Satellite-borne microwave scatterometer azimuth resolution enhancement method based on image reconstruction technology

Abstract

The invention discloses a satellite-borne microwave scatterometer azimuth resolution enhancement method based on an image reconstruction technology, and the method comprises the steps: dividing a range-direction wave beam footprint into a plurality of resolution units according to the range-direction resolution demands, converting the power of the resolution units into a backscattering coefficient through calibration; determining an observation position corresponding to the backscattering coefficient through geographic positioning based on observation geometric information; one-dimensional uniform grids are continuously divided in a segmented manner along the ground footprint track formed by each distance to the center of the corresponding beam of the distinguishing unit during antenna scanning; and reconstructing the backscattering coefficient in the grid by applying a scatterometer image reconstruction algorithm based on the distance-direction high-resolution backscattering coefficient and the projection information of the corresponding spatial response function in the grid region. By overcoming the limitation of the existing processing method, the azimuth resolution of the spaceborne microwave scatterometer can be improved reliably and efficiently in real time, and the method is a practical data processing method of the spaceborne microwave scatterometer.

Description

Technical Field

[0001] The invention relates to a resolution enhancement method for a satellite-borne microwave scatterometer, belonging to the technical field of space microwave remote sensing. Background Art

[0002] Spaceborne microwave scatterometers of fan-beam and pencil-beam systems usually use linear frequency modulation signal de-skewing processing technology to achieve range pulse compression, and use the correspondence between the signal frequency and the delay (distance) after de-skewing to divide the echo power into multiple high-resolution measurement units in the range direction. However, due to the limitation of the inherent beam width of the antenna, high-resolution measurement cannot be achieved in the azimuth direction. In order to improve the azimuth resolution of the backscatter coefficient observation data, two main processing methods are currently used:

[0003] (1) Scatterometer image reconstruction method: By introducing a numerical iterative algorithm to solve the observation equation model, the noise signal amplification in the reconstruction process is suppressed, and two-dimensional high-resolution reconstruction in distance and azimuth is achieved. However, this method requires the accumulation of forward and backward observation data from multiple tracks or multiple days or the same track in the reconstruction area. Therefore, it cannot meet the needs of real-time high-resolution detection of sea surface wind fields. It can only be applied to land and the North and South Poles where the scattering characteristics remain basically unchanged for a long time and are insensitive to the instrument observation azimuth. In addition, this method has low processing efficiency due to the need to accumulate data.

[0004] (2) A one-dimensional high-resolution processing method in azimuth based on strip measurement data; used to solve the problem of insufficient azimuth resolution in the forward and backward observation areas of high-resolution and high-precision scatterometers using range-Doppler processing technology. This method is based on the overdetermined or completely determined observation equation model established by the scatterometer in the forward and backward observation areas at a high sampling rate, and uses an improved adaptive regularization algorithm to achieve one-dimensional high-resolution reconstruction in azimuth. However, this method relies on a high pulse sampling rate to ensure that the observation matrix model is positive or overdetermined, and is only applied to observation data in the forward and backward observation areas. Summary of the invention

[0005] The present invention provides a method for enhancing the azimuth resolution of a satellite-borne microwave scatterometer. The method can generate backscatter coefficients with high resolution in range and azimuth at all scanning azimuth positions of the scatterometer, and can improve the azimuth resolution of the satellite-borne microwave scatterometer in real time, reliably and efficiently.

[0006] The technical solution of the present invention includes:

[0007] A method for enhancing azimuth resolution of a spaceborne microwave scatterometer based on image reconstruction technology, comprising:

[0008] Step 1: High-resolution processing in range: After the range resolution unit is subdivided, the backscatter coefficient of the range resolution unit is calibrated and the geolocation processing of the range resolution unit is performed, the backscatter coefficient of the range resolution unit is obtained and the latitude Lat of the center of the footprint of each range resolution unit is determined. p and Lon p , and determine the corresponding ground incidence angle Eia p and geographic azimuth Azi p , and retain the antenna scanning angle Scan that reflects the forward and backward states of the observation data p ;

[0009] Step 2: azimuth one-dimensional resolution enhancement processing: through azimuth one-dimensional reconstruction parameter calculation, original observation data reorganization, azimuth high-resolution reconstruction and reconstruction data geographic geometry auxiliary information;

[0010] A backscatter coefficient with high resolution in both the range and azimuth directions is obtained, and at the same time, the longitude, latitude, ground incidence angle, geographic azimuth and antenna scanning angle information corresponding to the high-resolution backscatter coefficient is obtained.

[0011] Optionally, step 1, distance high-resolution processing specifically includes:

[0012] The range beam footprint is divided into several resolution units according to the range resolution requirement. The resolution unit power is converted into backscatter coefficients through calibration based on the corresponding relationship between echo power slice distribution and range resolution unit and system parameters. At the same time, the observation position corresponding to the backscatter coefficient is determined through geographic positioning based on the observation geometry information. The number of measurement units that the scatterometer can provide in the range direction for each observation, that is, the number of range resolution units, is calculated by the range resolution unit subdivision step. Each range resolution unit corresponds to a power slice within a certain range in the echo measurement value. The power slice corresponding to each range resolution unit is accumulated and calibrated through the range resolution unit backscatter coefficient calibration step to determine the backscatter coefficient corresponding to the range resolution unit. Finally, the geographical location and other information corresponding to the backscatter coefficient of each range resolution unit are determined through the range resolution unit geographic positioning step, including latitude, longitude, ground incidence angle, geographical azimuth and antenna scanning angle, to provide input for the one-dimensional resolution enhancement processing in azimuth. The backscatter coefficient of each range resolution unit and the corresponding geographical location data are arranged in the order of the original scanning observation time of the scatterometer.

[0013] Optionally, in step 2, the azimuth one-dimensional resolution enhancement processing specifically includes:

[0014] The surface coverage of the specific range resolution unit participating in the azimuth resolution enhancement processing along the scanning footprint direction is determined through the azimuth one-dimensional reconstruction parameter calculation step, and then the number of observations of the specific range resolution unit participating in the azimuth high-resolution reconstruction is determined each time; before the azimuth high-resolution reconstruction, the antenna scanning angle and the geographic azimuth are converted into angles, and then the range resolution unit backscatter coefficient, latitude, longitude, adjusted antenna scanning angle and geographic azimuth, and incident angle are reorganized in four dimensions according to the scanning start angle and the front and back sight marks within the scanning cycle as input for the azimuth high-resolution reconstruction processing step; the high-resolution reconstruction processing step is sequentially performed in the distance resolution unit dimension, the scanning footprint direction, and the scanning footprint direction. The input data is traversed in the scanning period dimension and the fore-and-aft dimension, and the backscatter coefficient of the range resolution unit is processed by the scatterometer image reconstruction algorithm based on the reconstruction parameters and the spatial response function information, so that the backscatter coefficient of each range resolution unit is enhanced in the azimuth direction; the latitude, longitude, antenna scanning angle and geographic azimuth information corresponding to the backscatter coefficient processed with the azimuth resolution enhancement are calculated from the latitude, longitude, adjusted antenna scanning angle and geographic azimuth of the range resolution unit through the reconstruction data geographic geometry auxiliary information calculation step, and finally the backscatter coefficient with enhanced resolution in both the range and azimuth directions, as well as the corresponding latitude, longitude, antenna scanning angle and geographic azimuth information are obtained.

[0015] Optionally, the distance to high resolution processing includes:

[0016] (1) Distance resolution unit subdivision

[0017] Subdivide the ground footprint into a number of ground resolution units N Ele :

[0018]

[0019] S Ele ΔR is the ground footprint of the 3dB beam from the antenna. Ele The required resolution is for distance;

[0020] (2) Calibration of backscatter coefficient of range resolution unit

[0021] In the real-time processing of remote sensing data, the echo power slice accumulation range and illumination integral parameter I corresponding to each range resolution unit of each pulse observation are obtained by searching the power slice accumulation range and illumination integral parameter table based on the real-time sub-satellite point latitude and antenna scanning angle. 0r , calculate the echo power P of each pulse distance resolution unit 0r,p :

[0022]

[0023] r1 and r2 represent the start and end numbers of the echo power slices contained in the rth range resolution unit, respectively. i represents the echo power slice number. The above formula means that the echo power slices ranging from r1 to r2 are selected for accumulation. r = 1, 2…N Ele , p represents the scanning observation point number;

[0024] Based on the internal calibration data, P 0r,p After noise reduction processing to obtain pure signal power P sr,p , combined with the illumination integration factor and radar system parameters, the range resolution unit backscatter coefficient is obtained

[0025]

[0026] X1 is a fixed parameter related to the system, and X2 is a parameter related to the calibration within the system;

[0027] (3) Distance resolution unit geolocation

[0028] Each pulse observation distance is geographically located in each resolution unit, and the center latitude of each distance resolution unit footprint is determined. p and Lon p , and determine the corresponding ground incidence angle Eia p and geographic azimuth Azi p , and retain the antenna scanning angle Scan that reflects the forward and backward states of the observation data p .

[0029] Optionally, the one-dimensional resolution enhancement processing in azimuth includes:

[0030] (1) Calculation of one-dimensional reconstruction parameters in azimuth

[0031] At a specific satellite position vector and the velocity vector When the antenna beam center viewing angle and antenna beam center azimuth are θ0 and When the conversion matrix M from the earth fixed coordinate system to the antenna coordinate system is calculated EA :

[0032] M EA =M Ecf2Ant M Ins (4);

[0033] Among them, M Ecf2Ant is the transformation matrix related to the satellite position vector, velocity vector, beam center viewing angle and beam center azimuth, M Ins Represents the instrument installation offset matrix;

[0034] Assume that the beam angle corresponding to each resolution unit is θ r ,r=1,2…N Ele , the azimuth of each resolution unit is consistent with the azimuth of the antenna beam center. Assume that the antenna beam width of the resolution unit azimuth is 2φ0, and the azimuth angle traversal range used by the resolution unit azimuth projection is The traversal interval is Δφ, Covering the first zero point position of the antenna azimuth pattern; according to the satellite position vector Velocity Vector Radar instrument installation matrix, resolution unit beam viewing angle θ r , resolution unit projection traversal azimuth j=1,2…M, M is the number of azimuth subdivisions of the resolution unit, and the ground projection point corresponding to the distance resolution unit is calculated. from arrive The connecting vector is:

[0035]

[0036] Vector The modulus of is the distance from the satellite to the projection point on the ground:

[0037]

[0038] Use the transformation matrix M from the earth fixed coordinate system to the antenna coordinate system EA Will Convert to antenna coordinate system;

[0039]

[0040] Depend on Calculated vector Elevation angle and azimuth angle in antenna coordinate system:

[0041]

[0042]

[0043] It can be determined by looking up the table The corresponding antenna two-way gain G rj , combined with the slope distance R rj Get the spatial response function RSF of each range resolution unit rj :

[0044]

[0045] About RSF rj according to Normalized to get the normalized spatial response function RSF rj,n :

[0046]

[0047] Correspondingly, the ground projection point coordinates of the distance resolution unit Calculate the orientation of each projection point relative to the central projection point The distance D Rsf,rj , the relative distance D of the center projection point Rsf,cen = 0; through the sign conversion will correspond to the azimuth Part D Rsf,rj The space function is converted to a negative value to facilitate azimuth reconstruction, that is:

[0048]

[0049] When the microwave scatterometer is working normally, the satellite position corresponding to different observation points changes as the satellite platform moves. When the conical scanning observation method is used, the satellite position corresponding to any fixed observation point Other scanning observation points and (m=nM,n-M+1…n+M,m≠n) The connecting line vector is: n represents the serial number of any fixed observation point, and m represents the serial number of 2M observation points adjacent to the observation point before and after:

[0050]

[0051] The corresponding beam viewing angle is:

[0052]

[0053] The corresponding beam angle α mn The change should be controlled within each distance resolution viewing angle range Δα r , and thus determine the observation points involved in the reconstruction Among the 2M observation points adjacent to each other in time, there are M0 observation points that meet the condition that the coverage area is located in the same range resolution unit. The width ΔS of the effective reconstruction area in azimuth is:

[0054] ΔS=M0ΔP (15);

[0055] ΔP is the ground surface distance between the centers of adjacent pulse beam footprints along the scanning direction;

[0056] (2) Reconstruction of original observation data

[0057] Adjust the raw data antenna scanning angle and geographic azimuth to within ±180°:

[0058]

[0059] Scan p and Azi p They represent the scanning azimuth and the geographical azimuth of the center of the beam footprint corresponding to the backscattering coefficient obtained after high-resolution processing in the range direction, and p represents the serial number of the scanning observation point;

[0060] Calculate the forward and backward look marks for each scan observation based on the antenna scan angle:

[0061]

[0062] The backscatter coefficient, latitude, longitude, adjusted antenna scanning angle and geographic azimuth, incident angle, and fore-aft and aft-aft identification of each range resolution unit obtained through range high-resolution processing are arranged according to the periodic scanning observation sequence of the scatterometer;

[0063] The data obtained by range-direction high-resolution processing are reorganized according to the following method: the forward-sight scanning azimuth angle closest to -90° in each scanning cycle is used as the scanning start angle, and the backscatter coefficient, latitude, longitude, adjusted antenna scanning angle and geographic azimuth, incident angle, forward and backward sight marks are periodically arranged according to the scanning start angle value, and the data is divided into forward-sight state data and backward-sight state data according to the forward-sight and backward-sight marks in each cycle;

[0064] The reorganized data is divided into four dimensions. The first dimension corresponds to different distance resolution units, the second dimension corresponds to different scanning cycles, the third dimension corresponds to front and rear vision, and the fourth dimension corresponds to different observation points in front or rear vision.

[0065] (3) Azimuth high-resolution reconstruction

[0066] Select a specific distance resolution unit and forward or backward observation data within the scanning cycle, and divide the continuous scanning observation point beam footprint coverage evenly into N P intervals, each interval length is the width of the azimuth reconstruction region ΔS, and the Pth interval is expressed as [S P S P +ΔS],S P represents the distance between the starting point of the Pth interval to be reconstructed and the center of the footprint of the first scanning point along the scanning trajectory;

[0067] Reconstruct the area based on coverage [S P S P +ΔS] scanning observation point ground projection coordinates calculate The distance D from the center of the footprint relative to the starting observation point of the scanning observation segment along the scanning trajectory Azi,p ;

[0068] According to the target resolution requirement of one-dimensional reconstruction in azimuth, the Pth interval to be reconstructed [S P S P +ΔS] is evenly divided into N Q intervals, each interval [R q R q +ΔR] represents the azimuth resolution unit with a size of ΔR, and the starting point of the interval is R q Indicates the distance between the starting point of the azimuth resolution unit and the center of the footprint of the starting observation point of the scanning observation segment;

[0069] For each observation point Its normalized spatial response function RSF ij,n The corresponding relative distance is D Rsf,j , then RSF ij,n The distance from the starting point of the scanning observation segment is D Azi,p +D Rsf,j , search distance value D Azi,p +D Rsf,j Located in the interval [R q R q +ΔR] and average them as the spatial response function value of the p-th observation point falling into the q-th azimuth resolution unit;

[0070]

[0071] Ind1 and Ind2 represent the interval [R q R q +ΔR] internal distance value D Azi,p +D Rsf,j The start and end range of the serial number, Q p =Ind2-Ind1+1;

[0072] Observation point The corresponding backscatter coefficient Combined with spatial response function RSF ij,n Application of Scatterometer Image Reconstruction Algorithm SIR to Realize Interval [S P S P +ΔS] in each azimuth resolution unit [R q R q +ΔR] backscatter coefficient reconstruction;

[0073] (4) Calculation of auxiliary information of geographic geometry of reconstructed data

[0074] Select the azimuth resolution unit [R q Rq +ΔR] is taken as the center of the reconstructed scattering coefficient, that is, The Lon corresponding to this position q , Lat q 、Antenna scanning angle Scan q 、Geographical azimuth Azi q and the angle of incidence Eia q The corresponding information Lon of the original observation data involved in the reconstruction of the backscatter coefficient of the area can be used p , Lat p 、Antenna scanning angle Scan p 、Geographical azimuth Azi p and the angle of incidence Eia p Determined by interpolation;

[0075] Finally, steps (3) to (4) are repeated to traverse all distance resolution units, scanning cycles, forward-looking and backward-looking observation data to achieve one-dimensional azimuth high-resolution reconstruction of all observation data. The processed backscatter coefficient has high resolution in both distance and azimuth.

[0076] Optionally, the scatterometer image reconstruction algorithm SIR realizes the interval [S p S p +ΔS] in each azimuth resolution unit [R q R q +ΔR] backscatter coefficient reconstruction, the specific reconstruction process is as follows:

[0077] a. The vector composed of the backscatter coefficients of each resolution unit in azimuth is S, s q represents the scattering coefficient of the qth resolution unit, and initializes the vector to an arbitrary vector S 0 ;

[0078] b. k+1th iteration:

[0079]

[0080] in:

[0081] λ is the iterative tuning factor, h pq is the average value of the normalized spatial response function of the pth original observation point involved in the reconstruction falling within the qth resolution unit;

[0082] In each iteration, the original observation data participate in the update of the backscattering coefficient of the azimuth resolution unit;

[0083] c. Repeat step b. As the number of iterations increases, the difference between the results of the previous and subsequent iterations gradually decreases until convergence.

[0084] The advantages of the present invention compared with the prior art are:

[0085] The present invention divides and reconstructs the grid area in real time along the center track of the beam footprint formed by antenna scanning and applies the SIR algorithm to perform resolution enhancement processing. It does not need to accumulate multi-day or multi-track or forward and backward observation data. Compared with the existing two-dimensional reconstruction processing method, it has higher real-time performance and simplifies the two-dimensional reconstruction problem into a one-dimensional problem.

[0086] The method provided by the present invention performs azimuth resolution enhancement processing at all scanning positions in each scanning cycle, which has higher coverage than the existing method that only processes in the front and rear view areas.

[0087] All scanning positions except internal calibration measurement can be covered in each scanning cycle.

[0088] The data processed by the method provided by the present invention for azimuth resolution enhancement synchronously retains the observed geometric information in the original data, which is convenient for various applications of the data after resolution enhancement, while the traditional processing method discards the observed geometric information of the original data involved in the reconstruction. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present disclosure but do not constitute a limitation of the present disclosure. In the accompanying drawings:

[0090] Figure 1 It is a flow chart of the azimuth resolution enhancement method of a spaceborne microwave scatterometer based on image reconstruction technology of the present invention;

[0091] Figure 2 This is the result diagram before distance high-resolution processing of Example 1;

[0092] Figure 3 This is a result diagram after distance-to-high-resolution processing in Example 1;

[0093] Figure 4 This is a result image after one-dimensional high-resolution reconstruction in azimuth of Example 1;

[0094] Figure 5 For example, the one-dimensional resolution enhancement processing and front and back scattering coefficient distribution map of isolated islands near Antarctica in azimuth direction; the one-dimensional reconstruction and front and back scattering coefficient distribution map in azimuth direction;

[0095] Figure 6 The scattering coefficient distribution map after one-dimensional resolution enhancement processing in azimuth for an isolated island near Antarctica processed in Example 1; and the scattering coefficient distribution map after one-dimensional reconstruction in azimuth. DETAILED DESCRIPTION

[0096] The present invention will be further explained and illustrated below in conjunction with the accompanying drawings and specific implementation methods.

[0097] The invention discloses a method for enhancing the azimuth resolution of a satellite-borne microwave scatterometer based on image reconstruction technology, relates to satellite-borne microwave scatterometer data processing, and belongs to the field of space microwave remote sensing technology. The invention comprises: firstly, dividing the range beam footprint into a plurality of resolution units according to the range resolution requirement, converting the resolution unit power into a backscatter coefficient through calibration based on the corresponding relationship between the echo power slice distribution and the range resolution unit and the system parameters, and determining the observation position corresponding to the backscatter coefficient through geographic positioning based on the observation geometry information. Based on the fact that the beam footprint corresponding to the range high-resolution backscatter coefficient has a certain overlap characteristic in the scanning direction, the ground footprint track formed by the beam center corresponding to each range resolution unit during antenna scanning is segmented and continuously divided into a one-dimensional uniform grid, and based on the projection information of the range high-resolution backscatter coefficient and the corresponding spatial response function in the grid area, a scatterometer image reconstruction algorithm (Scatterometer Image Reconstruction, SIR) is applied to realize the reconstruction of the backscatter coefficient in the grid, thereby improving the azimuth resolution, and finally obtaining a backscatter coefficient product with high resolution in both the range and azimuth directions. The present invention overcomes the limitations of existing processing methods and can improve the azimuth resolution of a satellite-borne microwave scatterometer in real time, reliably and efficiently. It is a practical satellite-borne microwave scatterometer data processing method.

[0098] The technical solution of the present invention is carried out in the following two steps:

[0099] Step 1: High-resolution processing in the range direction. The high-resolution processing in the range direction produces a backscatter coefficient product with high spatial resolution in the range direction through the steps of range resolution unit subdivision, range resolution unit backscatter coefficient calibration and range resolution unit geolocation processing, which serves as the basis for one-dimensional resolution enhancement processing in the azimuth direction.

[0100] Step 2: One-dimensional resolution enhancement processing in azimuth: Based on the high-resolution backscatter coefficient in range, the one-dimensional resolution enhancement processing in azimuth produces a backscatter coefficient product with high spatial resolution in azimuth through the steps of one-dimensional reconstruction parameter calculation in azimuth, reorganization of original observation data, high-resolution reconstruction in azimuth, and calculation of geographic geometric auxiliary information of reconstruction data.

[0101] Specifically, they include:

[0102] The azimuth resolution enhancement method of spaceborne microwave scatterometer based on image reconstruction technology processes the scatterometer raw data in the range and azimuth directions in turn, and finally obtains the backscatter coefficients and related information with high resolution in both the range and azimuth directions.

[0103] The number of measurement units that the scatterometer can provide in the range direction for each observation, i.e., the number of range resolution units, is calculated by the range resolution unit subdivision step. Each range resolution unit corresponds to a power slice within a certain range of the echo measurement value. The power slice corresponding to each range resolution unit is accumulated and calibrated by the range resolution unit backscatter coefficient calibration step to determine the backscatter coefficient corresponding to the range resolution unit. Finally, the geographical location and other information corresponding to the backscatter coefficient of each range resolution unit are determined by the range resolution unit geographic positioning step, including latitude, longitude, ground incidence angle, geographical azimuth and antenna scanning angle, to provide input for the one-dimensional resolution enhancement processing in azimuth. The backscatter coefficient of each range resolution unit and the corresponding geographical location data are arranged in the order of the original scanning observation time of the scatterometer.

[0104] The azimuth one-dimensional resolution enhancement processing first determines the surface coverage of the specific range resolution unit participating in the azimuth resolution enhancement processing along the scanning footprint direction through the azimuth one-dimensional reconstruction parameter calculation step, and then determines the number of specific range resolution unit observations participating in the azimuth high-resolution reconstruction each time. Before the azimuth high-resolution reconstruction, the antenna scanning angle and the geographic azimuth are converted into angles, and then the range resolution unit backscatter coefficient, latitude, longitude, adjusted antenna scanning angle and geographic azimuth, and incident angle are reorganized in four dimensions according to the scanning start angle and the front-back sight mark in the scanning cycle, as input for the azimuth high-resolution reconstruction processing step. The high-resolution reconstruction processing step traverses the input data in the range resolution unit dimension, scanning cycle dimension, and front-back sight dimension in turn, and processes the range resolution unit backscatter coefficient using the scatterometer image reconstruction algorithm based on the reconstruction parameters and spatial response function information to achieve azimuth resolution enhancement of the backscatter coefficient of each range resolution unit. The latitude, longitude, antenna scanning angle and geographic azimuth information corresponding to the backscatter coefficient processed by the azimuth resolution enhancement processing are calculated from the latitude, longitude, adjusted antenna scanning angle and geographic azimuth of the range resolution unit through the reconstruction data geographic geometry auxiliary information calculation step. After the above processing steps, the backscatter coefficient with enhanced resolution in both the range and azimuth directions, as well as the corresponding latitude, longitude, antenna scanning angle and geographic azimuth information are finally obtained.

[0105] After the above processing steps, a backscatter coefficient product with high resolution in both range and azimuth directions is obtained.

[0106] 1. Distance-to-high-resolution processing

[0107] (1) Distance resolution unit subdivision

[0108] According to the actual satellite orbit parameters and the wind field radar scanning observation geometry information, the 3dB beam width in the range is projected to obtain the ground footprint of the range beam. According to the requirements of the backscatter coefficient range resolution in the sea surface wind field inversion application, the ground footprint is subdivided into several ground resolution units N. Ele , N Ele Determined by the antenna distance to the 3dB beam ground footprint and the resolution required by the application:

[0109]

[0110] S Ele ΔR is the ground footprint of the 3dB beam from the antenna. Ele To meet the required resolution in range, different range resolution units have the same ground footprint.

[0111] According to the correspondence between the ground footprint unit and the observation angle, slant range, antenna gain and echo power slice position, the resolution unit echo power slice accumulation range, illumination integration parameters and beam angle lookup table are established.

[0112] (2) Calibration of backscatter coefficient of range resolution unit

[0113] In the real-time processing of remote sensing data, the echo power slice accumulation range and illumination integral parameter I corresponding to each range resolution unit of each pulse observation are obtained by searching the power slice accumulation range and illumination integral parameter table based on the real-time sub-satellite point latitude and antenna scanning angle. 0r , calculate the echo power P of each pulse distance resolution unit 0r,p :

[0114]

[0115] r1 and r2 represent the start and end numbers of the echo power slices contained in the rth range resolution unit, respectively; p represents the pulse observation sequence number; i represents the echo power slice number. The above formula means selecting the echo power slices ranging from r1 to r2 for accumulation, r = 1, 2…N Ele , p represents the pulse observation sequence number;

[0116] Based on the internal calibration data, P 0r,p After noise reduction processing to obtain pure signal power P sr,p , combined with the illumination integration factor and radar system parameters, the range resolution unit backscatter coefficient is obtained:

[0117]

[0118] X1 is a fixed parameter related to the system, and X2 is a parameter related to the calibration within the system.

[0119] (3) Distance resolution unit geolocation

[0120] Based on the information such as antenna scanning angle, satellite ephemeris and radar installation offset matrix at the time of pulse observation, and combined with the beam viewing angle obtained by table lookup, each pulse observation distance resolution unit is geographically located to determine the center latitude Lat of each distance resolution unit footprint. p , Lon p , and determine the corresponding ground incidence angle Eia p and geographic azimuth Azi p , and retain the antenna scanning angle Scan that reflects the forward and backward states of the observation data p .

[0121] 2. Azimuth one-dimensional resolution enhancement processing

[0122] (1) Calculation of one-dimensional reconstruction parameters in azimuth

[0123] At a specific satellite position vector and the velocity vector When the antenna beam center viewing angle and antenna beam center azimuth are θ0 and When the conversion matrix M from the earth fixed coordinate system to the antenna coordinate system is calculated EA :

[0124] M EA =M Ecf2Ant M Ins (4);

[0125] Among them, M Ecf2Ant is the transformation matrix related to the satellite position vector, velocity vector, beam center viewing angle and beam center azimuth, M Ins Represents the instrument installation offset matrix.

[0126] Assume that the beam angle corresponding to each resolution unit is θ r ,r=1,2…N Ele , the azimuth of each resolution unit is consistent with the azimuth of the antenna beam center. Assume that the antenna beam width of the resolution unit azimuth is 2φ0, and the azimuth angle traversal range used by the resolution unit azimuth projection is The traversal interval is Cover the first zero point position of the antenna azimuth pattern. According to the satellite position vector Velocity Vector Radar instrument installation matrix, resolution unit beam pointing θ r , resolution unit projection traversal azimuth j=1,2…M, M is the number of azimuth subdivisions of the resolution unit; the ground projection point corresponding to the distance resolution unit is calculated from arrive The connecting vector is:

[0127]

[0128] Vector The modulus of is the distance from the satellite to the projection point on the ground:

[0129]

[0130] Use the transformation matrix M from the earth fixed coordinate system to the antenna coordinate system EA Will Transform to antenna coordinate system

[0131]

[0132] Depend on Calculated vector Elevation angle and azimuth angle in antenna coordinate system:

[0133]

[0134]

[0135] It can be determined by looking up the table The corresponding antenna two-way gain G rj , combined with the slope distance R rj Get the spatial response function RSF of each range resolution unit rj :

[0136]

[0137] About RSF rj according to Normalized to get the normalized spatial response function RSF rj,n :

[0138]

[0139] Correspondingly, the ground projection point coordinates of the distance resolution unit Calculate the orientation of each projection point relative to the central projection point The distance D R sf,rj, relative distance D of the center projection point Rsf,cen = 0. By converting the sign, it will correspond to the azimuth Part D Rsf,rj The space function is converted to a negative value to facilitate azimuth reconstruction, that is:

[0140]

[0141] When the microwave scatterometer is working normally, the satellite position corresponding to different observation points changes as the satellite platform moves. When the conical scanning observation method is used, the satellite position corresponding to any fixed observation point Other scanning observation points and The connection vector is:

[0142]

[0143] The corresponding beam viewing angle is:

[0144]

[0145] In order to ensure the accuracy of one-dimensional high-resolution reconstruction in azimuth, the area covered by the observation points participating in the reconstruction must be located in the same distance resolution unit, that is, the corresponding viewing angle α mn The change should be controlled within each distance resolution viewing angle range Δα r , and thus determine the observation points involved in the reconstruction The effective reconstruction area width in azimuth is that there are M0 observation points in the 2M adjacent observation points before and after the above time that meet the condition that the coverage area is located in the same distance resolution unit:

[0146] ΔS=M0ΔP (15)

[0147] ΔP is the ground distance between the centers of adjacent pulse beam footprints along the scanning direction.

[0148] (2) Reconstruction of original observation data

[0149] To ensure the continuity of the antenna scanning angle and geographic azimuth of the reconstructed high-resolution data and avoid the antenna scanning angle and geographic azimuth jump problem caused by interpolation when reconstructing in the area formed by the continuous scanning footprint, the original data antenna scanning angle and geographic azimuth are converted to the range of ±180°:

[0150]

[0151]

[0152] Scan p and Azi p They represent the scanning azimuth corresponding to the pulse observation point and the geographical azimuth of the beam footprint center respectively.

[0153] Calculate the forward and backward identification of each raw observation data according to the antenna scanning angle:

[0154]

[0155] The backscatter coefficient, latitude, longitude, adjusted antenna scanning angle and geographic azimuth, incident angle, front and back sight marks of each range resolution unit obtained by range high-resolution processing are arranged according to the periodic scanning observation sequence of the scatterometer; the data obtained by range high-resolution processing are reorganized according to the following method: the forward-looking scanning azimuth with the closest distance in each scanning cycle is used as the scanning starting angle, and the backscatter coefficient, latitude, longitude, adjusted antenna scanning angle and geographic azimuth, incident angle, front and back sight marks are arranged periodically according to the scanning starting angle value, and the data is divided into two parts of forward-looking and backward-looking state data according to the forward-looking and backward-looking marks in each cycle; the original data (backscatter coefficient, latitude, longitude, antenna scanning angle, geographic azimuth) are arranged according to the antenna periodic scanning sequence, and each scanning cycle data consists of forward-looking and backward-looking state observation data, and the starting point of each scanning cycle data corresponds to the forward-looking scanning azimuth with the closest distance of -90°. The reorganized data is divided into four dimensions. The first dimension corresponds to different distance resolution units, the second dimension corresponds to different scanning cycles, the third dimension corresponds to forward and backward vision within a cycle, and the fourth dimension corresponds to different observation points in the forward or backward vision.

[0156] (3) Azimuth high-resolution reconstruction

[0157] Select a specific distance resolution unit and forward or backward observation data within the scanning cycle, and divide the continuous scanning observation point beam footprint coverage evenly into N P intervals, each of which is equal to the width of the azimuth reconstruction region ΔS. p The interval is expressed as [S p S p +ΔS],S p Indicates the first p The distance between the starting point of the interval to be reconstructed and the center of the footprint of the first scanning point.

[0158] Reconstruct the area based on coverage [S p S p +ΔS] the ground projection coordinates of the pulse observation point calculate The distance D from the center of the footprint of the starting observation point of the scanning observation segment (forward observation segment and rearward observation segment) along the scanning trajectory Azi,p .

[0159] According to the target resolution requirement of one-dimensional reconstruction in azimuth, the pth interval [S p S p +ΔS] is evenly divided into N Qintervals, each interval [R q R q +ΔR] represents the azimuth resolution unit with a size of ΔR, and the starting point of the interval is R q It indicates the distance from the starting point of the azimuth resolution unit to the center of the footprint of the starting observation point of the scanning observation segment (forward observation segment and rearward observation segment).

[0160] For each observation point Its normalized spatial response function RSF ij,n The corresponding relative distance is D Rsf,j , then RSF ij,n The distance from the starting point of the scanning observation segment (forward observation segment and rearward observation segment) is D Azi,p +D Rsf,j , search distance value D Azi,p +D Rsf,j Located in the interval [R q R q +ΔR] and average them as the spatial response function value of the p-th observation point falling into the q-th azimuth resolution unit.

[0161]

[0162] Ind1 and Ind2 represent the interval [R q R q +ΔR] internal distance value D Azi,p +D Rsf,j The start and end range of the serial number, Q p =Ind2-Ind1+1.

[0163] Observation point The corresponding backscatter coefficient Combined with spatial response function RSF ij,n The Scattermeter Image Reconstruction (SIR) algorithm can be used to achieve the interval [S p S p +ΔS] in each azimuth resolution unit [R q R q +ΔR] backscattering coefficient reconstruction. The SIR algorithm uses a numerical iteration method to reconstruct the azimuth resolution unit [R q R q +ΔR] backscattering coefficient, the specific reconstruction process is as follows:

[0164] a. The vector composed of the backscatter coefficients of each resolution unit in azimuth is S, s q Represents the scattering coefficient of the qth resolution unit. Initialize the vector to an arbitrary vector S 0 .

[0165] b. k+1th iteration

[0166]

[0167] in:

[0168] λ is the iterative tuning factor, h pq It is the average value of the normalized spatial response function of the pth original observation point involved in the reconstruction falling within the qth resolution unit.

[0169] In each iteration, the original observation data participates in updating the backscattering coefficient of the azimuth resolution unit.

[0170] c. Repeat step b. As the number of iterations increases, the difference between the results of the previous and subsequent iterations gradually decreases until convergence.

[0171] (4) Calculation of auxiliary information of geographic geometry of reconstructed data

[0172] Select the azimuth resolution unit [R q R q +ΔR] is taken as the center of the reconstructed scattering coefficient, that is, The Lon corresponding to this position q , Lat q 、Antenna scanning angle Scan q 、Geographical azimuth Azi q and the angle of incidence Eia q The corresponding information Lon of the original observation data involved in the reconstruction of the backscatter coefficient of the area can be used p , Lat p 、Antenna scanning angle Scan p 、Geographical azimuth Azi p and the angle of incidence Eia p Determined by interpolation.

[0173] Finally, steps (3) to (4) are repeated to traverse all range resolution units, scanning cycles, forward-looking and backward-looking observation data, and one-dimensional azimuth high-resolution reconstruction of all observation data is achieved. The processed backscatter coefficient has high resolution in both range and azimuth. Because the reconstruction process is performed along the antenna beam scanning footprint, the reconstructed data has the same observation geometry characteristics as the original data, and there is no need to separately geolocate the reconstruction area and calculate auxiliary information such as observation geometry, which improves data processing efficiency.

[0174] Embodiment 1:

[0175] The azimuth resolution enhancement method of spaceborne microwave scatterometer based on image reconstruction technology was tested and verified on a Windows system PC (129G memory, 3.1GHz main frequency, Core i7 processor, and hard disk space of more than 100G). Matlab software was used for data processing software code writing and test verification.

[0176] The original measured data of the spaceborne microwave scatterometer are used as input and processed according to the method described above. The effectiveness of the azimuth resolution enhancement method of the spaceborne microwave scatterometer based on image reconstruction technology is verified by comparing the changes in the backscattering coefficient distribution before and after each main processing step.

[0177] The scattering coefficient distribution of the original observation data of Greenland in some areas before and after high-resolution processing in the range direction and after one-dimensional high-resolution reconstruction in the azimuth direction is shown in the figure below. Figure 2 , 3 and 4: Figure 2 Before the distance is processed to high resolution, Figure 3 After the distance is processed to high resolution, Figure 4 After one-dimensional high-resolution reconstruction in azimuth; by comparing the backscatter coefficient distribution map of Greenland before, after and after high-resolution processing in range, it can be seen that after processing by the azimuth resolution enhancement method of the satellite-borne microwave scatterometer based on image reconstruction technology, the island boundary coastline is clearer and the detailed information of the local backscatter distribution of the island is more obvious.

[0178] The distribution of scattering coefficients before and after one-dimensional resolution enhancement processing of isolated islands near Antarctica is shown in the figure below. Figure 5 and 6 As shown: Figure 5 is the backscattering coefficient distribution before azimuth one-dimensional reconstruction, Figure 6 This is the distribution of backscattering coefficients after azimuth one-dimensional reconstruction. Comparing the changes in the distribution of backscattering coefficients of isolated islands before and after azimuth one-dimensional high-resolution reconstruction, it is shown that the scattering characteristics of isolated islands are more significant after azimuth one-dimensional high-resolution reconstruction, thus achieving the purpose of resolution enhancement.

[0179] The above data processing test results verify the effectiveness of the azimuth resolution enhancement method of the spaceborne microwave scatterometer based on image reconstruction technology in achieving spatial resolution enhancement of the backscattering coefficient.

[0180] The contents not described in detail in the specification of the present invention belong to the common knowledge of those skilled in the art.

[0181] Although the present invention has been described in detail above with general descriptions and specific embodiments, it is obvious to those skilled in the art that some modifications or improvements can be made to the present invention. Therefore, these modifications or improvements made without departing from the spirit of the present invention all belong to the protection scope of the present invention.

Claims

1. A method for enhancing azimuth resolution of a spaceborne microwave scatterometer based on image reconstruction technology, characterized in that: include: Step 1: High-resolution processing in range: After the range resolution unit is subdivided, the backscatter coefficient of the range resolution unit is calibrated and the geolocation processing of the range resolution unit is performed, the backscatter coefficient of the range resolution unit is obtained and the latitude Lat of the center of the footprint of each range resolution unit is determined. p and Lon p , and determine the corresponding ground incidence angle Eia p and geographic azimuth Azi p , and retain the antenna scanning angle Scan that reflects the forward and backward states of the observation data p ; Step 2: azimuth one-dimensional resolution enhancement processing: through azimuth one-dimensional reconstruction parameter calculation, original observation data reorganization, azimuth high-resolution reconstruction and reconstruction data geographic geometry auxiliary information; A backscatter coefficient with high resolution in both the range and azimuth directions is obtained, and at the same time, the longitude, latitude, ground incidence angle, geographic azimuth and antenna scanning angle information corresponding to the high-resolution backscatter coefficient is obtained.

2. The method for enhancing azimuth resolution of a spaceborne microwave scatterometer based on image reconstruction technology according to claim 1, characterized in that: Step 1: distance high-resolution processing specifically includes: The range beam footprint is divided into several resolution units according to the range resolution requirement. The resolution unit power is converted into backscatter coefficients through calibration based on the corresponding relationship between echo power slice distribution and range resolution unit and system parameters. At the same time, the observation position corresponding to the backscatter coefficient is determined through geographic positioning based on the observation geometry information. The number of measurement units that the scatterometer can provide in the range direction for each observation, that is, the number of range resolution units, is calculated by the range resolution unit subdivision step. Each range resolution unit corresponds to a power slice within a certain range in the echo measurement value. The power slice corresponding to each range resolution unit is accumulated and calibrated through the range resolution unit backscatter coefficient calibration step to determine the backscatter coefficient corresponding to the range resolution unit. Finally, the geographical location and other information corresponding to the backscatter coefficient of each range resolution unit are determined through the range resolution unit geographic positioning step, including latitude, longitude, ground incidence angle, geographical azimuth and antenna scanning angle, to provide input for the one-dimensional resolution enhancement processing in azimuth. The backscatter coefficient of each range resolution unit and the corresponding geographical location data are arranged in the order of the original scanning observation time of the scatterometer.

3. The method for enhancing azimuth resolution of a spaceborne microwave scatterometer based on image reconstruction technology according to claim 1 or 2, characterized in that: Step 2: azimuth one-dimensional resolution enhancement processing specifically includes: The surface coverage of the specific range resolution unit participating in the azimuth resolution enhancement processing along the scanning footprint direction is determined through the azimuth one-dimensional reconstruction parameter calculation step, and then the number of observations of the specific range resolution unit participating in the azimuth high-resolution reconstruction is determined each time; before the azimuth high-resolution reconstruction, the antenna scanning angle and the geographic azimuth are converted into angles, and then the range resolution unit backscatter coefficient, latitude, longitude, adjusted antenna scanning angle and geographic azimuth, and incident angle are reorganized in four dimensions according to the scanning start angle and the front and back sight marks within the scanning cycle as input for the azimuth high-resolution reconstruction processing step; the high-resolution reconstruction processing step is sequentially performed in the distance resolution unit dimension, the scanning footprint direction, and the scanning footprint direction. The input data is traversed in the scanning period dimension and the fore-and-aft dimension, and the backscatter coefficient of the range resolution unit is processed by the scatterometer image reconstruction algorithm based on the reconstruction parameters and the spatial response function information, so that the backscatter coefficient of each range resolution unit is enhanced in the azimuth direction; the latitude, longitude, antenna scanning angle and geographic azimuth information corresponding to the backscatter coefficient processed with the azimuth resolution enhancement are calculated from the latitude, longitude, adjusted antenna scanning angle and geographic azimuth of the range resolution unit through the reconstruction data geographic geometry auxiliary information calculation step, and finally the backscatter coefficient with enhanced resolution in both the range and azimuth directions, as well as the corresponding latitude, longitude, antenna scanning angle and geographic azimuth information are obtained.

4. The method for enhancing azimuth resolution of a spaceborne microwave scatterometer based on image reconstruction technology according to claim 1 or 2, characterized in that: The distance to high resolution processing includes: (1) Distance resolution unit subdivision Subdivide the ground footprint into a number of ground resolution units N Ele : S Ele ΔR is the ground footprint of the 3dB beam from the antenna. Ele The required resolution is for distance; (2) Calibration of backscatter coefficient of range resolution unit In the real-time processing of remote sensing data, the echo power slice accumulation range and illumination integral parameter I corresponding to each range resolution unit of each pulse observation are obtained by searching the power slice accumulation range and illumination integral parameter table based on the real-time sub-satellite point latitude and antenna scanning angle. 0r , calculate the echo power P of each pulse distance resolution unit 0r,p : r1 and r2 represent the start and end numbers of the echo power slices contained in the rth range resolution unit, respectively. i represents the echo power slice number. The above formula means that the echo power slices ranging from r1 to r2 are selected for accumulation. r = 1, 2…N Ele , p represents the scanning observation point number; Based on the internal calibration data, P 0r,p After noise reduction processing to obtain pure signal power P sr,p , combined with the illumination integration factor and radar system parameters, the range resolution unit backscatter coefficient is obtained X1 is a fixed parameter related to the system, and X2 is a parameter related to the calibration within the system; (3) Distance resolution unit geolocation Each pulse observation distance is geographically located in each resolution unit, and the center latitude of each distance resolution unit footprint is determined. p and Lon p , and determine the corresponding ground incidence angle Eia p and geographic azimuth Azi p , and retain the antenna scanning angle Scan that reflects the forward and backward states of the observation data p .

5. The method for enhancing azimuth resolution of a spaceborne microwave scatterometer based on image reconstruction technology according to claim 1 or 2, characterized in that: The one-dimensional resolution enhancement process in azimuth includes: (1) Calculation of one-dimensional reconstruction parameters in azimuth At a specific satellite position vector and velocity vector When the antenna beam center viewing angle and antenna beam center azimuth are θ0 and When the conversion matrix M from the earth fixed coordinate system to the antenna coordinate system is calculated EA : M EA =M Ecf 2 Ant M Ins (4); Among them, M Ecf 2 Ant is the transformation matrix related to the satellite position vector, velocity vector, beam center viewing angle and beam center azimuth, M Ins Represents the instrument installation offset matrix; Assume that the beam angle corresponding to each resolution unit is θ r ,r=1,2…N Ele , the azimuth of each resolution unit is consistent with the azimuth of the antenna beam center. Assume that the antenna beam width of the resolution unit azimuth is 2φ0, and the azimuth angle traversal range used by the resolution unit azimuth projection is The traversal interval is Δφ, Covering the first zero point position of the antenna azimuth pattern; according to the satellite position vector Velocity Vector Radar instrument installation matrix, resolution unit beam viewing angle θ r , resolution unit projection traversal azimuth j=1,2…M, M is the number of azimuth subdivisions of the resolution unit, and the ground projection point corresponding to the distance resolution unit is calculated. from arrive The connecting vector is: Vector The modulus of is the distance from the satellite to the projection point on the ground: Use the transformation matrix M from the earth fixed coordinate system to the antenna coordinate system EA Will Convert to antenna coordinate system; Depend on Calculated vector Elevation angle and azimuth angle in antenna coordinate system: By looking up the table, it can be determined The corresponding antenna two-way gain G rj , combined with the slope distance R rj Get the spatial response function RSF of each range resolution unit rj : About RSF rj according to Normalized to get the normalized spatial response function RSF rj,n : Correspondingly, the ground projection point coordinates of the distance resolution unit Calculate the orientation of each projection point relative to the central projection point The distance D Rsf,rj , the relative distance D of the center projection point Rsf,cen = 0; through the sign conversion will correspond to the azimuth Part D Rsf,rj The space function is converted to a negative value to facilitate azimuth reconstruction, that is: When the microwave scatterometer is working normally, the satellite position corresponding to different observation points changes as the satellite platform moves. When the conical scanning observation method is used, the satellite position corresponding to any fixed observation point Other scanning observation points and (m=nM,n-M+1…n+M,m≠n) The connecting line vector is: n represents the serial number of any fixed observation point, and m represents the serial number of 2M observation points adjacent to the observation point before and after: The corresponding beam viewing angle is: The corresponding beam angle α mn The change should be controlled within each distance resolution viewing angle range Δα r , and thus determine the observation points involved in the reconstruction Among the 2M observation points adjacent to each other in time, there are M0 observation points that meet the condition that the coverage area is located in the same range resolution unit. The width ΔS of the effective reconstruction area in azimuth is: ΔS=M0ΔP (15); ΔP is the ground surface distance between the centers of adjacent pulse beam footprints along the scanning direction; (2) Reconstruction of original observation data Adjust the raw data antenna scanning angle and geographic azimuth to within ±180°: Scan p and Azi p They represent the scanning azimuth and the geographical azimuth of the center of the beam footprint corresponding to the backscattering coefficient obtained after high-resolution processing in the range direction, and p represents the serial number of the scanning observation point; Calculate the forward and backward look marks for each scan observation based on the antenna scan angle: The backscatter coefficient, latitude, longitude, adjusted antenna scanning angle and geographic azimuth, incident angle, and fore-aft and aft-aft identification of each range resolution unit obtained through range high-resolution processing are arranged according to the periodic scanning observation sequence of the scatterometer; The data obtained by range-direction high-resolution processing are reorganized according to the following method: the forward-sight scanning azimuth angle closest to -90° in each scanning cycle is used as the scanning start angle, and the backscatter coefficient, latitude, longitude, adjusted antenna scanning angle and geographic azimuth, incident angle, forward and backward sight marks are periodically arranged according to the scanning start angle value, and the data is divided into forward-sight state data and backward-sight state data according to the forward-sight and backward-sight marks in each cycle; The reorganized data is divided into four dimensions. The first dimension corresponds to different distance resolution units, the second dimension corresponds to different scanning cycles, the third dimension corresponds to front and rear vision, and the fourth dimension corresponds to different observation points in front or rear vision. (3) Azimuth high-resolution reconstruction Select a specific distance resolution unit and forward or backward observation data within the scanning cycle, and divide the continuous scanning observation point beam footprint coverage evenly into N P intervals, each interval length is the width of the azimuth reconstruction region ΔS, and the Pth interval is expressed as [S P S P +ΔS], S P represents the distance between the starting point of the Pth interval to be reconstructed and the center of the footprint of the first scanning point along the scanning trajectory; Reconstruct the area based on coverage [S P S P +ΔS] scanning observation point ground projection coordinates calculate The distance D from the center of the footprint relative to the starting observation point of the scanning observation segment along the scanning trajectory Azi,p ; According to the target resolution requirement of one-dimensional reconstruction in azimuth, the Pth interval to be reconstructed [S P S P +ΔS] is evenly divided into N Q intervals, each interval [R q R q +ΔR] represents the azimuth resolution unit with a size of ΔR, and the starting point of the interval is R q Indicates the distance between the starting point of the azimuth resolution unit and the center of the footprint of the starting observation point of the scanning observation segment; For each observation point Its normalized spatial response function RSF ij,n The corresponding relative distance is D Rsf,j , then RSF ij,n The distance from the starting point of the scanning observation segment is D Azi,p +D Rsf,j , search distance value D Azi,p +D Rsf,j Located in the interval [R q R q +ΔR] and average them as the spatial response function value of the p-th observation point falling into the q-th azimuth resolution unit; Ind1 and Ind2 represent the interval [R q R q +ΔR] internal distance value D Azi,p +D Rsf,j The start and end range of the serial number, Q p =Ind2-Ind1+1; Observation point The corresponding backscatter coefficient Combined with spatial response function RSF ij,n Application of Scatterometer Image Reconstruction Algorithm SIR to Realize Interval [S P S P +ΔS] in each azimuth resolution unit [R q R q +ΔR] backscatter coefficient reconstruction; (4) Calculation of auxiliary information of geographic geometry of reconstructed data Select the azimuth resolution unit [R q R q +ΔR] is taken as the center of the reconstructed scattering coefficient, that is, The Lon corresponding to this position q , Lat q 、Antenna scanning angle Scan q 、Geographical azimuth Azi q and the angle of incidence Eia q The corresponding information Lon of the original observation data involved in the reconstruction of the backscatter coefficient of the area can be used p , Lat p 、Antenna scanning angle Scan p 、Geographical azimuth Azi p and the angle of incidence Eia p Determined by interpolation; Finally, steps (3) to (4) are repeated to traverse all distance resolution units, scanning cycles, forward-looking and backward-looking observation data to achieve one-dimensional azimuth high-resolution reconstruction of all observation data. The processed backscatter coefficient has high resolution in both distance and azimuth.

6. The method for enhancing azimuth resolution of a spaceborne microwave scatterometer based on image reconstruction technology according to claim 5, characterized in that: The scatterometer image reconstruction algorithm SIR achieves the interval [S p S p +ΔS] in each azimuth resolution unit [R q R q +ΔR] backscatter coefficient reconstruction, the specific reconstruction process is as follows: a. The vector composed of the backscatter coefficients of each resolution unit in azimuth is S, s q represents the scattering coefficient of the qth resolution unit, and initializes the vector to an arbitrary vector S 0 ; b. k+1th iteration: in: λ is the iterative tuning factor, h pq is the average value of the normalized spatial response function of the pth original observation point involved in the reconstruction falling within the qth resolution unit; In each iteration, the original observation data participate in the update of the backscattering coefficient of the azimuth resolution unit; c. Repeat step b. As the number of iterations increases, the difference between the results of the previous and subsequent iterations gradually decreases until convergence.

Images