A SAR motion compensation method and device in large dynamic scenes

Abstract

This application relates to a method and apparatus for SAR motion compensation in highly dynamic scenarios, pertaining to the field of radar imaging. The invention addresses the problem that existing technologies cannot meet the requirements of missiles for SAR imaging motion compensation under highly maneuverable conditions and highly dynamic scenarios. The method, executed by a missile, includes: obtaining sub-aperture data corresponding to a SAR image; determining, based on the sub-aperture data, the number of strong points in the image, the position information of the strong points in the image, and the current entropy value of the image; determining a predicted entropy value for the image based on the position information of the strong points in the image; determining a modulation frequency for motion compensation based on the number of strong points, the predicted entropy value, and the current entropy value; and performing motion compensation using the determined modulation frequency. The technical solution provided by the application can achieve accurate motion compensation in various shooting scenarios and ensure the timeliness of motion compensation.

Description

Technical Field

[0001] The present invention relates to the field of radar imaging, and is a SAR motion compensation method and device in a large dynamic scene. Background Art

[0002] Synthetic Aperture Radar (SAR) has the characteristics of all-day, all-weather and high-resolution imaging. With the development of SAR imaging technology, missile-borne SAR, which combines SAR with missile scene matching and positioning, has become a research hotspot in recent years.

[0003] The probability of a missile detecting and recognizing a target is closely related to the focusing effect of the SAR image. Since it is difficult for a missile to maintain uniformly accelerated linear motion during flight, its maneuverability in flight can cause envelope and phase distortion in the echo signal, affecting the precise focusing of the radar image. Therefore, it is necessary to extract high-precision missile motion parameters for motion error compensation.

[0004] The missile's flight path will pass through a variety of terrain, but existing compensation methods are usually only applicable to a single terrain or a single flight state (such as constant-speed flight). Therefore, they cannot meet the missile's requirements for SAR imaging motion compensation under large maneuvering conditions and large dynamic scenes. Summary of the Invention

[0005] In view of the above analysis, this application aims to propose a SAR motion compensation method and device in a large dynamic scene, so as to realize motion compensation of the missile according to the imaging scene of the flight process.

[0006] The purpose of this application is mainly achieved through the following technical solutions:

[0007] On the one hand, the present application provides a SAR motion compensation method in a large dynamic scene, which is executed by a missile, comprising:

[0008] Obtain sub-aperture data corresponding to the SAR image;

[0009] Determining the number of strong points in the image, strong point position information in the image, and a current entropy value of the image based on the sub-aperture data;

[0010] Determining a predicted entropy value of the image based on strong point position information in the image;

[0011] determining a modulation rate for motion compensation based on the number of strong points, the predicted entropy value, and the current entropy value;

[0012] Motion compensation is performed using the determined modulation rate.

[0013] Furthermore, the sub-aperture data is a two-dimensional data matrix, including: range data and azimuth data; the strong points include range strong points and contrast strong points.

[0014] Further, determining the number of range-oriented strong points in the image and the position information of the range-oriented strong points in the image based on the sub-aperture data includes:

[0015] performing pulse compression, time domain correction, range walk correction, and range curvature correction on the range data in sequence to obtain processed sub-aperture data;

[0016] Superimposing the processed sub-aperture data along the azimuth direction to obtain one-dimensional range data;

[0017] Determining the amplitude mean and amplitude variance of the one-dimensional distance data;

[0018] The number of the range-direction strong points and their position information in the image are determined according to the amplitude mean and the amplitude variance.

[0019] Furthermore, determining the number of contrast-intensive points in the image and the position information of the contrast-intensive points in the image based on the sub-aperture data includes:

[0020] performing azimuth compression on the azimuth data to obtain compressed azimuth data;

[0021] Determining the amplitude mean and amplitude variance of the compressed azimuth data;

[0022] determining the contrast intensity of the range data according to the amplitude mean and the amplitude variance;

[0023] determining a mean and a variance of the contrast intensity of the range data;

[0024] The number of contrast-strong points and their position information in the image are determined according to the mean and variance of the contrast intensity.

[0025] Furthermore, determining the current entropy value of the image based on the sub-aperture data includes:

[0026] performing range compression on the azimuth data;

[0027] determining the probability of occurrence of each pixel in the image based on the compressed azimuth data;

[0028] The current entropy value is obtained according to the probability of occurrence of each pixel in the image.

[0029] Furthermore, the predicted entropy value includes an MD entropy value and a contrast entropy value.

[0030] Furthermore, the MD entropy value of the image is determined based on the contrast strength point, including:

[0031] Determining first target azimuth data according to the position information of the strong point;

[0032] performing azimuth compression on the first target azimuth data to obtain first compressed azimuth data;

[0033] dividing the first compressed azimuth data into two parts according to the number of matrix elements of the sub-aperture data in the azimuth direction;

[0034] Determining a translation amount of the two portions of data based on Fourier transform, normalization, and cross-correlation;

[0035] determining the MD modulation frequency according to the translation amount;

[0036] Determining an MD filter coefficient according to the MD modulation frequency;

[0037] performing azimuth compression on the first target azimuth data based on the MD filter coefficient;

[0038] Obtaining an MD entropy value based on the re-compressed first target azimuth data; and determining a contrast entropy value of the image, including:

[0039] Determining the second target azimuth data according to the position information of the contrast-strong point;

[0040] performing azimuth compression on the second target azimuth data to obtain second compressed azimuth data;

[0041] Determining a frequency modulation search sequence according to the MD frequency modulation rate and a preset step size;

[0042] Constructing a filter function with the frequency modulation search sequence as an independent variable set;

[0043] filtering the second compressed azimuth data according to the filtering function;

[0044] Determining the amplitude mean and amplitude variance of the second compressed azimuth data after filtering;

[0045] determining, according to the amplitude mean and amplitude variance of the second compressed azimuth data, a contrast corresponding to each modulation rate in the modulation rate search sequence;

[0046] Determine the modulation frequency corresponding to the maximum contrast as the contrast modulation frequency;

[0047] determining a contrast filter coefficient according to the contrast modulation frequency;

[0048] Based on the contrast filter coefficient, the second target azimuth data is azimuthally compressed; and a contrast entropy value is obtained according to the re-compressed second target azimuth data.

[0049] Furthermore, determining an optimal method for motion compensation based on the number of strong points, the predicted entropy value, and the current entropy value includes:

[0050] When the number of the contrast strong points is greater than a first threshold and the contrast entropy value is less than the current entropy value, determining the sum of the MD modulation frequency and the contrast modulation frequency as the modulation frequency;

[0051] When the distance to the strong point is greater than a second threshold, the number of the contrast strong points is less than a first threshold, and the MD entropy value is less than the current entropy value, determining the MD modulation frequency to be the modulation frequency;

[0052] Otherwise, the modulation rate is determined using inertial navigation data.

[0053] On the other hand, the present application also provides a SAR motion compensation device in a large dynamic scene, comprising: an acquisition module, a data processing module and a motion compensation module;

[0054] The acquisition module is used to acquire sub-aperture data corresponding to the image;

[0055] The data processing module is used to determine the number of strong points in the image, the position information of the strong points in the image, and the current entropy value of the image based on the sub-aperture data; determine the predicted entropy value of the image based on the strong point position information in the image; and determine the modulation rate for motion compensation based on the number of strong points, the predicted entropy value, and the current entropy value;

[0056] The motion compensation module is configured to perform motion compensation using the determined modulation rate.

[0057] Furthermore, the sub-aperture data is a two-dimensional data matrix, including: range data and azimuth data;

[0058] The strong point includes at least one of a distance strong point and a contrast strong point;

[0059] When the strong points include range-direction strong points, the data processing module is configured to sequentially perform pulse compression, time domain correction, range walk correction, and range curvature correction on the range data; superimpose the range data along the azimuth direction to obtain one-dimensional range data; determine an amplitude mean and an amplitude variance of the one-dimensional range data; and determine the number of the range-direction strong points and their position information in the image based on the amplitude mean and the amplitude variance;

[0060] When the strong points include contrast-strong points; the data processing module is used to perform azimuth compression on the sub-aperture data to obtain compressed azimuth data; determine the amplitude mean and amplitude variance of the compressed azimuth data; determine the contrast intensity of the range data based on the amplitude mean and amplitude variance; determine the mean and variance of the contrast intensity of the range data; and determine the number of contrast-strong points and their position information in the image based on the mean and variance of the contrast intensity.

[0061] Compared with the existing technology, this application can achieve at least one of the following technical effects:

[0062] 1. Classify the shooting scenes of radar images based on the two dimensions of strong points and image entropy, and select the corresponding modulation rate under the corresponding shooting scene to facilitate subsequent accurate motion compensation.

[0063] 2. Based on the MD algorithm, contrast algorithm and inertial navigation data, motion compensation is achieved solely with the missile's own resources and data, ensuring the timeliness of motion compensation.

[0064] Other features and advantages of the present application will be described in the subsequent description, and some will become apparent from the description or be understood by practicing the present application. The purpose and other advantages of the present application can be realized and obtained by the structures particularly pointed out in the written description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] The accompanying drawings are only for the purpose of illustrating particular embodiments and are not to be considered as limiting the present application. Like reference symbols denote like components throughout the drawings.

[0066] Figure 1 A flowchart of a SAR motion compensation method in a large dynamic scene provided by an embodiment of the present application;

[0067] Figure 2 This is a schematic diagram of the structure of a SAR motion compensation device in a large dynamic scene provided by an embodiment of the present application. DETAILED DESCRIPTION

[0068] The preferred embodiments of the present application are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of the present application and are used together with the embodiments of the present application to illustrate the principles of the present application, and are not used to limit the scope of the present application.

[0069] High-maneuverability guided missiles cannot maintain a constant speed during flight, so the missile needs to perform motion compensation on the captured images.

[0070] The flight paths of guided missiles in highly dynamic scenarios often need to be temporarily set or even constantly changed. Therefore, the scene being filmed cannot be predicted in advance, and is therefore dynamic. In contrast, other aircraft, such as airplanes, have stable flight paths, making their scenes predictable. Therefore, missiles face more complex scenes than other aircraft. Complex scenes make it difficult to ensure effective motion compensation. For example, when passing through cities and villages, the difference between images captured between adjacent capture cycles is often significant, facilitating motion compensation. However, in scenes such as deserts, grasslands, and Gobi deserts, the difference between images captured between adjacent capture cycles is often less pronounced, making motion compensation less effective.

[0071] Furthermore, missiles are less easily steered than other aircraft, making them more susceptible to deviation. To compensate for this drawback, missiles require rapid, accurate positioning during flight. For example, aircraft can manually steer their trajectory, allowing them to transmit image data to a server for high-quality image processing. This process takes time, which can be sufficient to cause irreversible deviations in the missile's trajectory, leading to unpredictable outcomes. Therefore, missiles place high demands on the timeliness of motion compensation.

[0072] In summary, when performing motion compensation on missile SAR imaging, it is necessary to consider the shooting scene and the timeliness of the information.

[0073] The key to SAR imaging is the selection of strong points in the scene. Strong points are the objects being photographed. Based on the principles of SAR imaging, strong points usually have at least one of the following characteristics:

[0074] 1. In the shooting scene, the distance between the subject and other objects makes the subject easy to identify, such as individual plants, animals and rocks in the desert or Gobi, and sculptures in the square in the city.

[0075] 2. The shape, appearance, or other attributes of the subject in the scene create a sharp contrast with its surroundings. For example, the color and length of an airport runway contrast sharply with the surrounding environment, or the length of a river contrasts sharply with the objects on both sides.

[0076] Different shooting scenes contain different numbers, locations, and characteristics of strong points, all of which affect the effectiveness of motion compensation. Existing compensation methods are typically only applicable to a single terrain or a single flight state (such as constant speed flight), and therefore are not well suited for missile motion compensation.

[0077] Based on the above theoretical and technical problems, the embodiment of the present application provides a SAR motion compensation method in a large dynamic scene, including the following steps:

[0078] Step 1: Obtain sub-aperture data corresponding to the SAR image;

[0079] In the embodiment of the present application, the sub-aperture data is specifically SAR echo data, and the image is a SAR radar image. The sub-aperture data is a two-dimensional data matrix including range data and azimuth data.

[0080] Step 2: Determine the number of strong points in the image, the position information of the strong points in the image, and the current entropy value of the image based on the sub-aperture data.

[0081] In the embodiment of the present application, based on the characteristics of the aforementioned strong points, the strong points in the image include: range strong points and contrast strong points;

[0082] Methods for determining the number of strong points and strong point location information in the direction of strong points include:

[0083] The range data are sequentially processed with pulse compression, time domain correction, range walk correction, and range curvature correction. The sub-aperture data are superimposed along the azimuth direction to obtain one-dimensional range data. The amplitude mean and amplitude variance of the one-dimensional range data are determined. Based on the amplitude mean and amplitude variance, the number of range strong points and their position information in the image are determined.

[0084] Specifically, 1) for the sub-aperture data x(t r ,t m ) in the range direction data are sequentially subjected to pulse compression, time domain correction, range movement correction, and range bending correction to obtain the processed sub-aperture data x(t r ,t m ); where t r Representing distance data, t m Representing azimuth data;

[0085] 2) For the processed sub-aperture data x(t r ,t m ), and superimpose along the azimuth direction to obtain the one-dimensional distance data x1(t r ), calculate x1(t r )'s amplitude mean x 1_mean , amplitude variance x 1_std , extract x1(t r ) is greater than x 1_mean +3 x 1_std The point is the distance strong point, and then the number of distance strong points N is obtained. r_q , and the location of the strong point Index_MD.

[0086] The method for determining the number of strong points and the position information of the strong points of contrast includes:

[0087] Azimuth compression is performed on the azimuth data to obtain compressed azimuth data; the amplitude mean and amplitude variance of the compressed azimuth data are determined; the contrast intensity of the range data is determined based on the amplitude mean and amplitude variance; the mean and variance of the contrast intensity of the range data are determined; and the number of contrast-strong points and their position information in the image are determined based on the mean and variance of the contrast intensity.

[0088] Specifically, construct the azimuth dechirp processing factor, in

[0089]

[0090] R is the distance unit coordinate, v is the carrier and speed The spatial squint angle θ is the Doppler center frequency f estimated by dc The obtained value, f dc =2vsinθ0 / λ, λ is the signal wavelength;

[0091] The dechirp processing factor is multiplied with the azimuth data to achieve azimuth compression of the azimuth data and obtain the compressed azimuth data x2(t m );

[0092] Calculate x2(t m )'s amplitude mean x 2_mean , amplitude variance x 2_std , and obtain the contrast of the compressed azimuth data

[0093] Calculate Q db (t r ) db_mean , variance Q db_std , extract Q db (t r ) is greater than Q db_mean +3·Q db_std The point with strong contrast is the point with strong contrast, and then the number of points with strong contrast N is obtained. Qdb And the location of the strong point Index_DB.

[0094] The process of determining the current entropy value includes:

[0095] The azimuth data is compressed in the distance direction; the probability of each pixel appearing in the image is determined based on the compressed azimuth data; and the current entropy value is obtained based on the probability of each pixel appearing in the image.

[0096] Specifically, the azimuth data is taken and dechirped along the range direction to achieve range compression;

[0097] For an M×N radar image, its pixels are {x m,n}, where m is the azimuth number and n is the distance number; let the probability of pixel occurrence be in The current entropy value is:

[0098]

[0099] Step 3: Based on the strong point position information in the image, determine the predicted entropy value of the image.

[0100] In the embodiment of the present application, the predicted entropy value includes: MC entropy value and contrast entropy value; MC is the abbreviation of sub-aperture correlation method.

[0101] Strong points in the image include: range-oriented strong points;

[0102] The method for determining an MD entropy value includes: determining first target azimuth data based on position information of a strong point; performing azimuth compression on the first target azimuth data to obtain first compressed azimuth data; dividing the first compressed azimuth data into two parts based on the number of matrix elements of sub-aperture data in the azimuth direction; determining a translation amount of the two parts of data based on Fourier transform, normalization, and cross-correlation; determining an MD frequency modulation rate based on the translation amount; determining an MD filter coefficient based on the MD frequency modulation rate; and performing azimuth compression on the target azimuth data based on the MD filter coefficient to obtain an MD entropy value.

[0103] Specifically, 1) according to the position information Index_MD and Index_DB of the strong point, determine the first target direction data x(t m ) is multiplied by the azimuth matching filter coefficient to achieve azimuth compression and obtain the first compressed azimuth data y(t m ):

[0104]

[0105] 2) Divide the first compressed azimuth data into two partial sub-apertures, the data is y1(t m )=y(t m1 ) and y2(t m )=y(t m2 ), the value range of m1 is (1, Na / 2), and the value range of m2 is (Na / 2+1, Na), where Na is the number of elements corresponding to the azimuth direction.

[0106] 3) For y1(t m ) and y2(t m ) are Fourier transformed to the azimuth frequency domain y1(f m ) and y2(f m );

[0107] 4) For y1(fm ) and y2(f m ) Perform normalization processing and modulus value;

[0108] 5) Use the correlation function to find y1(f m ) and y2(f m ) translation,

[0109] z=(FFT(y2(f m )·conj(FFT(y1(f m ))), according to z, determine the amplitude peak position, which is the translation Δn a ;

[0110] 6) Using Δn a The estimated MD modulation frequency is calculated. The corresponding MD filter coefficient is Where PRF is the repetition frequency of SAR.

[0111] 7) First target direction data x(t m ) and MD filter coefficient Multiply them together and perform dechirp processing, that is, perform azimuthal compression;

[0112] 8) Calculate the MD entropy value S_md based on the compressed first target azimuth data.

[0113] The method for determining the contrast entropy value includes:

[0114] According to the position information of the contrast strong point, the second target azimuth data is determined; the second target azimuth data is azimuthally compressed to obtain second compressed azimuth data; according to the MD modulation frequency and the preset step size, a modulation frequency search sequence is determined; a filter function is constructed with the modulation frequency search sequence as an independent variable set; the second compressed azimuth data is filtered according to the filter function; the amplitude mean and amplitude variance of the filtered second compressed azimuth data are determined; according to the amplitude mean and amplitude variance of the filtered second compressed azimuth data, the contrast corresponding to each modulation frequency in the modulation frequency search sequence is determined; the modulation frequency corresponding to the maximum contrast is determined as the contrast modulation frequency; according to the contrast modulation frequency, a contrast filter coefficient is determined; based on the contrast filter coefficient, the second target azimuth data is azimuthally compressed; and a contrast entropy value is obtained according to the compressed azimuth data.

[0115] Specifically,

[0116] 1) Based on the contrast strong point position Index_DB, determine the second target azimuth data. The second target azimuth data z(t m ) is multiplied by the MD filter coefficient to achieve azimuth compression and obtain the second compressed azimuth data s(t m ):

[0117]

[0118] 2) In a certain step Obtain frequency modulation search sequence ΔK = [-2Δk:2Δk];

[0119] 3) Construct a filter function based on ΔK as the independent variable set use With s(t m ) performs matched filtering to determine the amplitude mean I of the second compressed azimuth data after filtering mean , amplitude variance I std , get the contrast of the second compressed azimuth data after filtering Among them, Q db_jj There is a one-to-one correspondence with each modulation frequency of ΔK.

[0120] 4) Determine the modulation frequency corresponding to the maximum contrast as the contrast modulation frequency ΔK DB ,max(Q db_jj ) corresponding to ΔK i =ΔK DB , then the contrast filter coefficient is

[0121] 5) Second target direction data x(t m ) is multiplied by the contrast filter coefficient to perform dechirp processing, i.e. azimuth compression, and the contrast entropy value S_db is calculated based on the compressed azimuth data.

[0122] Step 4: Determine the modulation rate for motion compensation based on the number of strong points, the predicted entropy value, and the current entropy value.

[0123] In an embodiment of the present application, a method for determining a modulation rate for motion compensation includes:

[0124] When the number of contrast-strong points is greater than the first threshold and the contrast entropy is less than the current entropy, the modulation rate is the sum of the MD modulation rate and the contrast modulation rate;

[0125] When the distance to the strong point is greater than the second threshold, the number of contrast strong points is less than the first threshold, and the MD entropy value is less than the current entropy value, the modulation frequency is the MD modulation frequency;

[0126] In other cases, the inertial navigation data is used to determine the modulation rate.

[0127] The first threshold is the frequency modulation threshold estimated by the MD method, and the second threshold is the frequency modulation threshold estimated by the contrast method.

[0128] Use inertial navigation data to determine the modulation rate, specifically:

[0129]

[0130] ΔK a is the frequency modulation, a r_i is the inertial navigation acceleration during the echo collection process corresponding to each sub-aperture data, and λ is the signal wavelength.

[0131] Step 5: Perform motion compensation using the determined modulation rate.

[0132] In the embodiment of the present application, after the modulation rate is determined, motion compensation can be implemented using technical means available to those skilled in the art.

[0133] In the embodiment of the present application, the influence of the current shooting scene on motion compensation is evaluated based on the strong points and image entropy, and the influence is divided into three cases according to the evaluation results.

[0134] In the first case, there are enough points with strong contrast. During motion compensation, the points with strong contrast can also be used as distance strong points. In this case, the sum of the MD modulation frequency and the contrast modulation frequency is taken as the modulation frequency.

[0135] In the second case, the number of strong contrast points is insufficient, but the number of strong distance points is sufficient. Due to the insufficient number of strong contrast points, the obtained contrast modulation frequency may have a large error, so only the MD modulation frequency is used as the modulation frequency.

[0136] In the third case, the number of strong contrast points and the number of strong range points in other scenarios are insufficient, and there are huge errors in the MD modulation frequency and the contrast modulation frequency. In this case, the inertial navigation data is used to determine the modulation frequency.

[0137] Through the above method, the missile can distinguish the shooting scenes without relying on external forces (such as servers) and select the corresponding frequency modulation for motion compensation, taking into account the timeliness of the data and performing motion compensation more accurately.

[0138] The embodiment of the present application provides a SAR motion compensation device in a large dynamic scene, including: an acquisition module 201, a data processing module 202 and a motion compensation module 203;

[0139] The acquisition module 201 is used to acquire sub-aperture data corresponding to the image;

[0140] The data processing module 202 is configured to determine the number of strong points in the image, the position information of the strong points in the image, and the current entropy value of the image based on the sub-aperture data; determine a predicted entropy value of the image based on the position information of the strong points in the image; and determine a modulation rate for motion compensation based on the number of strong points, the predicted entropy value, and the current entropy value.

[0141] The motion compensation module 203 is configured to perform motion compensation using the determined modulation rate.

[0142] In the embodiment of the present application, the strong points in the image include: range-wise strong points;

[0143] The sub-aperture data is a two-dimensional data matrix, including: range data and azimuth data;

[0144] The data processing module 202 is used to perform pulse compression, time domain correction, range walk correction and range curvature correction on the range data in sequence; superimpose the range data along the azimuth direction to obtain one-dimensional range data; determine the amplitude mean and amplitude variance of the one-dimensional range data; and determine the number of range strong points and their position information in the image based on the amplitude mean and the amplitude variance.

[0145] In the embodiment of the present application, the strong points in the image include: strong contrast points;

[0146] The sub-aperture data is a two-dimensional data matrix, including: range data and azimuth data;

[0147] The data processing module 202 is used to perform azimuth compression on the sub-aperture data to obtain compressed azimuth data; determine the amplitude mean and amplitude variance of the compressed azimuth data; determine the contrast intensity of the range data based on the amplitude mean and amplitude variance; determine the mean and variance of the contrast intensity of the range data; and determine the number of contrast-strong points and their position information in the image based on the mean and variance of the contrast intensity.

[0148] The above is only a preferred specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any changes or replacements that can be easily thought of by any technician familiar with this technical field within the technical scope disclosed in this application should be covered by the scope of protection of the present application.

Claims

1. A SAR motion compensation method in a large dynamic scene, characterized by: Executed by missiles, including: Obtain sub-aperture data corresponding to the SAR image; Determining the number of strong points in the image, strong point position information in the image, and a current entropy value of the image based on the sub-aperture data; Determining a predicted entropy value of the image based on strong point position information in the image; determining a modulation rate for motion compensation based on the number of strong points, the predicted entropy value, and the current entropy value; performing motion compensation using the determined modulation rate; The sub-aperture data is a two-dimensional data matrix, including: range data and azimuth data; the strong points include range strong points and contrast strong points; Determining the number of range-oriented strong points in the image and position information of the range-oriented strong points in the image according to the sub-aperture data includes: performing pulse compression, time domain correction, range walk correction, and range curvature correction on the range data in sequence to obtain processed sub-aperture data; Superimposing the processed sub-aperture data along the azimuth direction to obtain one-dimensional range data; Determining the amplitude mean and amplitude variance of the one-dimensional distance data; Determining the number of the range-direction strong points and their position information in the image according to the amplitude mean and the amplitude variance; Determining the number of contrast-intensive points in the image and position information of the contrast-intensive points in the image based on the sub-aperture data includes: performing azimuth compression on the azimuth data to obtain compressed azimuth data; Determining the amplitude mean and amplitude variance of the compressed azimuth data; determining the contrast intensity of the range data according to the amplitude mean and the amplitude variance; determining a mean and a variance of the contrast intensity of the range data; The number of contrast-strong points and their position information in the image are determined according to the mean and variance of the contrast intensity.

2. The method according to claim 1, characterized in that Determining the current entropy value of the image according to the sub-aperture data includes: performing range compression on the azimuth data; determining the probability of occurrence of each pixel in the image based on the compressed azimuth data; The current entropy value is obtained according to the probability of occurrence of each pixel in the image.

3. The method according to claim 1, characterized in that The predicted entropy value includes an MD entropy value and a contrast entropy value.

4. The method according to claim 3, characterized in that The strong points determine the MD entropy value of the image, including: Determining first target azimuth data according to the position information of the range strong point; performing azimuth compression on the first target azimuth data to obtain first compressed azimuth data; dividing the first compressed azimuth data into two parts according to the number of matrix elements of the sub-aperture data in the azimuth direction; determining a translation amount of the data of the two parts based on Fourier transform, normalization, and cross-correlation; determining the MD modulation frequency according to the translation amount; Determining an MD filter coefficient according to the MD modulation frequency; performing azimuth compression on the first target azimuth data based on the MD filter coefficient; Obtaining an MD entropy value based on the re-compressed first target azimuth data; and determining a contrast entropy value of the image, including: Determining the second target azimuth data according to the position information of the contrast-strong point; performing azimuth compression on the second target azimuth data to obtain second compressed azimuth data; Determining a frequency modulation search sequence according to the MD frequency modulation rate and a preset step size; Constructing a filter function with the frequency modulation search sequence as an independent variable set; filtering the second compressed azimuth data according to the filtering function; Determining the amplitude mean and amplitude variance of the second compressed azimuth data after filtering; determining, according to the amplitude mean and amplitude variance of the second compressed azimuth data, a contrast corresponding to each modulation rate in the modulation rate search sequence; Determine the modulation frequency corresponding to the maximum contrast as the contrast modulation frequency; determining a contrast filter coefficient according to the contrast modulation frequency; Based on the contrast filter coefficient, the second target azimuth data is azimuthally compressed; and a contrast entropy value is obtained according to the re-compressed second target azimuth data.

5. The method according to claim 4, characterized in that: The determining of an optimal method for motion compensation based on the number of strong points, the predicted entropy value, and the current entropy value comprises: When the number of the contrast strong points is greater than a first threshold and the contrast entropy value is less than the current entropy value, determining the sum of the MD modulation frequency and the contrast modulation frequency as the modulation frequency; When the distance to the strong point is greater than a second threshold, the number of the contrast strong points is less than a first threshold, and the MD entropy value is less than the current entropy value, determining the MD modulation frequency to be the modulation frequency; Otherwise, the modulation rate is determined using inertial navigation data.

6. A SAR motion compensation device in a large dynamic scene, characterized in that: include: Acquisition module, data processing module and motion compensation module; The acquisition module is used to acquire sub-aperture data corresponding to the image; The data processing module is used to determine the number of strong points in the image, the position information of the strong points in the image and the current entropy value of the image based on the sub-aperture data; and determine the predicted entropy value of the image based on the position information of the strong points in the image; determining a modulation rate for motion compensation based on the number of strong points, the predicted entropy value, and the current entropy value; The motion compensation module is used to perform motion compensation using the determined modulation rate; The sub-aperture data is a two-dimensional data matrix, including: range data and azimuth data; The strong points include distance strong points and contrast strong points; When the strong points include range-direction strong points, the data processing module is configured to sequentially perform pulse compression, time domain correction, range walk correction, and range curvature correction on the range data; superimpose the range data along the azimuth direction to obtain one-dimensional range data; determine an amplitude mean and an amplitude variance of the one-dimensional range data; and determine the number of the range-direction strong points and their position information in the image based on the amplitude mean and the amplitude variance; When the strong points include contrast-strong points; the data processing module is used to perform azimuth compression on the sub-aperture data to obtain compressed azimuth data; determine the amplitude mean and amplitude variance of the compressed azimuth data; determine the contrast intensity of the range data based on the amplitude mean and amplitude variance; determine the mean and variance of the contrast intensity of the range data; and determine the number of contrast-strong points and their position information in the image based on the mean and variance of the contrast intensity.

Images