: Read in L2 scientific data, auxiliary data, and engineering telemetry data;

[0036] Step 2: Perform airglow background removal, aurora boundary identification and conjugate crossing analysis on the received L2 aurora image data to generate L3 data products.

[0037] The present invention has the following advantages: the present invention uses image processing software for SMILE satellite UVI observation data, including image background removal, image display, image geometry correction, image radiation correction, image projection, airglow background removal, aurora boundary recognition, conjugate crossing analysis and other functions, and finally generates image data for scientific research. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 is a flowchart of the process of the present invention;

[0039] Figure 2 It is a data processing interface diagram of the present invention;

[0040] Figure 3 It is a flow chart of L1 data processing of the present invention;

[0041] Figure 4 is a flow chart of L2 data processing of the present invention;

[0042] Figure 5 This is a flow chart of L3 data processing of the present invention.

[0043] Figure 6 The scientific data matrix format diagram of the UVI camera of the present invention;

[0044] Figure 7 This is a schematic diagram of the optical projection principle of the present invention; DETAILED DESCRIPTION

[0045] The present invention provides a method for processing data of a SMILE satellite ultraviolet aurora imager, including an image display area and a function button area. The software operation process is as follows: Figure 1 , the method comprises the following steps:

[0046] Step 1: Start the software, open the UV1 data processing software, and enter the main interface. Figure 2 As shown;

[0047] Step 2: Set the operation mode. Click "Auto" to set it to automatic, and click "Manual" to set it to manual;

[0048] Step 3. Set the relevant file paths: Use the "Task List Path" button to select the UVI data distribution list folder that needs to be monitored; use the "L0 Level Data Path" button to select the top-level folder that stores UVI's L0 level data and auxiliary data; use the "Data Output Path" button to select the top-level folder that stores all levels of data products produced by UVI;

[0049] Step 4: Select automatic mode or manual processing mode according to your needs and start the processing;

[0050] Step 5. Users can view the generated Aurora QL, L1-L3 level data in the corresponding folder;

[0051] Step 6. After the task is completed, click the "Stop" button to stop monitoring the task path and exit the software.

[0052] In this embodiment, the specific processing steps of the automatic mode in step 1 are as follows:

[0053] Step 1. Click the "Auto" button to set it to automatic mode;

[0054] Step 2. For the automatic mode, after completing the above settings, click the "Start" button, and the software will start monitoring the folder selected by the "Task List Path" button, automatically identify the newly added data processing task list file *.json in the folder, and automatically grab the file for full-process automatic processing. The processing starts from L0 data by default, including reading the original aurora data in the file, generating aurora quick-view images, correcting the original data and generating L1 data, performing physical dimension conversion, pixel-by-pixel projection positioning, and coordinate system conversion based on the L1 data to generate L2 data products;

[0055] Step 3: After the data processing task is completed, if there is a new task, repeat the above process. If there is no new task, the software is in monitoring and waiting state. The user can also click the "Stop" button to stop monitoring and exit the program.

[0056] In this embodiment, the specific processing steps of the manual mode in step 1 are as follows:

[0057] Step 1. Click the "Manual" button to set it to manual mode;

[0058] Step 2: For manual mode, the user must first send a data processing request to the SMILE satellite science data processing subsystem integrated data processing software as needed. The data level that can be requested is L0, L1, or L2.

[0059] Step 3. Set the time period for requesting data. Click the "Start Time" and "End Time" buttons to set the start and end time. After clicking the "Start Time" button, the start time selection interface will pop up. The user can set the time by clicking the corresponding drop-down menu or button. After setting, click the "OK" button. The selected start time will be displayed in the text box after "Start Time". After clicking the "End Time" button, the end time selection interface will pop up. Similar to the start time selection interface, the user can set the time by clicking the corresponding drop-down menu or button. After setting, click the "OK" button. The selected start time will be displayed in the text box after "End Time".

[0060] Step 4. Select the data level and version you need to request. Select the drop-down menus behind "Data Level" and "Data Version". The default values ​​are L0 and V01 respectively.

[0061] Step 5. Click the "Request Science Data" button to send the data processing task list file to the SMILE satellite science data processing subsystem integrated data processing software, and then click the "Start" button to start monitoring the UVI data distribution list folder to see if there is a corresponding data list file sent by the system;

[0062] Step 6: Wait for the SMILE satellite science data processing subsystem to integrate the data processing software to respond to the list file, generate the data requested by the user based on it, and send the data processing task file to the software distribution list folder;

[0063] Step 7. After the file monitoring module responds, a pop-up window will display the status: "Request for scientific data completed!", and then the user can click the relevant button in the lower right corner to process the data;

[0064] Step 8. If the data requested by the user is of level L0, the user should click the "Process L0 data" button to process the L0 data into L2 data. The functions of this button include L0 quick view generation, L1 data processing and L2 data processing;

[0065] Step 9. If the data requested by the user is of level L1, the user should click the "Process L1 Data" button to process the L1 data into L2 data. The functions included in this button include physical dimension conversion, pixel-by-pixel projection positioning, coordinate system conversion, etc. based on L1 data to generate L2 data products.

[0066] Step 10. If the data requested by the user is of level L2, the user should click the "Process L2 Data" button to process the L2 data into L3 data. The functions of this button include airglow background removal, aurora boundary identification, and conjugate crossing analysis to generate L3 data products.

[0067] Step 11: After the data processing task is completed, if there is a new task, repeat the above process; if there is no new task, the user clicks the "Stop" button to stop monitoring and exit the program.

[0068] In this embodiment, the specific steps of L1 data processing in step eight of the manual processing mode are as follows:

[0069] Step 1: Read in L0 level scientific data, auxiliary data and engineering telemetry data;

[0070] Step 2: Correct the received L0 aurora image data, including dark current correction, image bright spot and dark spot removal, image plane non-uniformity correction, image plane geometric distortion correction, and generate L1 level data products, such as Figure 3 shown.

[0071] In this embodiment, the specific steps of L2 data processing in step eight of the manual processing mode are as follows:

[0072] Step 1: Read in L1 scientific data, auxiliary data and engineering telemetry data;

[0073] Step 2: L2 data processing: Perform physical dimension conversion, pixel-by-pixel projection positioning, and coordinate system conversion on the received L1 aurora image data to generate L2 data products, such as Figure 4shown.

[0074] In this embodiment, the specific steps of L3 data processing in step 10 of the manual processing mode are as follows:

[0075] Step 1: Read in L2 scientific data, auxiliary data, and engineering telemetry data;

[0076] Step 2: Perform airglow background removal, aurora boundary identification, and conjugate crossing analysis on the received L2 aurora image data to generate L3 data products, such as Figure 5 shown.

[0077] In this embodiment, the CCD dark current correction algorithm: The scientific data arrangement format of the UVI camera is as follows: Figure 6 As shown. The data matrix is ​​512 rows × 514 columns, where the 1st and 514th columns store the CCD dark current values. The dark current in the 1st column represents the dark current value of the scientific data in the 2nd to 257th columns, and the dark current in the 514th column represents the dark current value of the scientific data in the 258th to 513th columns. Let the L0 level scientific data matrix be I0[512×514], and the dark current corrected scientific data matrix be I 11 [512×512]. Dark current value I in column 1 DCleft , the dark current value I in column 514 DCright , then I 11 For: I 11 [1:512,j]=I0[1:512,j]-I DCleft (j=2,3,…,257), I 11 [1:512,j]=I0[1:512,j]-I DCright (j=258,259,…,513).

[0078] In this embodiment, the algorithm for removing bright and dark spots from an image is as follows: due to physical defects and other reasons, bright and dark spots may exist in an enhanced CCD image, which are removed by a two-dimensional median filtering algorithm. A 20x20 pixel window is selected for filtering, and a mirror complement method is used for the image edge. In each 20x20 pixel window, the median of the 400 pixel values ​​is calculated and replaced with the pixel value at the center of the window. After filtering is completed, the difference between the filtered image and the original image is calculated. The upper and lower thresholds are both selected as 50, that is, the pixel values ​​in the original image that are higher than 50 in the filtered image are replaced by the corresponding pixel values ​​in the filtered image, and the pixel values ​​in the original image that are lower than 50 in the filtered image are replaced by the corresponding pixel values ​​in the filtered image, thereby completing the removal of bright and dark spots. Image I 11 The image after removing bright spots and dark spots is I 12 .

[0079] In this embodiment, the geometric distortion correction algorithm: Ideally, a standard square object appears as a standard square image on the image plane. However, due to the presence of optical system aberrations and detector distortion, the image plane is no longer a standard square. The deviation between the actual image and the standard square image is called geometric distortion. In order to achieve accurate image projection positioning, it is necessary to eliminate the position error caused by distortion. For coaxial optical systems, a polynomial can generally be used to more accurately characterize the distribution of distortion on the entire image plane, and then correct it. In the laboratory detection and calibration process, a two-dimensional N-order polynomial can be established by measuring the image of the standard grid point array,

[0080] In this embodiment, N is 4. Let the distorted image I 12 The coordinates of (x d ,y d ), undistorted image I 13 The coordinates are (x, y), and the specific correction formula is: Among them, P i,j and Q i,j are polynomial coefficients. In image I 12 The above formula can be used to eliminate the geometric distortion in the image and generate a distortion-free image I 13 .

[0081] In this embodiment, the image plane non-uniformity correction algorithm: The non-uniformity of the CCD image comes from the inherent optical aberration of the optical system itself and the corresponding non-uniformity of different CCD pixels. In the laboratory calibration process, by imaging a uniform surface light source, the image plane response non-uniformity matrix can be obtained. The matrix is ​​normalized to obtain the flat field calibration matrix R[512×512], the maximum value of which is 1, and the values ​​of other matrix elements are greater than 0 and less than 1. Let the scientific image after flat field calibration be I L1 [512×512], then: I L1 [i,j]=I 13 [i,j] / R[i,j], i = 1, 2, ..., 512; j = 1, 2, ..., 512, this image is the final processing result of L1 level scientific data.

[0082] In this embodiment, physical quantity interpretation is performed according to the definitions of auxiliary data and engineering data, and scientific data and auxiliary data are stored as CDF format files according to the requirements of the ground application system.

[0083] In this embodiment, physical dimension conversion: according to the experimentally calibrated instrument sensitivity coefficient S and the instrument sensitivity variation coefficient S1 obtained in the on-orbit test (when there is no on-orbit test, this value is configured as 1.0 by default). The image is converted from DN value to the radiation intensity unit Rayleigh. The image after physical dimension conversion I L2 For: IL2 =I L1 / (S*T*S1) where T is the image exposure time.

[0084] In this embodiment, pixel positioning: Pixel positioning is to project the corrected image along the light direction corresponding to the pixel onto a spherical surface at an altitude of 110 km to obtain the geographic longitude and latitude coordinates of the projection point. The parameters required in the projection process include: detector pixel size, optical system focal length, instrument coordinate system conversion matrix (instrument geometry calibration matrix, etc.), satellite coordinate system conversion matrix (instrument installation matrix), geospatial coordinate system conversion matrix (satellite attitude and position data), etc. After the L1 data is corrected, the optical instrument can be simplified to Figure 7 The system shown. The dots in the figure represent the principal points of the optical system. The distance from the principal point to the image is the equivalent focal length of the system, which is obtained by the optical test calibration process. The ray formed by each pixel in the CCD image and the principal point is the light corresponding to the pixel. After a series of coordinate transformations of the ray vector, the intersection of the ray and the aurora projection sphere can be obtained, thereby realizing the pixel positioning.

[0085] In this embodiment, the image I L2 Take any pixel P[x,y] as an example. The pixel distance Lp from point P to the center of the image is:

[0086]

[0087] Among them, (x c ,y c ) represents the pixel position of the optical axis in the CCD. Then the angle of point P relative to the optical axis, that is, the pitch angle θ of the light corresponding to point P in the image coordinate system (ICS) is: Where f is the focal length of the optical system, and pixel is the CCD pixel size. The azimuth angle φ of point P in ICS is: Therefore, the unit vector V of the light corresponding to point P in ICS can be calculated ICS for:

[0088]

[0089] Then the unit vector representation of the light vector in GEO is:

[0090]

[0091] Among them, T ICS-CCS is the transformation matrix from the image coordinate system ICS to the camera coordinate system (CCS), T CCS-SCSis the transformation matrix from the camera coordinate system CCS to the satellite coordinate system (SCS), T SCS-GSE is the transformation matrix from the satellite coordinate system SCS to the GSE coordinate system (Geocentric Solar Ecliptic, GSE), T GSE-GEO T is the transformation matrix from the GSE coordinate system to the GEO coordinate system (GEO graphic coordinate system, GEO). ICS-CCS and T CCS-SCS All of them are obtained and solidified during laboratory tests and payload installation tests. T is calculated based on the three Euler angles of the satellite attitude in the GSE coordinate system. SCS-GEO Let the three Euler angles be β, θ, γ, and the regularization method be 123. Then the transformation matrix is:

[0092]

[0093] In this embodiment, T GSE-GEO The calculation process is as follows: Let the input universal time be UT, and its corresponding modified Julian day be MJD. First, calculate the modified Julian century number of this time relative to 12:00:00 on January 1, 2000, T0 = (MJD-51544.5) / 36525.0. First, calculate the transformation matrix T1 from GSE to GEI (Geocentric Equatorial Inertial). This matrix consists of two rotations, rotating around the Z axis by an angle of α and rotating around the X axis by an angle of ε. The calculation method of these two angles is

[0094] M=357.528+35999.05T0+0.04107UT

[0095] Λ=280.460+36000.772T0+0.04107UT

[0096] α=Λ+(1.915-0.0048T0)sinM+0.020sin2M

[0097] ε=23.439-0.013T0

[0098] Then the rotation matrix T1 is:

[0099]

[0100] The conversion matrix T2 from GEI to GEO consists of a rotation, rotating around the Z axis by an angle λ, which is calculated as:

[0101] λ=100.416+36000.770T0+15.04107UT

[0102] Then the rotation matrix T2 is:

[0103]

[0104] Finally, we get:

[0105] T GSE-GEO =T2T1

[0106] According to the satellite position (Xsat, Ysat, Zsat) and V GEO The intersection point Pg of the light ray and the 110km altitude sphere can be calculated. First, based on the satellite position and V GEO Establish the parameter representation equation of light in GEO:

[0107]

[0108] Where t is the parameter of the line, t=0 represents the satellite position, and t>0 represents the line extending toward the earth. Solve for the relationship between the line and the radius R 110 =R E +110km(R E =6371.0km), we can get the light projection point P corresponding to pixel P. g The geographical longitude and latitude [GLAT, GLON], P g The coordinates are converted to the geomagnetic coordinate system, and the magnetic longitude and latitude [MLAT, MLON] and corrected magnetic longitude and latitude [AACGMLAT, AACGMLT] of the projection point can be obtained. After all pixels are projected, six positioning matrices can be generated. The matrix format is consistent with the scientific image format, which is 512×512. They are the projection point latitude matrix M in the geographic coordinate system, GLAT , the projection point longitude matrix M in the geographic coordinate system GLON , the magnetic latitude matrix M of the projection point in the magnetic coordinate system MLAT And the magnetic longitude matrix M in the magnetic coordinate system MLON , the projection point magnetic latitude matrix M under the altitude-adjusted corrected geomagnetic coordinate system (AACGM) AACGMLAT , the projection point magnetic local time matrix M in the AACGM coordinate system AACGMMLT The coordinate system conversion can be done using a public program module. When the SMILE satellite is at the apogee of the orbit (generally at an altitude greater than 80,000 km), part of the light from the UVI camera will extend into space and have no intersection with the 110 km altitude sphere. At the pixels corresponding to this part of the light, the corresponding values ​​of the six projection point matrices will be filled with -999.9.

[0109] In this embodiment, Keogram data is generated: in GEO, the grid range is 50°N~90°N, and the resolution varies with the change of SMILE satellite altitude. In AACGM, the grid range is from 40°N to 90°N, and the resolution also varies with the change of SMILE satellite altitude. Based on the AACGM map, three variables "keogram_ns", "keogram_ew", and "uvi_roi" will be generated, which are the intensity curves in the north-south direction, the intensity curves in the east-west direction, and the average intensity within the characteristic range. "keogram_ns" will be at 23h MLT (MLT is magnetic local time) from 40°-90°AACGM latitude with a resolution of 0.4°; "keogram_ew" along the magnetic latitude from 20h MLT to 02h MLT at 66°AACGM latitude, with a resolution of 0.05h. "uvi_roi" refers to the average value of the intensity in the area defined by 23±1hMLT and 66°±3°AACGM latitude, which is used to monitor the outbreak of most substorms. These three variables will be stored in the CDF file of L2.

[0110] In this embodiment, after completing pixel positioning and coordinate conversion, scientific data, positioning data, and auxiliary data are stored as CDF format files according to the requirements of the ground application system. The label of the L2 aurora data file is UVI-AURORA-GEO, and the CDF format is a self-analyzing file. Users can easily identify the variables and their physical meanings according to the variable attributes.

[0111] In this embodiment, L3 data is mainly processed in depth on the basis of L2 scientific data according to the needs of scientific teams or users to obtain richer scientific information. It mainly includes functions such as image airglow background removal, aurora boundary identification, and analysis of other satellite conjugate crossings, and retains the ability to expand functions in the future according to overall and user requirements. In principle, all L3 data inherits the content of L2 data, and new content is added to the data as needed.

[0112] In this embodiment, the image airglow background is removed: the image airglow background removal plan uses the GLOW model to calculate the LBH band airglow radiation intensity in the field of view in real time. Specifically, the time code when the scientific data was taken is extracted according to the auxiliary data in the L2 file, and the corresponding solar activity index is queried according to the time code, including the F107 index of the day, the F107 index F107p of the previous day, and the 81-day average F107 index F107a; at the same time, the daily average Ap index at that time is queried according to the time code. Generate a global input file in accordance with the GLOW input file format at 1°×1° geographic longitude and latitude intervals. When calculating airglow, all the sedimentation electron parameters are set to 0. The GLOW model is called to calculate the global LBH band airglow body radiance matrix Ver[361×181×102], with longitude dimension 360 (0~360°, interval 1°), latitude dimension 181 (-90°~90°, interval 1°), and altitude dimension 102 (80~640km, interval 1-10km varies with altitude). Read the light data in the L2 scientific data file, select the pixel with value 1 according to the mask matrix in the file, and use the corresponding light vector to calculate the body radiance integral on the path, with the integral altitude range of 80~640km. Thus, a global airglow image (data unit: Rayleigh) without auroral radiation is obtained. The airglow image is stored in the L3 scientific data for users to choose and use. At the same time, a quick view and thumbnail with the airglow background subtracted are generated.

[0113] In this embodiment, aurora boundary identification: aurora boundary identification is carried out on the basis of scientific data after removing the airglow background. The boundary identification algorithm is implemented in the AACGM coordinate grid. The relationship between the energy of the faint luminous point and the total energy in each small area of ​​0.5h×1° in the grid is checked in turn. If it is less than 44%, it is eliminated. For the pixel window to be tested, all pixels are extracted along the MLT to construct a vector. If all points of the vector are greater than zero (no zero values ​​are mixed in it), it is regarded as a continuous area. Even if it meets the above-mentioned criterion of less than 44%, it is still retained. The fuzzy clustering algorithm performs segmentation operations and clusters into 6 categories. Each cluster center is tested and the noise point area is eliminated. Along different MLT directions, the equatorial boundary and the polar boundary are found.

[0114] In this embodiment, noise points are removed: discrete noise points are inevitable in the image after the airglow background is removed. For stray noise points outside the auroral egg area observed by UVI, it is assumed that auroral precipitation (low-energy or high-energy precipitation particles) occurs continuously in the specified area, which is reasonable because diffuse aurora always occurs continuously. Then the aurora data within each 0.5h×1° small area in the magnetic grid is checked in turn, the sum of the intensities of all points in the above small grid is calculated, and then the relationship between the intensity of the weak luminous point and the total intensity is determined. If it is ≤44%, it is removed and these points are regarded as noise points. In order to ensure that the pixels near the aurora boundary are not erased, we introduce a criterion for judging the continuity of the geomagnetic latitude. For the pixel window to be checked, all pixels are extracted along the MLT to construct a vector. If all points of the vector are greater than zero (no zero values ​​are mixed in it), we regard it as a continuous area. For the continuous area of ​​the aurora boundary, we still retain it even if the above ≤44% criterion is met.

[0115] In this embodiment, fuzzy clustering algorithm: Cluster analysis is a multivariate analysis method in mathematical statistics, which quantitatively determines the similarity / closeness between samples, thereby objectively dividing the sample types (classifying samples). Taking into account the fuzzy boundary characteristics of the far-ultraviolet aurora image, the fuzzy local information C-means clustering segmentation algorithm is implemented on the complete aurora egg image. The purpose of this approach is to eliminate the non-auroral components in the aurora egg. This method fuses local spatial information and grayscale information in a fuzzy local similarity test method, which ensures insensitivity to noise and target details. This method is achieved by applying the variance objective function: A fuzzy factor is introduced into Thus, the local spatial information and intensity information are fused in a fuzzy local similarity test, which ensures that the target details are not only insensitive to noise, but also can be guaranteed. Where N is the total number of pixels, c is the number of clusters (2≤c <N),x i is the grayscale value of the i-th pixel in the local pixel window (3×3 pixels by default), v k is the center of cluster k, u ki is the pixel x i The membership degree in the k class, m is the weight factor (m = 2), N i The expression falls into x i The neighboring pixels within the pixel window centered at the center, d ij is the spatial Euler distance between pixels i and j. In order to make the target J m Reaching the minimum value, u ki and v k Calculated by the following two equations:

[0116]

[0117] Each aurora image is clustered into 6 categories (labeled as category 1, category 2, ..., category 6 from strong to weak) using the fuzzy clustering algorithm, and then the regions are merged to remove the background area. The criteria for regional merging are as follows: let the intensity of each cluster center be v1, v2, ..., v6 respectively. If v3>80, the aurora region is defined to include category 1, category 2 and category 3; if v2>80, the aurora region includes category 1 and category 2. The above two cases correspond to the situation where the aurora intensity is very strong. For general aurora intensity, we find the maximum value of the difference between adjacent cluster center values. If the v1-v2 value is the largest, the aurora region is defined to include only category 1; if the v2-v3 value is the largest; the aurora region is defined to include category 1 and category 2; if the v3-v4 value is the largest, the aurora region is defined to include category 1, category 2 and category 3. If the aurora region obtained by the above method contains multiple independent blocks, we set the unselected aurora data points to zero, thereby separating the aurora data from the background data.

[0118] In this embodiment, the boundary is found: the aurora image after discrete noise removal and fuzzy clustering segmentation is smoothed using a 3×3 pixel window to facilitate subsequent processing. With the geomagnetic pole as the center, a ray is drawn from 0h MLT to 24hMLT to find non-zero values ​​of aurora data distributed in the direction of the ray, wherein the maximum geomagnetic latitude corresponding to these non-zero values ​​is the poleward boundary of the aurora oval in this direction, and the corresponding minimum geomagnetic latitude is the equatorial boundary in this direction.

[0119] In this embodiment, conjugate crossing analysis: The conjugate crossing analysis function is mainly to search for satellites whose foot points along the magnetic field lines are within the range of the aurora egg according to the time when the aurora image was taken, and the position of the projected foot points. At the same time, according to the information of the fixed ground aurora station, the field of view of the ground all-sky camera is superimposed to provide convenience for data users. The main method is: according to the time code in the L2 data, the different satellite orbits in the database are searched. The satellites planned to be included include Fengyun series, MMS magnetosphere detection satellites, Arase magnetosphere detection satellites, DMSP series ionosphere detection satellites, GOLD satellites, ICON satellites, GOES series geosynchronous orbit satellites, and POES series polar orbit satellites. The database is planned to be established and updated separately. The Geopack toolkit is used to track the magnetic field lines of each satellite orbit position at this moment (the T96 magnetic field model is planned to be used by default, and the software will be set to be configurable. The optional models include: IGRF, T89, TS05, TA15, etc.), and the satellite position is mapped to an altitude of 110km. If the position is within the identified auroral egg boundary, it is determined that there is a conjugate crossing, and the footpoint information is stored in the L3 file as a structure type, including the satellite name, orbital position, and projected footpoint position. According to the time code in the L2 data, the position of the ground auroral station in the image is calculated (the auroral station data location information is put into the data processing software as a configuration file in advance, mainly including THEMIS GBO stations, SMILE GBO stations, etc., which can be updated and maintained). If its field of view intersects with the UVI aurora image, the information of the ground station is stored in the L3 data, mainly including the station name and station location.

[0120] In this embodiment, the L3 scientific data content is based on the complete inheritance of the L2 scientific data, and adds airglow background data, auroral oval boundary data, conjugate satellite footpoint data, ground station location data, etc. The corresponding quick view / thumbnail contains two sub-images, one sub-image is the image without airglow background removal, and the other sub-image is the image without airglow background removal. Both sub-images are superimposed with latitude and longitude or continent boundary grids, superimposed auroral boundaries, and superimposed with conjugate satellite names and positions (if they exist at the time), ground station positions and field of view ranges (if they exist at the time).

[0121] Although the specific implementation of the present invention is described in detail in conjunction with the drawings, it should not be understood as limiting the scope of protection of this patent. Within the scope described in the claims, various modifications and variations that can be made by those skilled in the art without creative work still fall within the scope of protection of this patent.

Claims

1. A method for processing data of a SMILE (Solar wind Magnetosphere Ionosphere Link Explorer) satellite ultraviolet aurora imager, comprising an image display area and a function button area, characterized in that: The method comprises the following steps: Step 1: Start the software, open the UVI (Ultraviolet Imager) data processing software, and enter the main interface; Step 2: Set the operation mode. Click "Auto" to set it to automatic mode, and click "Manual" to set it to manual mode. Step 3, set the relevant file paths: "Task list path" button selects the UVI data distribution list folder that needs to be monitored; "L0 level data path" button selects the top-level folder that stores UVI's L0 level data and auxiliary data; "Data output path" button selects the top-level folder that stores the data products of all levels produced by UVI; Step 4: Select automatic mode or manual mode according to your needs and start the processing; The specific processing steps of the automatic mode in step 4 are as follows: Step 4.

1. Click the "Auto" button to set to automatic mode; Step 4.2, for automatic mode, after the above settings are completed, click the "Start" button, the software starts to monitor the folder selected by the "Task List Path" button, automatically identifies the newly added data processing task list file *.json in the folder, and automatically grabs the file for full-process automatic processing. The processing starts from L0 level data by default, including reading the original aurora data in the file, generating aurora quick view images, correcting the original data and generating L1 data, where the received L0 level aurora image data is corrected, including dark current correction, image bright and dark spot removal, image plane non-uniformity correction, image plane geometric distortion correction, physical dimension conversion, pixel-by-pixel projection positioning, and coordinate system conversion are performed on the basis of L1 data to generate L2 level data products; Step 4.3: After the data processing task is completed, if there is a new task, repeat the above process. If there is no new task, the software is in monitoring and waiting state. The user clicks the "Stop" button to stop monitoring and exit the program. The specific processing steps of the manual mode in step 4 are as follows: Step 4.

4. Click the "Manual" button to set to manual mode; Step 4.5: For manual mode, the user must first send a data processing request to the SMILE satellite science data processing subsystem integrated data processing software as needed, with the requested data level being L0, L1, or L2; Step 4.6, set the time period for requesting data, click the "Start Time" and "End Time" buttons to set the start and end time. After clicking the "Start Time" button, the start time selection interface will pop up, and the user can set the time by clicking the corresponding drop-down menu or button. After setting, click the "OK" button, and the selected start time will be displayed in the text box after "Start Time". After clicking the "End Time" button, the end time selection interface will pop up, and the user can set the time by clicking the corresponding drop-down menu or button. After setting, click the "OK" button, and the selected start time will be displayed in the text box after "End Time"; Step 4.7, select the data level and version you need to request, select the drop-down menu after "Data level" and "Data version", the default is L0 level data and V01 version respectively; Step 4.

8. Click the "Request Science Data" button to send the data processing task list file to the SMILE satellite science data processing subsystem integrated data processing software, then click the "Start" button to start monitoring the UVI data distribution list folder to see if there is a corresponding data list file sent by the system; Step 4.9, wait for the SMILE satellite science data processing subsystem integrated data processing software to respond to the list file, generate the data requested by the user based on it, and send the data processing task file to the software distribution list folder; Step 4.10: After the file monitoring module responds, a pop-up window will display the status: "Request for scientific data completed!", and then the user clicks the relevant button in the lower right corner to process the data; Step 4.11, if the data requested by the user is of level L0, the user should click the "Process L0 data" button to process the L0 data into L2 data. The functions of the "Process L0 data" button include L0 quick view generation, L1 data processing and L2 data processing; The specific steps of L1 data processing are as follows: Step 4.11.1, read in L0 level scientific data, auxiliary data and engineering telemetry data; Step 4.11.2, correct the received L0-level aurora image data, including dark current correction, image bright spot and dark spot removal, image plane non-uniformity correction, image plane geometric distortion correction, to generate L1-level data products; Step 4.12, if the data requested by the user is of level L1, the user should click the "Process L1 data" button to process the L1 data into L2 data. The functions of the "Process L1 data" button include physical dimension conversion, pixel-by-pixel projection positioning, and coordinate system conversion based on L1 data to generate L2 data products; The specific steps of L2 data processing are as follows: Step 4.12.1, read in L1 level scientific data, auxiliary data and engineering telemetry data; Step 4.12.2, L2 data processing: Perform physical dimension conversion, pixel-by-pixel projection positioning, and coordinate system conversion on the received L1 aurora image data to generate L2 data products; Step 4.13, if the data requested by the user is of level L2, the user should click the "Process L2 data" button to process the L2 data into L3 data. The functions of the "Process L2 data" button include airglow background removal, aurora boundary identification, and conjugate crossing analysis to generate L3 data products; The specific steps of L3 data processing are as follows: Step 4.13.1, read in L2 level scientific data, auxiliary data and engineering telemetry data; Step 4.13.2: Perform airglow background removal, aurora boundary identification and conjugate crossing analysis on the received L2 aurora image data to generate L3 data products; Step 4.14: After the data processing task is completed, if there is a new task, repeat the above process; if there is no new task, the user clicks the "Stop" button to stop monitoring and exit the program; Step 5. The user views the generated Aurora QL, L1-level data, L2-level data, and L3-level data in the corresponding folders; Step 6. After the task is completed, click the "Stop" button to stop monitoring the task path and exit the software.

Images