SAR image domain false target generation method based on time-frequency rotation
By embedding a false target template in the SAR image domain and using fractional Fourier transform and image background spectrum fusion, a false target that approximates the real target is generated. This solves the problems of large computational load and large difference between false targets in the existing technology, and improves the effectiveness and realism of deception and interference.
Patent Information
- Application Number
- CN202511889634.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-15
- Publication Date
- 2026-02-13
AI Technical Summary
Existing SAR deception jamming methods involve large computational loads and complex algorithms when generating false targets, making it difficult to efficiently achieve deception jamming of spaceborne SAR. Furthermore, the false targets generated by traditional methods differ significantly from real targets in terms of motion state and statistical characteristics, making them difficult to identify through deception jamming.
A false target generation method based on time-frequency rotation in the SAR image domain is adopted. By embedding false target templates into SAR Level 1A SLC image data and using fractional Fourier transform to control the degree of defocus, combined with image background spectrum fusion and amplitude histogram matching, false targets that approximate real targets are generated.
This method enables the generation of false targets that closely approximate real targets in SAR images, improving image detail and realism. It also solves the problem of large differences between false targets and real targets in traditional methods, thus enhancing the effectiveness of deception and interference.
Smart Images
Figure CN121522583A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of radar technology, specifically relating to a method for generating false targets in the SAR image domain based on time-frequency rotation. Background Technology
[0002] Synthetic Aperture Radar (SAR) possesses all-weather, high-resolution imaging capabilities and is widely used for target reconnaissance and identification. After receiving SAR signals, ground-based jammers modulate and generate false echoes with characteristics of real targets, then relay these echoes back to the SAR system, creating SAR images containing false targets and effectively jamming SAR reconnaissance. Existing SAR deception jamming methods generate false echoes through signal frequency storage and forwarding, as well as template convolution, but these methods require significant computation and involve complex algorithms. They can only effectively deceive and jam airborne SAR systems; given the massive amounts of data generated by spaceborne SAR systems, they are insufficient for efficient deception jamming.
[0003] Traditional deception jamming is implemented in the echo domain, requiring high precision in the motion parameters of the SAR platform and involving complex algorithm derivations. Furthermore, existing false target generation methods often only process amplitude and phase, failing to consider the target's motion state and statistical characteristics. When a target moves, its scattering characteristics change with velocity, causing defocusing in the SAR image. Therefore, false targets generated by traditional methods differ significantly from real targets in geometry, defocusing degree, or statistical distribution, making them difficult to detect through deception jamming identification algorithms or manual interpretation, thus failing to achieve effective jamming. Deception jamming also lacks corresponding image domain simulation technology. This technology has low-cost advantages in areas such as echo generation algorithm inversion and providing false target templates, and also helps establish an inverse mapping model from the image domain to the signal domain. Since there are currently no simulation methods or patents for moving target deception jamming in the image domain, researching this technology will help provide a large number of realistic false target templates and promote the echo domain algorithm inversion for moving target deception jamming. Summary of the Invention
[0004] To overcome the shortcomings of existing technologies, this invention provides a SAR image domain false target generation method based on time-frequency rotation. This method involves locally processing SAR Level 1A SLC image data and embedding false target templates. Simultaneously, fractional Fourier transform is used to control the defocusing degree of the false target in the azimuth direction. Refined generation of deception interference is achieved through the fusion of the false target with the image background. Amplitude histogram matching ensures that the amplitude distribution of the false target is consistent with the real image, while preserving the target's phase characteristics. This generates false targets in the SAR image that closely approximate the real target in terms of geometric shape, defocusing degree, and statistical characteristics. By fusing the image background spectrum and using histogram matching based on the statistical characteristics of the real target's amplitude, this invention effectively avoids the problems of large differences in amplitude characteristics and obvious background fragmentation between the defocused false target image and the real target, thus improving the refinement and realism of the false targets in the image.
[0005] The technical solution adopted by this invention to solve its technical problem is as follows: Step 1: Receive synthetic aperture radar single-view complex SLC image data and represent the image region to be processed in the SLC image data as a complex matrix. Simultaneously input the SLC image template of the dummy target. , use fake target image template After converting to a complex matrix, it is clipped to a size of Matrix; Step 2: In The required Region of Interest (ROI) is determined for pseudo-target fusion; For the false target image template obtained in step 1 Calculate its amplitude statistical characteristics, and calculate the detection threshold based on the mean and standard deviation. For false target image templates Filtering is performed to remove background noise components from the target spectrum; detection threshold. Represented as: (1) (2) (3) in, It's a template. The numerical value of the element in the middle. It is the mean of all elements. Standard deviation of all elements The total number of elements; Using detection thresholds Determine if a given element is background noise, and set the background noise to zero to obtain the template. : (4) That is, when the template The value in Greater than the detection threshold It is then determined to be the target element; Step 3: The size obtained in Step 2 is False target image template Perform time-frequency rotation; perform fractional Fourier transform along the azimuth direction; Step 4: Perform a range-to-Fourier transform on the fake target template after time-frequency rotation in Step 3 to obtain the two-dimensional range spectrum in the range frequency domain and azimuth time domain. Simultaneously, for the ROI selected in step 2, the size is... Perform a range-to-Fourier transform on the image background region to obtain the reference background range spectrum. The two-dimensional range spectrum of the false target and the range spectrum of the reference background are fused according to the energy distribution ratio. This allows for the adjustment of the spectral characteristics of the false target; the formula for generating the spectral fusion is as follows: (7) in, This is the image distance spectral energy ratio coefficient. The reference background distance spectral energy ratio coefficient; Step 5: Analyze the two-dimensional range spectrum of the false target obtained in Step 4. The false target template Z, after fusing background features, is obtained through inverse Fourier transform; a reference region R is selected, amplitude normalization is performed, and the amplitude histogram is calculated. Histogram sequence interval index The calculation method is as follows: (8) (9) in, This represents the floor function. This represents the normalized amplitude over the reference interval. Indicates the length of the histogram interval. This represents the number of intervals in the histogram. When When the value is 1, it is classified into the last interval; Step 6: Use the histogram obtained in Step 5 Histogram matching is performed on the false targets to correct the amplitude statistical characteristics of the defocused false target template. Simultaneously, the phase component φ of the pre-matched false target template is added azimuthally to ensure that the imaging coherence remains consistent with the target morphology. Finally, the processed false target template is fused into the real image to obtain a SAR image containing the defocused false targets. The mapping function for histogram matching is expressed as: (10) in, Let be the cumulative distribution function of the image to be matched. For the first The number of pixels in each interval For reference histogram The cumulative distribution function, It is the histogram index of the image to be matched. It is the histogram index of the matched image.
[0006] Preferably, the formula for calculating the fractional Fourier transform is expressed as follows: (5) (6) in, It is the kernel function of FrFT. Represents a one-dimensional signal In the FrFT domain sampling points, It is a natural constant. The imaginary unit; Represents a time variable. Indicates changing the angle. This indicates the order of the fractional Fourier transform.
[0007] An electronic device includes: a processor and a memory; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to cause the electronic device to perform the above-described spoof target generation method.
[0008] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned method for generating spoofed targets.
[0009] A chip includes a processor for retrieving and running a computer program from a memory, causing a device on which the chip is installed to perform the aforementioned spoof target generation method.
[0010] A computer program product includes a computer storage medium storing a computer program, the computer program including instructions executable by at least one processor, which, when executed by the at least one processor, implement the aforementioned spoofing target generation method.
[0011] The beneficial effects of this invention are as follows: This invention utilizes a SAR deception jamming image generation method based on fractional Fourier transform to achieve refined generation of SAR deception jamming images containing false targets. By performing azimuth-direction fractional Fourier transform processing on the image to achieve defocusing under target motion conditions, and by fusing the image background spectrum and using histogram matching based on the statistical characteristics of the real target amplitude, the problem of large differences in amplitude characteristics between the defocused false target image and the real target, and obvious background fragmentation, is effectively avoided, thus improving the refinement and realism of the false targets in the image. Attached Figure Description
[0012] Figure 1 This is a flowchart of the method of the present invention.
[0013] Figure 2 This is the original SAR image.
[0014] Figure 3 This is a false target template.
[0015] Figure 4 This is a two-dimensional distance spectrum of a false target.
[0016] Figure 5 The image shows the two-dimensional distance spectrum of the dummy target after time-frequency rotation.
[0017] Figure 6 This is a two-dimensional distance spectrum diagram after background spectrum fusion.
[0018] Figure 7 This is a normalized magnitude histogram for the reference region.
[0019] Figure 8 This is a template image of the false target after time-frequency rotation and spectral processing.
[0020] Figure 9 SAR image with defocused decoys added. Detailed Implementation
[0021] The present invention will be further described below with reference to the accompanying drawings and embodiments.
[0022] This invention proposes a method for generating false targets in the SAR image domain based on time-frequency rotation. The Fractional Fourier Transform (FRFT) can achieve rotation at any angle in the time-frequency plane. Combining spectral processing methods such as FRFT with histogram matching allows for simultaneous adjustment of false target features at both the phase and amplitude levels, thereby generating more realistic false targets. This method involves local processing of SAR Level 1A SLC image data and embedding false target templates. Simultaneously, the FRFT is used to control the defocusing degree of the false target in the azimuth direction. The fusion of the false target with the image background achieves refined generation of deception interference. Amplitude histogram matching ensures that the amplitude distribution of the false target is consistent with the real image, while preserving the target's phase characteristics. This generates false targets in the SAR image that closely approximate the real target in terms of geometry, defocusing degree, and statistical characteristics. While pre-implementation experiments for deception interference are costly, using the proposed technique to implement interference can assist in interpretation, verify the ability of deception interference identification methods, and reverse-engineer the interference signal generation process.
[0023] To achieve the above objectives, the technical solution of the present invention includes the following steps: Step 1: Receive Synthetic Aperture Radar Single-Look-Complex (SLC) image data and represent the image region to be processed in the SLC data as a complex matrix. Simultaneously input the SLC image template of the dummy target. , use fake target template After converting to a complex matrix, it is clipped to a size of The matrix.
[0024] Step 2: In The desired region of interest (ROI) is determined for pseudo-target fusion. The pseudo-target template obtained in step 1 is then used... Calculate its amplitude statistical characteristics, and calculate the detection threshold based on the mean and standard deviation. For fake target templates Filtering is performed to remove background noise components from the target spectrum, so that the background features of the selected ROI can be fused into the false target template later. Detection threshold. It can be represented as: (1) (2) (3) in, It's a template. The numerical value of the element in the middle. It is the mean of all elements. Standard deviation of all elements The total number of elements is determined using a detection threshold. Determine if a given element is background noise, and set the background noise to zero to obtain the template. .
[0025] (4) That is, when the template The value in Greater than the detection threshold It is determined to be the target element at that time.
[0026] Step 3: The size obtained in Step 2 is False target template Perform time-frequency rotation. Perform a fractional Fourier transform along the azimuth direction. Adjust the order of the FrFT. The value of can produce defocus variations ranging from slight to significant, thus simulating the motion characteristics of the target under different imaging conditions. The formula for calculating the fractional Fourier transform can be expressed as: (5) (6) in, It is the kernel function of FrFT. Represents a one-dimensional signal In the FrFT domain sampling points, It is a natural constant. The imaginary unit; Represents a time variable. Indicates changing the angle. This indicates the order of the fractional Fourier transform.
[0027] Step 4: Perform a range-to-Fourier transform on the fake target template after time-frequency rotation in Step 3 to obtain the two-dimensional range spectrum in the range frequency domain and azimuth time domain. Meanwhile, for the ROI selected in step 2, the size is... Perform a range-to-Fourier transform on the image background region to obtain the range spectrum of the reference background. The range spectrum of the false target template and the range spectrum of the reference background are fused according to the energy distribution ratio. This allows for the adjustment of the spectral characteristics of the false target. The formula for generating the spectral fusion is as follows: (7) in, This is the image distance spectral energy ratio coefficient. The reference background distance spectral energy ratio coefficient.
[0028] Step 5: Analyze the false target range spectrum obtained in Step 4. The false target template Z, after fusing background features, is obtained through inverse Fourier transform. A reference region R is selected, amplitude normalization is performed, and the amplitude histogram is calculated. Histogram sequence interval index The calculation method is as follows: (8) (9) in, This represents the floor function. This represents the normalized amplitude over the reference interval. Indicates the length of the histogram interval. This represents the number of intervals in the histogram. When When the value is 1, it is classified into the last interval.
[0029] Step 6: Use the histogram obtained in Step 5 Histogram matching is performed on the false targets to correct the amplitude statistics of the defocused false target template. Simultaneously, the phase component *f* of the pre-matched false target template is added azimuthally to ensure that the imaging coherence remains consistent with the target morphology. Finally, the processed false target template is fused into the real image to obtain a SAR image containing the defocused false targets. The mapping function of histogram matching can be expressed as: (10) in, Let be the cumulative distribution function of the image to be matched. For the first The number of pixels in each interval For reference histogram The cumulative distribution function, It is the histogram index of the image to be matched. It is the histogram index of the matched image.
[0030] Example: Step 1: Figure 1 The diagram shows a flowchart of a SAR image domain false target generation method based on time-frequency rotation according to an embodiment of the present invention. Measured SLC image data of SAR is received, and a portion of the SAR image with the defocused false target region added is taken, denoted as... Size is ,like Figure 2 As shown. Simultaneously receive fake target templates. ,like Figure 3 As shown.
[0031] Step 2: Select the ROI with added false targets in the SAR image obtained in Step 1. The size of the ROI should be the same as the size of the false target template. Simultaneously, the false target template obtained in step 1... Perform filtering, calculate the mean and standard deviation of the amplitude, and calculate the filter mask matrix based on statistical characteristics. Background noise points below a certain threshold are filtered out to allow for the subsequent fusion of background features of the selected ROI into the false target template. Detection threshold It can be represented as: (1) (2) (3) in, It's a template. The numerical value of the element in the middle. It is the mean of all elements. Standard deviation of all elements The total number of elements is determined using a detection threshold. Determine if a given element is background noise, and set the background noise to zero to obtain the template. .
[0032] (4) That is, when the template The value in Greater than the detection threshold If the value is determined to be a target image element, its original value is retained; if it is less than or equal to the detection threshold, it is considered a target image element. When a point is identified as background noise, its value is replaced with 0. The template matrix after noise removal based on mean and standard deviation is denoted as... .
[0033] Step 3: The size obtained in Step 2 is The dummy target template is rotated in time and frequency. A fractional Fourier transform (FrFT) is performed along the azimuth direction. The order of the FrFT is set according to the motion state requirements of the dummy target. The value of is used to achieve defocusing caused by target motion through azimuth-wise FrFT processing. In this example, the FrFT order is selected as 0.05. The formula for calculating the fractional Fourier transform can be expressed as: (5) (6) in, It is the kernel function of FrFT. Represents a one-dimensional signal In the FrFT domain sampling points, It is a natural constant. It is the imaginary unit. Represents a time variable. Indicates changing the angle. This indicates the order of the fractional Fourier transform.
[0034] Step 4: Apply Fast Fourier Transform (FFT) to the false target template processed by FrFT in Step 3 to obtain the two-dimensional spectrum of the image in the range frequency domain and the orientation time domain. ,like Figure 5 As shown. Simultaneously, a range-to-Fourier transform is performed on the background region near the ROI selected in step 2 that has the same size as the dummy target template, yielding the range spectrum of the reference background. The distance spectrum of the false target template and the distance spectrum of the reference background are fused according to their respective scaling factors. This allows for the adjustment of the spectral characteristics of the false target. The formula for generating the spectral fusion is as follows: (7) in, This is the image distance spectral energy ratio coefficient. The reference background distance spectral energy scaling factor is used. In this example, all selected coefficients are 0.9. The spectrum after background fusion is shown below. Figure 6 As shown.
[0035] Step 5: Analyze the range spectrum of the false target template obtained in Step 4. The false target image Z, obtained by inverse Fourier transform and incorporating background features, is then processed. A reference region R is selected, amplitude normalization is performed, and the histogram is calculated. ,like Figure 7 As shown, the processed false targets are subsequently corrected based on the reference histogram. Histogram sequence interval index. The calculation method is as follows: (8) (9) in, This represents the floor function. This represents the normalized amplitude over the reference interval. Indicates the length of the histogram interval. This represents the number of intervals in the histogram. When When the value is 1, it is classified into the last interval. This example divides the normalized magnitude histogram into 256 intervals.
[0036] Step 6: Use the histogram obtained in Step 5 Histogram matching is performed on the false targets to correct the amplitude statistical characteristics of the defocused false target template. Simultaneously, phase components of the pre-matched false target template are added azimuthally. This ensures that the imaging coherence matches the target morphology. The mapping function for histogram matching can be expressed as: (10) in, Let be the cumulative distribution function of the image to be matched. For the first The number of pixels in each interval For reference histogram The cumulative distribution function, It is the histogram index of the image to be matched. It is the histogram index of the matched image.
[0037] Finally, the processed false target template is fused into the real image to obtain a SAR image containing defocused false targets. The processed false target template is shown below. Figure 8 As shown, the SAR image after adding false targets is as follows: Figure 9 As shown.
Claims
1. A method for generating false targets in the SAR image domain based on time-frequency rotation, characterized in that, Includes the following steps: Step 1: Receive synthetic aperture radar single-view complex SLC image data and represent the image region to be processed in the SLC image data as a complex matrix. Simultaneously input the SLC image template of the dummy target. , use fake target image template After converting to a complex matrix, it is clipped to a size of Matrix; Step 2: In The required Region of Interest (ROI) is determined for pseudo-target fusion; For the false target image template obtained in step 1 Calculate its amplitude statistical characteristics, and calculate the detection threshold based on the mean and standard deviation. For false target image templates Perform filtering to remove background noise components from the target spectrum; Detection threshold Represented as: （1） （2） （3） in, It's a template. The numerical value of the element in the middle. It is the mean of all elements. The standard deviation of all elements The total number of elements; Using detection thresholds Determine if a given element is background noise, and set the background noise to zero to obtain the template. : （4） That is, when the template The value in Greater than the detection threshold It is then determined to be the target element; Step 3: The size obtained in Step 2 is False target image template Perform time-frequency rotation; perform fractional Fourier transform along the azimuth direction; Step 4: Perform a range-to-Fourier transform on the fake target template after time-frequency rotation in Step 3 to obtain the two-dimensional range spectrum in the range frequency domain and azimuth time domain. Simultaneously, for the ROI selected in step 2, the size is... Perform a range-to-Fourier transform on the image background region to obtain the reference background range spectrum. The two-dimensional range spectrum of the false target and the range spectrum of the reference background are fused according to the energy distribution ratio. This allows for the adjustment of the spectral characteristics of the false target; the formula for generating the spectral fusion is as follows: （7） in, This is the image distance spectral energy ratio coefficient. The reference background distance spectral energy ratio coefficient; Step 5: Analyze the two-dimensional range spectrum of the false target obtained in Step 4. The false target template Z, after fusing background features, is obtained through inverse Fourier transform; a reference region R is selected, amplitude normalization is performed, and the amplitude histogram is calculated. Histogram sequence interval index The calculation method is as follows: （8） （9） in, This represents the floor function. This represents the normalized amplitude over the reference interval. Indicates the length of the histogram interval. Represents the number of intervals in a histogram; when When the value is 1, it is classified into the last interval; Step 6: Use the histogram obtained in Step 5 Histogram matching is performed on the false targets to correct the amplitude statistical characteristics of the defocused false target template. Simultaneously, the phase component φ of the pre-matched false target template is added azimuthally to ensure that the imaging coherence remains consistent with the target morphology. Finally, the processed false target template is fused into the real image to obtain a SAR image containing the defocused false targets. The mapping function for histogram matching is expressed as: （10） in, Let be the cumulative distribution function of the image to be matched. For the first The number of pixels in each interval For reference histogram The cumulative distribution function, It is the histogram index of the image to be matched. It is the histogram index of the matched image.
2. The method for generating false targets in the SAR image domain based on time-frequency rotation according to claim 1, characterized in that, The formula for calculating the fractional Fourier transform is expressed as follows: （5） （6） in, It is the kernel function of FrFT. Represents a one-dimensional signal In the FrFT domain sampling points, It is a natural constant. The imaginary unit; Represents a time variable. Indicates changing the angle. This indicates the order of the fractional Fourier transform.
3. An electronic device, characterized in that, include: Processor and memory; The memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory to cause the electronic device to perform the method as described in any one of claims 1 to 2.
4. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 2.
5. A chip, characterized in that, include: A processor for retrieving and running a computer program from memory, causing a device on which the chip is mounted to perform the method as described in any one of claims 1 to 2.
6. A computer program product, characterized in that, The computer program product includes a computer storage medium storing a computer program, the computer program including instructions executable by at least one processor, which, when executed by the at least one processor, implement the method as described in any one of claims 1 to 2.