City DSM generation method and device for four-star formation single-pass SAR

Abstract

This invention discloses a method and apparatus for generating urban DSMs using single-pass SAR for four-satellite formations, belonging to the field of radar measurement technology. The method includes: data preprocessing, performing precise registration on the complex SAR image of the four-satellite formation and removing terrain phase; homogeneous pixel extraction, determining a set of homogeneous pixels for each pixel as an adaptive multi-view window based on amplitude information; structural phase compensation, estimating and compensating for the non-stationary structural phase of strong targets; target height calculation, constructing a covariance matrix using the compensated interferogram and calculating pixel height using spectral estimation; outlier removal based on energy entropy dispersion index, filtering and removing outlier pixels; and geocoding, combining target height and orbital information to solve for latitude and longitude, generating the final DSM. This invention achieves high-precision, high-resolution urban DSM reconstruction using only a single-pass SAR data, effectively avoiding temporal decorrelation problems and significantly improving surveying efficiency and product quality.

Description

Technical Field

[0001] This invention belongs to the field of radar measurement technology, specifically relating to a method and apparatus for generating urban DSMs for single-pass SAR of four-star formations. Background Technology

[0002] Synthetic Aperture Radar (SAR) is a radar technology that uses microwaves to achieve high-resolution imaging. It is capable of operating in all weather conditions and is widely used for acquiring three-dimensional information about the Earth's surface, particularly in areas such as urbanization monitoring and disaster prevention and mitigation. Digital Surface Models (DSMs) can reflect surface elevation information, including buildings and vegetation; their efficient and high-precision generation is currently a research hotspot in the SAR field.

[0003] Traditional spaceborne SAR systems typically rely on repeated orbit observations to acquire the multi-angle information needed to generate a DSM, meaning the satellite needs to fly over the same area multiple times at different times. This approach has significant drawbacks: first, the data acquisition cycle is long, ranging from months to years, making it difficult to meet the requirements for rapid response; second, ground targets may change within the intervals between multiple observations (such as building construction or vegetation growth), leading to so-called "temporal decorrelation," which severely reduces the accuracy and reliability of interferometry.

[0004] To address the aforementioned issues, existing technologies primarily employ two types of methods for DSM reconstruction. One type is the multi-baseline interferometry method, which estimates elevation by processing interferograms under different vertical baselines. This method is relatively simple to process and requires a small number of images. However, in urban areas, complex scattering caused by buildings and other structures leads to spatially non-stationary interferometric phase characteristics. Traditional multi-look averaging can induce coherent cancellation, resulting in phase estimation bias, ultimately manifesting as blurred DSM and numerous outliers. The other type is the tomographic SAR method, which can focus in the height dimension, theoretically providing more accurate elevation estimates. However, this method heavily relies on a large, long-term image stack to build sufficient observation redundancy. When the number of images is limited (e.g., during a single flight), its height dimension resolution and focusing performance deteriorate sharply, leading to problems such as sidelobe elevation and unstable peak values.

[0005] Therefore, the core advantage of the new generation of four-satellite formation SAR systems lies in their ability to simultaneously acquire multiple SAR images with spatial baselines during a single flight, fundamentally avoiding temporal decoherence. However, how to utilize this sparse data with a limited number of images and constrained observation geometry to stably generate high-precision, high-resolution urban DSMs while overcoming the shortcomings of traditional methods remains a pressing technical challenge. Summary of the Invention

[0006] To address the aforementioned technical issues, this invention provides a method and apparatus for generating urban DSMs using single-pass SAR for four-star formations. This method enables high-precision, high-resolution urban DSM reconstruction using only a single pass of data, effectively avoiding temporal decorrelation issues and significantly improving surveying efficiency and product quality.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0008] A method for generating city DSMs for single-pass SAR transit of four-star formations, the method comprising:

[0009] Step 1: Acquire four complex SAR images collected during the single-pass flight of the four-star formation and perform preprocessing;

[0010] Step 2: Based on the four preprocessed complex SAR images, a set of homogeneous pixels is selected for each center pixel within a fixed estimation window using the amplitude consistency discrimination method.

[0011] Step 3: For strong target pixels whose amplitude exceeds the threshold, estimate the phase gradient using a multi-baseline interferogram based on the set of homogeneous pixels corresponding to the strong target pixels, and compensate for the structural phase of each pixel.

[0012] Step 4: Construct the covariance matrix based on the multi-baseline interferogram and amplitude map obtained after compensating the phase of the structure, and calculate the target height for each pixel;

[0013] Step 5: Calculate the energy entropy dispersion index of each pixel based on the power spectrum function about height constructed when calculating the target height, and filter valid pixels based on the energy entropy dispersion index;

[0014] Step 6: Solve for the latitude and longitude of each effective pixel using the distance-Doppler equation to generate a digital surface model (DSM) in geographic coordinate system.

[0015] Furthermore, the preprocessing in step 1 includes: using the complex SAR image of one of the satellites as the main image, registering the complex SAR images of the other three satellites to the main image using the satellite's precise orbit information, SAR data parameters, and an external coarse digital elevation model, and performing interferometric processing on all registered image combinations to generate a multi-baseline interferogram and remove the terrain phase from it.

[0016] Furthermore, the amplitude consistency discrimination method in step 2 specifically involves: calculating the average amplitude of each pixel to be compared within the fixed estimation window in the four preprocessed images; if the ratio of the average amplitude of the pixel to be compared to the average amplitude of the pixel to be extracted in the four preprocessed images is within a preset threshold range, then the pixel to be compared is determined to be a homogeneous pixel of the pixel to be extracted.

[0017] Furthermore, step 3 includes: for each strong point target pixel, calculating the median phase gradient of all adjacent pixels in its homogeneous pixel set; based on the blur height and phase gradient statistical variance of different baseline interferograms, converting the median phase gradient into a height gradient, then performing weighted fusion on the height gradient to obtain the optimal height gradient estimate, and then converting the optimal height gradient estimate into the optimal phase gradient estimate; based on the optimal phase gradient estimate, calculating the structural phase of each pixel in the homogeneous pixel set, and subtracting it from the interferometric phase of the terrain phase removal before multi-view averaging processing.

[0018] Furthermore, the calculation of the target height of each pixel in step 4 specifically includes: processing the covariance matrix using the spectral estimation method to calculate the power spectral values ​​corresponding to different assumed heights; and determining the height value corresponding to the maximum value of the power spectral function as the target height of the pixel.

[0019] Furthermore, step 5 includes: calculating the energy entropy dispersion index value of each pixel according to the calculation formula of the energy entropy dispersion index; comparing the index value with a preset threshold, removing pixels with index values ​​lower than the preset threshold, and retaining the remaining pixels as valid pixels.

[0020] Furthermore, step 6 includes: substituting the target height, slant range information, azimuth time information, and satellite orbit parameters of each effective pixel into the range-Doppler equation to obtain the latitude and longitude coordinates corresponding to the effective pixel; and combining the latitude and longitude coordinates and elevation information of all effective pixels to generate a digital surface model in a geographic coordinate system.

[0021] On the other hand, the present invention provides a city DSM generation apparatus for single-pass SAR of four-star formations, comprising:

[0022] The preprocessing module is used to acquire and preprocess four complex SAR images collected during the single-pass flight of the four-star formation.

[0023] The first filtering module is used to filter a set of homogeneous pixels for each center pixel within a fixed estimation window based on the four preprocessed complex SAR images and using the amplitude consistency discrimination method.

[0024] The compensation module is used to estimate the phase gradient of strong target pixels whose amplitude exceeds a threshold, based on the set of homogeneous pixels corresponding to the strong target pixels, using a multi-baseline interferogram, and to compensate for the structural phase of each pixel.

[0025] The calculation module is used to construct the covariance matrix based on the multi-baseline interferogram and amplitude map obtained after compensating the phase of the structure, and to calculate the target height for each pixel;

[0026] The second filtering module is used to calculate the energy entropy dispersion index of each pixel based on the power spectrum function about height constructed when calculating the target height, and to filter valid pixels based on the energy entropy dispersion index.

[0027] The generation module is used to solve for the latitude and longitude of each effective pixel using the distance-Doppler equation, and generate a digital surface model (DSM) in geographic coordinate system.

[0028] Thirdly, the present invention provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs; wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the aforementioned method for generating city DSMs for single-pass SAR of four-star formations.

[0029] Fourthly, the present invention provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, enable the processor to implement the aforementioned method for generating city DSMs for single-pass SAR of four-star formations.

[0030] The beneficial effects of this invention are as follows:

[0031] First, this invention significantly improves data acquisition and processing efficiency. By utilizing complex SAR images acquired synchronously during a single overflight of a four-satellite formation, urban DSM reconstruction can be achieved, completely eliminating the reliance on traditional repeated orbits and long-term sequential observations. Compared to traditional methods, this method reduces data acquisition time from months or even years to within minutes of a single overflight, greatly improving surveying response speed and operational efficiency.

[0032] Secondly, this invention effectively improves the density and accuracy of DSM products. By introducing a homogeneous pixel extraction step, an adaptive multi-view window is constructed for each pixel, improving sample consistency. In particular, for strong point targets in urban scenes, a structural phase compensation technique is innovatively adopted to effectively suppress the coherent cancellation effect caused by non-stationary structural phase during multi-view reconstruction, thereby significantly improving the stability of high-resolution inversion and the grid resolution of the final DSM, achieving high-precision, high-resolution reconstruction.

[0033] Finally, this invention enhances the reliability and quality of DSM products. By calculating and utilizing the energy entropy dispersion index for outlier removal, this method can fully leverage the multidimensional information of multi-baseline data for internal quality control during the high-resolution stage, rather than post-processing filtering. This allows for more effective identification and removal of unreliable pixels, significantly reducing outliers in the final DSM and improving the overall consistency of the product. Attached Figure Description

[0034] Figure 1This is a flowchart of the city DSM generation method for single-pass SAR of four-star formations according to the present invention;

[0035] Figure 2 This is a diagram of the currently publicly available Copernicus DSM results;

[0036] Figure 3 This is a diagram of the DSM reconstruction results based on the traditional multi-baseline interferometric SAR method;

[0037] Figure 4 This is a diagram of the DSM reconstruction result based on the method of this invention. Detailed Implementation

[0038] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0039] like Figure 1 As shown, this invention proposes a method for generating city DSMs for single-pass SAR of four-star formations. The specific steps are as follows:

[0040] Step 1, Data Preprocessing: Register the four complex SAR images acquired during the single-flight transit of the four-satellite formation to the main image and remove terrain phase; including:

[0041] After acquiring the SAR data and parameters of the four-satellite formation, a radiometrically corrected complex SAR image of the four-satellite formation was input. Four SAR satellite images were acquired by the four satellites (A, B, C, and D), with the image acquired by satellite A used as the master image. Using precise satellite orbit information, SAR data parameters, and an external coarse digital elevation model, the images acquired by satellites B, C, and D were registered onto the image of satellite A. The registered images were then interferometrically analyzed pairwise to form interferogram combinations: AB, AC, AD, BC, BD, and CD. The terrain phase in the interferograms was removed using the external digital elevation model. The interferometry of two complex numbers p and q was achieved through the conjugate multiplication of p and q.

[0042] Step 2, Homogeneous Pixel Extraction: Based on the four preprocessed complex SAR images, a set of homogeneous pixels is selected for each center pixel within a fixed estimation window using the amplitude consistency discrimination method.

[0043] For the four precisely registered and preprocessed complex SAR images input, the amplitude is taken, given a size of... The estimation window is defined by using each pixel in the image to be extracted as the center of the estimation window. Pixels surrounding the pixel that are within the estimation window are compared to determine whether they are homogeneous pixels. Taking the parameter discrimination method as an example, a pixel to be extracted is selected. Location , Indicates the orientation of the image. Indicates the distance position of the image, for all images (four images in total) in The average value of the position is taken as the mean value. ;for A pixel to be compared within the homogeneous pixel estimation window. , The pixel position is in For all images (four images in total) in The average value of the position If the conditions are met Then determine for Homogeneous pixels. Among them and To set a threshold. The set of homogeneous pixels is represented as .

[0044] Step 3, Structural Phase Compensation: For strong target pixels whose amplitude exceeds the threshold, the phase gradient is estimated using a multi-baseline interferogram based on the set of homogeneous pixels corresponding to the strong target pixels, and the structural phase of each pixel is compensated.

[0045] Below, we will take the following as an example. Pixels of position For example, calculate its structural phase. Assume... The amplitude value exceeds the amplitude threshold, which meets the requirements for strong point targets.

[0046] Define pixel and the set of homogeneous pixels The union formed is For all the interferograms described in step 1, in The position of Height gradient estimated by amplitude interferogram Represented as:

[0047] (1)

[0048] in, Indicates the first In the interferogram, the union The set of interference phase gradients between any two adjacent points. This indicates taking the median value. For the first In the interferogram, the union The corresponding set of wavenumbers. Since gradient calculation has two directions, namely the azimuth and range directions along the image, the method for calculating the gradient along both directions is the same. Here, we will describe the method for calculating the gradient along one of the directions as an example.

[0049] The optimal weighted estimate Represented as:

[0050] (2)

[0051] in, The total number of interferograms, weights Represented as:

[0052] (3)

[0053] in, A set of phase gradients of homogeneous pixels The statistical variance Indicates the first The fuzziness height of the amplitude interferogram The normalization coefficient is expressed as:

[0054] (4)

[0055] Calculate the first Phase gradient of amplitude interferogram , is represented as:

[0056] (5)

[0057] in, Indicates the carrier frequency wavelength of radar electromagnetic waves. Indicates the slant distance. Indicates the angle of incidence. Indicates the first The vertical baseline length of the amplitude interferogram. It represents pi (π).

[0058] As mentioned above, since the phase gradient can be calculated along two directions, the calculation along the azimuth direction is defined. Record Calculated along the distance direction Record .

[0059] Finally, for the union One of them in the image Location samples In its first Structure phase in amplitude interferogram Represented as:

[0060] (6)

[0061] in, Let be a constant, such that the first Structure phase in amplitude interferogram In sample pixels The value is 0 when the position is active.

[0062] Subsequently, multi-view interferogram based on homogeneous pixels was performed, the first... Aspect interferogram in sample pixels The interferogram of position is represented as , Multi-view results are denoted as As a normalized interferogram sample, we have:

[0063] (7)

[0064] in, Represent an imaginary number, satisfying Define normalized interferogram samples. Formed by the interference of images X and Y, it can also be denoted as For example, the interference of images A and B can form a normalized interferogram sample that can be denoted as... ; and The images X and Y that form the interferogram are respectively represented in... The amplitude value of the position.

[0065] Step 4: Target Height Calculation: Based on the interferogram and amplitude map obtained from the above steps, construct the covariance matrix. Calculate the target height for each pixel using spectral estimation methods, taking the Capon spectral estimation method as an example. Power Spectrum Function The expression is:

[0066] (8)

[0067] Wherein, the guiding vector matrix , It is the natural logarithm. This indicates the conjugate transpose. The column vector is composed of the vertical wavenumbers of each interferogram. It is an integer index variable, with a value range of 1 to 1. , Possible target height values Quantity, This represents the normalized covariance matrix. Represented as:

[0068] (9)

[0069] in, The subscript letters XY (such as AB) represent the interferogram formed by the interference of images X and Y. The maximum height value is the final target height value.

[0070] Step 5: Outlier Removal Based on Energy Entropy Dispersion Index. Calculate the energy entropy dispersion index for each pixel based on the power spectrum function constructed when calculating the target height, and then filter valid pixels based on this index.

[0071] Energy Entropy Dispersion Index The calculation expression is:

[0072] (10)

[0073] in, This represents the logarithm with base 2, and the coefficient is... The expression is:

[0074] (11)

[0075] In particular, it is explained again It is an integer index variable, with a value range of 1 to 1. ,symbol Indicates the first One possible target height value. Based on this, the coefficient... Indicates by The uniquely determined coefficients calculated by equation (11), the power spectrum normalization coefficients Represented as:

[0076] (12)

[0077] Pixels with positions smaller than a given threshold are identified as abnormal pixels and discarded.

[0078] Step 6, Geocoding: Solve for the latitude and longitude of each effective pixel using the distance-Doppler equation to generate a digital surface model (DSM) in a geographic coordinate system.

[0079] The target height value has been calculated in the above steps. Furthermore, since the slant range, azimuth time, and satellite orbit information of each pixel in the image are known, the longitude and latitude information of each pixel can be solved by substituting them into the range-Doppler equation, thereby achieving geocoding, obtaining the coordinates of the target in the geodetic coordinate system, and thus obtaining the DSM.

[0080] Example:

[0081] A four-star formation commercial multi-baseline SAR system acquires four SAR images in a single observation and is designed to generate a 1:50,000 scale digital terrain model (DSM) (25-meter resolution grid). However, at this resolution, urban target morphology is difficult to identify. Based on the method of this invention, high-precision, high-resolution urban DSM reconstruction is generated from images acquired in a single flight. Taking the data of the Beijing Capital International Airport area as an example, this area has complex building features, and the Copernicus DSM, such as... Figure 2 As shown, the DSM reconstructed by the traditional multi-baseline interferometric SAR method is as follows: Figure 3 As shown, the city DSM obtained by processing using the method of the present invention is as follows: Figure 4 As shown, the DSM obtained by the method of this invention can finely characterize the structural features of the airport, with a grid resolution better than 3 m, which is significantly better than the 30 m grid resolution of the Copernicus DSM and the 10 m grid resolution of the DSM reconstructed by the traditional multi-baseline interferometric SAR method. Therefore, the method of this invention not only achieves higher grid resolution but also correctly reconstructs information about surrounding buildings.

[0082] On the other hand, the present invention provides a city DSM generation device for single-pass SAR of four-star formations, the various modules of which can implement the various steps of the aforementioned method, specifically including:

[0083] The preprocessing module is used to acquire and preprocess four complex SAR images collected during the single-pass flight of the four-star formation.

[0084] The first filtering module is used to filter a set of homogeneous pixels for each center pixel within a fixed estimation window based on the four preprocessed complex SAR images and using the amplitude consistency discrimination method.

[0085] The compensation module is used to estimate the phase gradient of strong target pixels whose amplitude exceeds a threshold, based on the set of homogeneous pixels corresponding to the strong target pixels, using a multi-baseline interferogram, and to compensate for the structural phase of each pixel.

[0086] The calculation module is used to construct the covariance matrix based on the multi-baseline interferogram and amplitude map obtained after compensating the phase of the structure, and to calculate the target height for each pixel;

[0087] The second filtering module is used to calculate the energy entropy dispersion index of each pixel based on the power spectrum function about height constructed when calculating the target height, and to filter valid pixels based on the energy entropy dispersion index.

[0088] The generation module is used to solve for the latitude and longitude of each effective pixel using the distance-Doppler equation, and generate a digital surface model (DSM) in geographic coordinate system.

[0089] Thirdly, the present invention provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs; wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the aforementioned method for generating city DSMs for single-pass SAR of four-star formations.

[0090] Fourthly, the present invention provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, enable the processor to implement the aforementioned method for generating city DSMs for single-pass SAR of four-star formations.

[0091] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for generating city DSMs for single-pass SAR transit of four-star formations, characterized in that, The method includes: Step 1: Acquire four complex SAR images collected during the single-pass flight of the four-star formation and perform preprocessing; Step 2: Based on the four preprocessed complex SAR images, a set of homogeneous pixels is selected for each center pixel within a fixed estimation window using the amplitude consistency discrimination method. Step 3: For strong target pixels with amplitudes exceeding a threshold, estimate the phase gradient using a multi-baseline interferogram based on the homogeneous pixel set corresponding to the strong target pixel, and compensate for the structural phase of each pixel; this includes: for each strong target pixel, calculating the median phase gradient of all its neighboring pixels in the homogeneous pixel set; based on the blur height and phase gradient statistical variance of different baseline interferograms, converting the median phase gradient into a height gradient, then performing weighted fusion on the height gradient to obtain the optimal height gradient estimate, and then converting the optimal height gradient estimate into the optimal phase gradient estimate; based on the optimal phase gradient estimate, calculating the structural phase of each pixel in the homogeneous pixel set, and subtracting it from the interferometric phase after terrain phase removal before multi-view averaging; Step 4: Construct the covariance matrix based on the multi-baseline interferogram and amplitude map obtained after compensating the phase of the structure, and calculate the target height for each pixel; Step 5: Calculate the energy entropy dispersion index of each pixel based on the power spectrum function constructed when calculating the target height, and filter valid pixels based on the energy entropy dispersion index; including: calculating the energy entropy dispersion index value of each pixel according to the calculation formula of the energy entropy dispersion index; comparing the index value with a preset threshold, removing pixels with index values ​​lower than the preset threshold, and retaining the remaining pixels as valid pixels; wherein, the energy entropy dispersion index... The calculation expression is: （10） in, This represents the logarithm with base 2, and the coefficient is... The expression is: （11） in, It is an integer index variable, with a value range of 1 to 1. ,symbol Indicates the first One possible target height value, Possible target height values Quantity, The power spectrum function, These are the power spectrum normalization coefficients; Step 6: Solve for the latitude and longitude of each effective pixel using the distance-Doppler equation to generate a digital surface model (DSM) in geographic coordinate system.

2. The method for generating a city DSM for single-pass SAR of a four-star formation according to claim 1, characterized in that, The preprocessing in step 1 includes: using the complex SAR image of one of the satellites as the main image, registering the complex SAR images of the other three satellites to the main image using the satellite's precise orbit information, SAR data parameters, and an external coarse digital elevation model, and performing interferometric processing on all registered image combinations to generate a multi-baseline interferogram and remove the terrain phase.

3. The method for generating city DSMs for single-pass SAR of four-star formations according to claim 1, characterized in that, The amplitude consistency discrimination method in step 2 is as follows: calculate the average amplitude of each pixel to be compared in the four preprocessed images within the fixed estimation window. If the ratio of the average amplitude of the pixel to be compared to the average amplitude of the pixel to be extracted in the four preprocessed images is within a preset threshold range, then the pixel to be compared is determined to be a homogeneous pixel of the pixel to be extracted.

4. The method for generating a city DSM for single-pass SAR of a four-star formation according to claim 1, characterized in that, The calculation of the target height of each pixel in step 4 specifically includes: processing the covariance matrix using the spectral estimation method, calculating the power spectral values ​​corresponding to different assumed heights; and determining the height value corresponding to the maximum value of the power spectral function as the target height of the pixel.

5. The method for generating a city DSM for single-pass SAR of a four-star formation according to claim 1, characterized in that, Step 6 includes: substituting the target height, slant range information, azimuth time information, and satellite orbit parameters of each effective pixel into the range-Doppler equation to obtain the latitude and longitude coordinates corresponding to the effective pixels; combining the latitude and longitude coordinates and elevation information of all effective pixels to generate a digital surface model in the geographic coordinate system.

6. A city DSM generation apparatus for single-pass SAR of four-star formations, used to perform the method according to any one of claims 1-5, characterized in that, include: The preprocessing module is used to acquire and preprocess four complex SAR images collected during the single-pass flight of the four-star formation. The first filtering module is used to filter a set of homogeneous pixels for each center pixel within a fixed estimation window based on the four preprocessed complex SAR images and using the amplitude consistency discrimination method. The compensation module is used to estimate the phase gradient of strong target pixels whose amplitude exceeds a threshold, based on the set of homogeneous pixels corresponding to the strong target pixels, using a multi-baseline interferogram, and to compensate for the structural phase of each pixel. The calculation module is used to construct the covariance matrix based on the multi-baseline interferogram and amplitude map obtained after compensating the phase of the structure, and to calculate the target height for each pixel; The second filtering module is used to calculate the energy entropy dispersion index of each pixel based on the power spectrum function about height constructed when calculating the target height, and to filter valid pixels based on the energy entropy dispersion index. The generation module is used to solve for the latitude and longitude of each effective pixel using the distance-Doppler equation, and generate a digital surface model (DSM) in geographic coordinate system.

7. An electronic device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When one or more programs are executed by the one or more processors, the one or more processors implement the city DSM generation method for single-pass SAR of four-star formation as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, It stores executable instructions that, when executed by a processor, enable the processor to implement the city DSM generation method for single-pass SAR of four-star formation as described in any one of claims 1-5.

Images