Target isar image evaluation method based on signal processing and image change

Abstract

The present application relates to the target ISAR image evaluation method based on signal processing and image change, belongs to ISAR image processing field. In view of the ISAR image imaging fuzzy under the condition of target posture change on ISAR image, feature extraction is difficult, leading to the problem of target threat assessment difficulty, a kind of target ISAR image evaluation method based on signal processing and image change is presented. Including: based on the rearrangement and elimination method of micro-motion to realize the de-micro-motion imaging of space micro-motion target;The ISAR image is carried out morphological processing open operation, using Canny operator carries out edge detection processing, carries out Hough transformation and carries out straight line detection, obtains the size and boundary of target;The echo model of compound micro-motion target of combined micro-motion is established, and the micro-motion period is estimated using the method based on the phase difference signal aggregation degree, realizes the de-micro-motion imaging of space target and micro-motion target feature extraction and solves the problem of target threat assessment difficulty.

Description

Technical Field

[0001] This invention relates to the field of ISAR signal processing, and more specifically to a method for evaluating target ISAR images based on signal processing and image changes. Background Technology

[0002] ISAR can perform high-resolution imaging of non-cooperative targets at long distances, under all-weather and all-time conditions, thus it has broad application prospects in fields such as space target detection and space situational awareness. However, further research is needed in the areas of motion-free imaging and feature extraction for slightly moving targets. When the same space target is stationary, its ISAR imaging results are relatively clear; however, when the space target is slightly moving, its ISAR imaging exhibits blurring and defocusing. Blurring and defocusing greatly complicate image feature extraction, and slightly moving space targets may pose a significant threat to space security. Therefore, existing technologies suffer from problems such as blurred ISAR images and difficulties in feature extraction when the target's attitude changes, leading to difficulties in target threat assessment. Summary of the Invention

[0003] The purpose of this invention is to address the problem of blurring and defocusing during ISAR imaging in existing technologies, which greatly hinders image feature extraction. This invention proposes a target ISAR image assessment method based on signal processing and image changes. This method utilizes micro-motion rearrangement and culling methods for micro-motion removal imaging to obtain a clearer target image. Morphological processing, including opening operations, is then applied to smooth the image. The size and boundary of the target are obtained through the Canny operator and Hough transform. Finally, the micro-motion period is estimated based on the phase difference signal aggregation method. The obtained target size, boundary, and micro-motion period information serve as key information for threat assessment, further ensuring space security.

[0004] The technical solution of this application is:

[0005] 1. Target ISAR image evaluation methods based on signal processing and image changes, including:

[0006] S1: The radar echo signal is processed using the RD algorithm to obtain the range cell-azimuth cell image. The obtained range cell-azimuth cell image is then subjected to two rounds of micro-motion rearrangement and removal using micro-Doppler echo separation technology to obtain the space micro-motion target ISAR image.

[0007] The range-azimuth image consists of Y range units and X azimuth units;

[0008] S2: Extract space target features from the ISAR image of the space micro-movement target obtained in S1;

[0009] S3: Establish a composite micro-motion target echo model combining translational micro-motion and use a method based on phase difference signal concentration to estimate the micro-motion period of the target;

[0010] S4: The threat level of the target object is assessed based on the spatial target characteristics obtained in S2 and the micro-motion cycle of the target object obtained in S3.

[0011] 2. An electronic device, characterized in that it comprises:

[0012] One or more processors;

[0013] Storage device for storing one or more programs.

[0014] When the one or more programs are executed by the one or more processors, the one or more processors execute the above-described target ISAR image evaluation method based on signal processing and image changes.

[0015] Compared with the prior art, this application has the following advantages:

[0016] The present invention provides a target ISAR image assessment method based on signal processing and image changes. It utilizes a micro-motion rearrangement and elimination method to achieve de-motion imaging of spatial micro-motion targets; performs morphological processing on the ISAR image, including opening operations, edge detection using the Canny operator, and line detection using Hough transform to obtain the target size and boundary; establishes a composite micro-motion target echo model combining translational and micro-motion, and estimates the micro-motion period using a method based on phase difference signal aggregation. This achieves de-motion imaging of spatial targets and extraction of micro-motion target features, and solves the problem of difficulty in assessing the threat level of targets. Attached Figure Description

[0017] Figure 1 This is a flowchart of the target ISAR image evaluation method based on signal processing and image changes according to the present invention;

[0018] Figure 2 This is a schematic diagram of the micro-Doppler echo signal separation and micro-motion removal imaging process of the present invention;

[0019] Figure 3 This is a schematic diagram of the spatial target feature extraction process of the present invention;

[0020] Figure 4 This is a schematic diagram of the micro-motion period estimation process based on the phase difference signal aggregation degree of the present invention;

[0021] Figure 5 .a is a three-dimensional scatter diagram of the micro-motion removal imaging model of the present invention;

[0022] Figure 5.b is a schematic diagram of the RD image before micro-motion removal imaging according to the present invention;

[0023] Figure 5 .c is the result image after the first micro-motion removal imaging of the present invention;

[0024] Figure 5 .d is the result of the first removal of micro-motion signals in the micro-motion removal imaging of the present invention;

[0025] Figure 5 .e is the result image after the second micro-motion removal imaging of the present invention;

[0026] Figure 5 .f is the result of the second removal of micro-motion signals in the micro-motion removal imaging of the present invention;

[0027] Figure 6 .a is a schematic diagram of the simulated target ISAR image for spatial target feature extraction according to the present invention;

[0028] Figure 6 .b is the result of the dilation processing of the simulated image for spatial target feature extraction according to the present invention;

[0029] Figure 6 .c is the result of erosion processing of the simulated image for spatial target feature extraction according to the present invention;

[0030] Figure 6 .d is the edge detection result of the simulation image for spatial target feature extraction in this invention;

[0031] Figure 6 .e is the simulation result of Hough domain marker points for spatial target feature extraction according to the present invention;

[0032] Figure 6 .f is the result of extracting a straight line from the simulated image domain of spatial target feature extraction according to the present invention;

[0033] Figure 7 The figure shows the simulation results of the phase differential signal concentration of the present invention. Detailed Implementation

[0034] Specific implementation method one: Combining Figure 1 This implementation method is described as follows: Figure 1 As shown, the target ISAR image evaluation method based on signal processing and image changes in this embodiment includes:

[0035] S1: The radar echo signal is processed using the RD algorithm to obtain the range cell-azimuth cell image, thereby obtaining the target's range and velocity information. The obtained range cell-azimuth cell image is then subjected to two rounds of micro-motion rearrangement and removal using micro-Doppler echo separation technology to obtain a clear ISAR image of the spatially micro-motion target.

[0036] The full name of the RD algorithm is the Range Doppler algorithm.

[0037] The range-azimuth image consists of Y range units and X azimuth units;

[0038] The radar echo signal refers to the radar echo received after the radar transmitted signal is scattered and reflected by the target object.

[0039] S2: Extract spatial target features from the ISAR image of the spatially moving target obtained in S1;

[0040] S3: Establish a composite micro-motion target echo model combining translational micro-motion and use a method based on phase difference signal concentration to estimate the micro-motion period of the target;

[0041] S4: Based on the space target characteristics obtained in S2 and the micro-motion period of the target object obtained in S3, assess the threat level of the target object. Space target characteristics generally include the target's size and boundary information; larger objects may have a higher threat level. For example, in the aerospace field, large space debris may pose a greater threat to satellites or other spacecraft because they can cause more severe damage. In military applications, target size can be correlated with potential firepower or the number of weapons carried. Boundaries refer to the shape or appearance of the target. Irregular or complex boundaries may indicate that the target has special functions or characteristics. Certain boundary shapes may indicate that the target has stealth or counter-reconnaissance capabilities, increasing its threat level.

[0042] Micro-periods can be understood as a target's behavioral pattern or motion characteristics. For example, a target that frequently changes its motion state may be more difficult to predict and intercept, thus posing a higher threat. In the aerospace field, micro-periods may be related to a target's orbital stability. An unstable orbit may imply a higher risk of collision.

[0043] Combining these two key pieces of information, the assessment process involves first assigning weights α1 and α2 to each item, and then assigning a threat score χ1 and χ2 to the target based on the specific information of each item. Finally, these scores are summed to obtain the target's total threat assessment score ψ = α1χ1 + α2χ1.

[0044] Specific Implementation Method Two: Combining Figure 2 This embodiment differs from specific embodiment one in that...

[0045] In step S1, the range cell-azimuth cell image is subjected to two rounds of micro-motion rearrangement and removal using micro-Doppler echo separation technology to obtain a clear ISAR image of a spatially moving target. The specific process is as follows:

[0046] S11: The range cell-azimuth cell image obtained using the RD algorithm is sorted in descending order according to the magnitude of the azimuth dimension signal in each range cell. The azimuth dimension signal of the top 10% of the magnitude is retained. Then, the retained azimuth dimension signal is restored to the original azimuth dimension arrangement order of the range cell-azimuth cell image. The range cell-azimuth cell image after initial micro-motion rearrangement and elimination is obtained.

[0047] S12: The range cell-azimuth cell images that have been initially rearranged and removed due to micro-motion are sorted in descending order according to the signal magnitude of the range dimension in each azimuth cell, and the signals of the range dimension with the first 30% of the magnitude are retained; then the signals of the retained range dimension are restored to the original range cell-azimuth cell image in the order of the range dimension, and a clear ISAR image of the spatially micro-motion target is obtained.

[0048] The distance unit represents the unit for dividing and measuring the distance between the target and the radar; different distance units contain combined radar echo signals at different distance positions;

[0049] The azimuth unit represents the unit that divides the measurement target and the radar in azimuth (or angle). Different azimuth units contain combined radar echo signals from different azimuths (or angles). Other steps and parameters are the same as in Specific Implementation Method 1.

[0050] Specific implementation method three: Combining Figure 2 and Figure 5 This embodiment differs from specific embodiment one or two in that:

[0051] In step S11, the combined radar echo signal contained in each range cell of the range cell-azimuth cell image is separated, and the separated signals are sorted in descending order according to the magnitude of the azimuth dimension, retaining the top M signals according to the magnitude of the azimuth dimension. Q1 The signal is then processed to restore the original azimuth dimension arrangement of the range cell-azimuth cell image; the initial rearrangement and removal of range cells-azimuth cell images is obtained; the specific process is as follows:

[0052] S111: Combine the azimuth dimension data of each distance cell in the distance cell-azimuth cell image into an azimuth dimension data set;

[0053] S112: Take the modulus of each azimuth dimension data in the azimuth dimension dataset, and then rearrange all azimuth dimension data according to the magnitude of the azimuth dimension modulus. The rearranged result is:

[0054] |Ψ k (0)|≥|Ψ k (1)|≥|Ψ k (2)|...≥|Ψ k (M-1)| (1)

[0055] Among them Ψ k ∈S k (m), |Ψ k (0)| represents S k The maximum modulus of (m), S k (m) represents the azimuth dimension data set;

[0056] S113: Select the group with the largest modulus. Each element is retained, among which Q1 represents the removal ratio, with a value of 90. `int` indicates rounding up. The retained azimuth dimension data is restored according to the original arrangement order of the range cell-azimuth cell image, resulting in the initial rearranged and removed range cell-azimuth cell image.

[0057] The original arrangement order of the range cell-azimuth cell images is: the arrangement order of the range cell-azimuth cell images obtained using the RD algorithm.

[0058] son Figure 5 a) is a three-dimensional scatter plot of the spatial target. Figure 5 b) shows the imaging results of a space target exhibiting slight movement after applying the RD algorithm. Figure 5 c) represents the first fine-tuning result obtained after processing with the distance-dimensional fine-tuning algorithm. Figure 5 .d) It was observed that the initial micro-motion removal process eliminated most of the micro-motion signals, i.e., the blur components in the imaging results. Other steps and parameters are the same as in specific implementation method one or two.

[0059] Specific implementation method four: Combination Figure 2 and Figure 5 This embodiment differs from specific embodiments one through three in that...

[0060] In step S12, the distance-azimuth unit images that were initially rearranged and eliminated after the micro-motion removal are sorted in descending order according to the magnitude of the distance dimension in each azimuth unit, and the images with the highest magnitudes according to the distance dimension are retained. The signal is then analyzed; the range dimension arrangement of the original range cell-azimuth cell image is restored to obtain a clear ISAR image of a spatially moving target; the specific process is as follows:

[0061] S121: Combine the distance dimension data of each azimuth cell in the distance cell-azimuth cell image into a distance dimension data set;

[0062] S122: Take the modulus of each distance dimension data in the distance dimension dataset, and then rearrange all distance dimension data according to the magnitude of the modulus value. The rearranged result is:

[0063] |ΨR (0)|≥|Ψ R (1)|≥|Ψ R (2)|...≥|Ψ R (M-1)| (2)

[0064] Among them Ψ R ∈S R (m), |Ψ R (0)| represents S R The maximum modulus of (m), S R (m) represents the distance dimension data set;

[0065] S123: Select the group with the largest modulus. Each element is retained, among which Q2 represents the removal ratio, with a value of 70. Int indicates rounding up. The retained elements are restored according to the original arrangement order of the range cell-azimuth cell image to obtain the ISAR image after motion removal.

[0066] The original arrangement order of the range cell-azimuth cell images is: the arrangement order of the range cell-azimuth cell images obtained using the RD algorithm.

[0067] son Figure 5 e) is the second motion-free imaging result obtained after processing with the azimuth-dimensional motion-free algorithm, which can be subdivided. Figure 5 f) It was observed that the two de-motion processing steps eliminated most of the motion signals, which is the blur component in the imaging results.

[0068] The other steps and parameters are the same as those in one of the specific implementation methods one to three.

[0069] Specific Implementation Method Five: Combining Figure 3-6 This embodiment differs from specific embodiments one through four in that, in step S2, spatial target feature extraction is performed on the ISAR image of the spatially moving target obtained in S1. The specific process is as follows:

[0070] S21: Perform morphological processing and opening operations on the ISAR image of the space micro-moving target obtained in S1;

[0071] S22: The ISAR image after morphological processing and opening operation of S21 is processed by the Canny operator, and edge detection processing is performed to obtain the edge information of the target object.

[0072] S23: Using the Hough transform algorithm, the ISAR image after edge detection processing in S22 is processed to perform line detection and Hough domain feature point labeling to obtain the size information of the target object.

[0073] The feature points are obtained by calculating the extreme points in the Hough domain; other steps and parameters are the same as in one of the specific implementation methods one to four.

[0074] Specific Implementation Method Six: Combination Figure 6 This embodiment differs from specific embodiments one through five in that...

[0075] The specific process of the opening operation in the S21 morphological processing is as follows: (e.g., sub-) Figure 6 a) shows an S1-based radar ISAR image after fine-tuning, where only the approximate location of the target is visible; the boundaries and size are unclear.

[0076] S211: Perform dilation operation on the input ISAR image.

[0077] The dilation operation makes an object grow by comparing the point we are interested in with the points in the cross-shaped configuration around it one by one. If the conditions of the operation are met, then the area around the point in the original image is filled with color.

[0078] Figure 6 .b is the result of the dilation processing of the simulated image for spatial target feature extraction according to the present invention;

[0079] S212: Perform corrosion operation on the corrosion result to obtain the outer contour of the target object.

[0080] The erosion operation reduces the size of an object along its edges, removing small details and noise. The principle of erosion is to slide a structuring element across the image and check the overlap between the structuring element and the image. If the structuring element is completely contained within its corresponding position in the image, the pixel at that position is preserved; otherwise, the pixel at that position is set to 0 (black). Image erosion works similarly to image dilation, and after processing, the overall outline of the target is visible.

[0081] Figure 6 .c is the result of the erosion processing of the simulation image for spatial target feature extraction according to the present invention; other steps and parameters are the same as those in one of the specific embodiments one to five.

[0082] Specific implementation method seven: Combining Figure 6 This embodiment differs from specific embodiments one through six in that it includes:

[0083] The specific process of edge detection processing of the ISAR image after morphological opening operation of S21 using the Canny operator in S22 is as follows:

[0084] S221: Comparative morphological processing of the gray values ​​between adjacent pixels in the ISAR image after opening operation, and calculation of the image difference using the gray values.

[0085] S222: Calculate the gradient value and gradient direction of each pixel in the ISAR image after morphological processing and opening operation using image difference;

[0086] S223: Using the gradient value and gradient direction of each pixel in the morphologically processed opening ISAR image, the direction perpendicular to the gradient direction of the partial derivative matrix in the x and y directions of the entire morphologically processed opening ISAR image is obtained. The direction perpendicular to the gradient direction is the edge direction.

[0087] S224: The edge direction is tracked using the Canny operator to obtain the edge-detected ISAR image. The edge detection is then used to obtain the image as shown in the image. Figure 6 The edge detection image shown in .d) is processed so that the size of the target object is roughly clear. At this point, the sizes of the satellite and solar panels are roughly clear. Other steps and parameters are the same as in any of the specific implementation methods one through six.

[0088] Specific implementation method eight: Combination Figure 6 This embodiment differs from specific embodiments one through seven in that...

[0089] The specific process of S23 using the Hough transform algorithm to mark Hough domain feature points in the ISAR image after edge detection processing in S22, and then marking feature lines in the image domain, is as follows:

[0090] S231: First, the Hough transform is used to transform the edge-detected ISAR image from the Cartesian coordinate system (xy) to the polar coordinate system (ρ-θ) in the Hough domain. Each point in the Hough domain represents a straight line in the image domain.

[0091] S232: Then, feature points in the ISAR image after edge detection processing are marked in the Hough domain. Finally, based on the feature points in the Hough domain, straight lines in the ISAR image after edge detection processing are marked in the image domain. Ultimately, spatial target features are extracted, including the size information and boundary information of the target object.

[0092] The feature points are obtained by calculating the extreme points in the Hough domain.

[0093] By selecting intersection points in the Hough domain using the Hough transform, lines can be marked in the image domain, ultimately yielding the sub-line. Figure 6 The size and boundaries of the target are clearly defined. Other steps and parameters are the same as in any of the specific implementation methods one through seven.

[0094] Specific Implementation Method Nine: Combining Figure 7This embodiment differs from specific embodiments one through eight in that, in step S3, a composite micro-motion target echo model combining translational and micro-motion is established, and the micro-motion period of the target is estimated using a method based on phase difference signal concentration, including:

[0095] S31: Establish the composite micro-motion echo signal model. The specific process is as follows:

[0096] First, the radar echo signal is modeled using the target equivalent scattering center substitution method. The radar echo signal includes the radar echo signal of the translational motion of the target object and the radar echo signal of the micro-motion of the target object.

[0097] If the target object consists of L scattering centers, then the distance from the l-th scattering center to the radar is:

[0098] r l (t)=r M,l (t)+r T (t) (3)

[0099] Where, distance r T (t) corresponds to the translational motion of the target object, with a distance r. M,l (t) corresponds to the micro-motion of the l-th scattering center. The micro-motion includes periodic and aperiodic micro-motions; the periodic micro-motions have a period, specifically expressed as:

[0100] r M,l (t+T M ) = r M,l (t) (4)

[0101] Among them, T M This refers to the periodicity of micro-motions. Periodic micro-motions include movements such as spin.

[0102] Within the radar echo coherent processing time, the target translational motion can be modeled using a polynomial as follows:

[0103]

[0104] Where Q takes the value 3, indicating that a third-order polynomial model is used, R0 represents the initial position, and a i The parameter represents the motion state in most cases.

[0105] The echo coherence processing time of the radar refers to the time during which the radar continuously sends and receives the returned echo signals, during which the radar can maintain coherence (i.e., maintain phase continuity).

[0106] radar echo signal Specifically, it is expressed as follows:

[0107]

[0108] Where p is the radar's transmitted signal, f c Let t be the carrier frequency and t be the slow time. For fast time, L is the number of scattering centers of the target, σ l Let τ be the RCS of the l-th scattering center, where l is the scattering center, and τ is the RCS of the l-th scattering center. l (t) represents the time delay of the l-th scattering center, exp is the exponential function, RCS is the radar cross section, j is the imaginary unit, and π is the mathematical constant pi.

[0109] The radar cross section (RCS) is obtained from the RCS calculation formula; the distance r from the l-th scattering center to the radar is... l (t) represents the variable in the formula for calculating the RCS of the l-th scattering center;

[0110] radar echo signal Distance compression is represented as:

[0111]

[0112] in, Let be the point spread function, in A peak is formed at this point. When the radar transmitted signal is a linear frequency modulated signal, the point spread function exhibits a sinc function form, p. r (0) = 1;

[0113] The location of the l-th scattering center of the target object is... Substituting into formula (7), we obtain the baseband signal s. r (t);

[0114] The baseband signal s r (t) is the radar echo signal of the translational motion of the target object plus the radar echo signal of the micro-motion of the target object, which is the composite micro-motion echo signal model.

[0115] S32: Calculate the phase difference signal based on the composite micro-motion echo signal model. The specific process is as follows:

[0116] Let the sampling interval of the radar echo signal be ΔT, and calculate the radar echo signal as a discrete phase difference signal s. PD (n,m):

[0117]

[0118] In the formula, n, m, and N are time variables, 0 ≤ m < N, 0 ≤ n < Nm. For baseband signal s r The conjugate signal, sr (n+m) represents the baseband signal with time t = n+m;

[0119] S33: Estimate the modulation frequency of the phase differential signal described in S32. As shown in the following formula:

[0120]

[0121] Among them, S β (k,m) represents the Fourier transform of the phase difference signal with time delay m after de-frequency modulation, N f The number of points in the Fourier transform. The expression represents the value of β when it reaches its maximum value, where β is the frequency modulation frequency, k and f are Fourier transform parameters, and the Fourier transform process of the phase difference signal after frequency modulation with a time delay of m is as follows:

[0122]

[0123] Where ΔT is the sampling interval of the radar echo signal.

[0124] S34: After demodulation, the phase differential signal degenerates into a linear frequency modulated signal, based on the modulation frequency of the phase differential signal estimated in S33. The result yields the concentration ε(m) of the linear frequency modulated signal:

[0125]

[0126]

[0127] S35: Estimate the micro-motion period of the target object based on the concentration ε(m) of the linear frequency modulated signal obtained in S34:

[0128] Peak detection is performed on the concentration degree ε(m) of the linear frequency modulated signal obtained by S34. The time corresponding to the position where the peak appears is the estimated micro-motion period of the target object.

[0129] Figure 7 This is a simulation result diagram of phase differential micro-motion period estimation. The micro-motion period set in the simulation is 1s. Based on the phase differential signal concentration results, the micro-motion period is determined to be 1s, which is consistent with the parameters set in the simulation, thus realizing the micro-motion period estimation. Other steps and parameters are the same as in specific implementation methods one to eight.

[0130] Specific Implementation Method Ten: An electronic device, characterized in that it includes:

[0131] One or more processors;

[0132] Storage device for storing one or more programs.

[0133] When the one or more programs are executed by the one or more processors, the one or more processors perform the target ISAR image evaluation method based on signal processing and image changes as described in any one of embodiments 1 to 9.

[0134] The above description is merely of preferred embodiments of the present invention. It should be understood that the present invention is not limited to the specific embodiments described above. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent substitutions, and improvements made to the above embodiments without departing from the scope of the present invention, based on the technical essence of the present invention, and within the spirit and principles of the present invention, shall still fall within the protection scope of the present invention.

Claims

1. A target ISAR image evaluation method based on signal processing and image changes, characterized in that, S1: The radar echo signal is processed using the RD algorithm to obtain the range cell-azimuth cell image. The obtained range cell-azimuth cell image is then subjected to two rounds of micro-motion rearrangement and removal using micro-Doppler echo separation technology to obtain the space micro-motion target ISAR image. S2: Extract space target features from the ISAR image of the space micro-movement target obtained in S1; S3: Establish a composite micro-motion target echo model combining translational micro-motion and use a method based on phase difference signal concentration to estimate the micro-motion period of the target; S4: Assess the threat level of the target object based on the spatial target characteristics obtained in S2 and the micro-motion cycle of the target object obtained in S3; In step S1, the range cell-azimuth cell image is subjected to two rounds of micro-motion rearrangement and removal using micro-Doppler echo separation technology to obtain a spatially moving target ISAR image. The specific process is as follows: S11: Sort the obtained range cell-azimuth cell image in descending order according to the magnitude of the azimuth dimension signal in each range cell, retain the azimuth dimension signal of the top 10% of the magnitude, and then restore the retained azimuth dimension signal to the original arrangement order of the azimuth dimension of the range cell-azimuth cell image; thus obtaining the range cell-azimuth cell image after initial micro-motion rearrangement and elimination. S12: The range cell-azimuth cell images that have been initially rearranged and removed due to micro-motion are sorted in descending order according to the signal magnitude of the range dimension in each azimuth cell, and the signals of the range dimension with the first 30% of the magnitude are retained; then the retained range dimension signals are restored to the original range cell-azimuth cell image in the order of the range dimension, and the space micro-motion target ISAR image is obtained.

2. The target ISAR image evaluation method based on signal processing and image changes according to claim 1, characterized in that: In step S11, the range cell-azimuth cell images obtained using the RD algorithm are sorted in descending order for each range cell according to the magnitude of the azimuth dimension, retaining the images with the highest magnitudes according to the azimuth dimension. The signal is then restored to the original azimuth dimension arrangement order of the range cell-azimuth cell image; thus, the range cell-azimuth cell image after initial micro-motion rearrangement and removal is obtained. The specific process is as follows: S111: Combine the azimuth dimension data of each distance cell in the distance cell-azimuth cell image into an azimuth dimension data set; S112: Take the modulus of each azimuth dimension data in the azimuth dimension dataset, and then rearrange all azimuth dimension data according to the magnitude of the azimuth dimension modulus. The rearranged result is: （1） in , express The maximum modulus, For directional dimension data set; S113: Select the group with the largest modulus. Each element is retained, among which , The value represents the removal ratio, which is 90. The integer part indicates rounding up. The retained azimuth dimension data is restored according to the original arrangement order of the distance cell-azimuth cell image to obtain the initial rearrangement and removal of the distance cell-azimuth cell image.

3. The target ISAR image evaluation method based on signal processing and image changes according to claim 1, characterized in that: In step S12, the distance-azimuth unit images that were initially rearranged and eliminated after the micro-motion removal are sorted in descending order according to the magnitude of the distance dimension in each azimuth unit, and the images with the highest magnitudes according to the distance dimension are retained. One signal; then restore the range dimension arrangement order of the original range cell-azimuth cell image to obtain the ISAR image of the spatially moving target; the specific process is as follows: S121: Combine the distance dimension data of each azimuth cell in the distance cell-azimuth cell image into a distance dimension data set; S122: Take the modulus of each distance dimension data in the distance dimension dataset, and then rearrange all distance dimension data according to the magnitude of the modulus value. The rearranged result is: （2） in , express The maximum modulus, For distance dimension data set; S123: Select the group with the largest modulus. Each element is retained, among which , This indicates the removal ratio, with a value of 70. Int means rounding up. The retained elements are restored according to the original arrangement order of the range cell-azimuth cell image to obtain the ISAR image after motion removal.

4. The target ISAR image evaluation method based on signal processing and image changes according to claim 1, characterized in that: The specific process for extracting space target features from the ISAR image of the space micro-motion target obtained in S1 in S2 is as follows: S21: Perform morphological processing and opening operations on the ISAR image of the space micro-moving target obtained in S1 to obtain the processed ISAR image; S22: Use the Canny operator to perform edge detection processing on the ISAR image after S21 to obtain the edge-detected ISAR image; S23: The Hough transform algorithm is used to mark the Hough domain feature points of the ISAR image obtained after edge detection processing in S22, and feature lines are marked in the image domain based on the feature points, and finally the spatial target features are extracted.

5. The target ISAR image evaluation method based on signal processing and image changes according to claim 4, characterized in that: The specific process of the morphological opening operation in S21 is as follows: S211: Perform dilation operation on the input ISAR image. S212: Perform an erosion operation on the dilation result to obtain an ISAR image after morphological processing and opening operation.

6. The target ISAR image evaluation method based on signal processing and image changes according to claim 4, characterized in that: The specific process of edge detection processing of the ISAR image after morphological opening operation of S21 using the Canny operator in S22 is as follows: S221: Compare the gray values ​​between adjacent pixels in the ISAR image after opening operation using morphological processing, and calculate the difference between the images using the gray values; S222: Calculate the gradient value and gradient direction of each pixel in the ISAR image after morphological processing and opening operation using image difference; S223: Obtain the entire morphologically processed ISAR image by using the gradient value and gradient direction of each pixel in the morphologically processed opening operation. and The partial derivative matrix of the direction is perpendicular to the gradient direction, and the direction perpendicular to the gradient direction is the edge direction. S224: The edge direction is tracked by the Canny operator to obtain the ISAR image after edge detection processing.

7. The target ISAR image evaluation method based on signal processing and image changes according to claim 4, characterized in that: The specific process of using the Hough transform algorithm to mark Hough domain feature points in the ISAR image after edge detection processing in S22 in S23, and then marking feature lines in the image domain, is as follows: S231: First, use Hough transform to convert the edge-detected ISAR image from Cartesian coordinates. Transforming the image domain into polar coordinates The Hough region, where each point in the Hough region represents a straight line in the image domain. S232: Then, feature points in the ISAR image after edge detection processing are marked in the Hough domain. Finally, based on the feature points in the Hough domain, straight lines in the ISAR image after edge detection processing are marked in the image domain. Ultimately, spatial target features are extracted, including the size information and boundary information of the target object. The feature points are obtained by calculating the extreme points in the Hough domain.

8. The target ISAR image evaluation method based on signal processing and image changes according to claim 1, characterized in that: In step S3, a composite micro-motion target echo model combining translational and micro-motion is established, and the micro-motion period of the target is estimated using a method based on phase difference signal concentration, including: S31: Establish a composite micro-motion echo signal model combining translational and micro-motion. The specific process is as follows: First, the radar echo signal is modeled using the target equivalent scattering center substitution method. The radar echo signal includes the radar echo signal of the translational motion of the target object and the radar echo signal of the micro-motion of the target object. Assuming the target object consists of L scattering centers, then the th scattering center... The distance from each scattering center to the radar is: （3） Among them, distance This corresponds to the translational motion of the target object, and the distance. Corresponding to the number The micro-motions of the scattering centers include periodic and aperiodic micro-motions. For periodic micro-motions: （4） in, This is the micro-motion cycle; Within the radar echo coherent processing time, the target translational motion can be modeled using a polynomial as follows: （5） Here, Q takes the value 3, indicating that a third-order polynomial model is used. Indicated as the initial position, Represented as parameters radar echo signal Specifically, it is expressed as follows: （6） in, For radar transmission signals, For carrier frequency, For slow time, For fast time, L is the number of scattering centers of the target. For the first Each scattering center , As the scattering center, For the first The time delay of each scattering center It is an exponential function, where RCS is the radar cross section. The imaginary unit, Pi; radar cross section According to the RCS calculation formula, the first... Distance from each scattering center to the radar To calculate the first Each scattering center Variables in the calculation formula; target radar echo signal Distance compression is represented as: （7） in, Let be the point spread function, in A peak is formed at this point. When the radar transmitted signal is a linear frequency modulated signal, the point spread function exhibits the following behavior: Functional form, ; The first of the target objects The locations where the scattering centers appear are: Substituting into formula (7), the baseband signal is obtained. ; The baseband signal The radar echo signal of the translational motion of the target object plus the radar echo signal of the micro-motion of the target object is the composite micro-motion echo signal model of translational combined with micro-motion. S32: Calculate the phase difference signal based on the composite micro-motion echo signal model of translational combined micro-motion. The specific process is as follows: Let the sampling interval of the radar echo signal be . The radar echo signal is calculated as a discrete phase differential signal. : （8） In the formula, n, m, and N are time variables. , , Baseband signal The conjugate signal, The baseband signal has a time t of n+m; S33: Estimate the modulation frequency of the phase differential signal described in S32. As shown in the following formula: （9） in, For delay The Fourier transform of the phase difference signal after frequency modulation The number of points in the Fourier transform. This indicates when the expression reaches its maximum value. value, For frequency modulation, k and f are Fourier transform parameters. The delay is Fourier transform of phase difference signal after frequency modulation The expression is: （10） in, The sampling interval for the radar echo signal. S34: After demodulation, the phase differential signal degenerates into a linear frequency modulated signal, based on the modulation frequency of the phase differential signal estimated in S33. The result at that time yields the concentration of the linear frequency modulated signal. : （11） in, （12） S35: The concentration of the linear frequency modulated signal obtained from S34 Estimate the micro-motion period of the target object: The concentration of the linear frequency modulated signal obtained from S34 Peak detection is performed, and the time corresponding to the location where the peak occurs is the estimated micro-motion period of the target object.

9. An electronic device, characterized in that, include: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors perform the target ISAR image evaluation method based on signal processing and image changes according to any one of claims 1 to 8.

Images