A method and system for correcting atmospheric disturbances of ground-based synthetic aperture slope radar

Abstract

The present invention relates to the field of radar survey technology, and more specifically, to a method and system for correcting atmospheric disturbances of a ground-based synthetic aperture slope radar, comprising: collecting radar imaging, performing phase unwrapping on the radar imaging, and obtaining PS points; eliminating the PS points using an improved morphological opening operation, obtaining a stable PS point distribution, and completing deformation monitoring and realizing deformation inversion based on the stable PS point distribution; the improved morphological opening operation comprises: obtaining an outlier value of a PS point using an outlier detection algorithm, and correcting the outlier value by the distance value between the PS point and its nearest PS point. The present invention analyzes the data distribution of PS points to obtain the core radius of morphological processing corresponding to different PS points, realizes adaptive morphological processing, realizes the elimination of isolated PS points, obtains more stable PS points, and improves the accuracy of deformation inversion.

Description

Technical Field

[0001] The present invention relates to the field of radar survey technology, and more specifically, to a method and system for correcting atmospheric disturbances of a ground-based synthetic aperture slope radar. Background Art

[0002] Ground-based synthetic aperture radar is derived from spaceborne synthetic aperture radar (GB-SAR) and is a microwave active detection radar technology. Deformation monitoring based on GB-SAR interferometric technology is local, all-day, all-weather, fixed-point continuous, and has good flexibility and operability. It has extremely broad application prospects in deformation monitoring of landslides, glaciers, mines, buildings, volcanoes and dams. Factors affecting the accuracy of GB-SAR interferometric measurement include system frequency instability, atmospheric delay, changes in target scattering characteristics and noise. Among them, the atmospheric delay under the radar line of sight of two GB-SAR images is one of the main factors affecting the deformation measurement accuracy of GBSAR interferometric technology. Studies have shown that when the temperature is 20℃, a change of 1% in relative humidity at a distance of 1km from the radar can lead to a measurement error of 2mm. Therefore, eliminating the atmospheric delay under the radar line of sight is the key to improving the accuracy of GB-SAR interferometric deformation monitoring.

[0003] At present, GB-SAR atmospheric correction methods are mainly divided into two categories: one is a method based on meteorological data, which corrects the atmospheric phase according to the relationship between the refractive index of electromagnetic wave propagation and the temperature, air pressure and humidity data in the scene; the other is a method based on a function model, which usually assumes that the atmospheric influence is strongly correlated in space, that is, the atmosphere is uniform in space. At this time, the atmospheric phase compensation is completed by using the stable PS point interferometric phase through a parameterized model method.

[0004] In the related technology, for example, the Chinese patent application document with publication number CN117075056A discloses a ground-based synthetic aperture slope radar atmospheric correction method, which screens stable atmospheric estimation PS points through multi-threshold thresholds, and then uses an iterative algorithm to further eliminate unstable atmospheric estimation PS points to achieve accurate selection of higher quality PS points. On this basis, GB-SAR atmospheric correction is achieved through an atmospheric correction method based on PS points and parameterized models, which greatly improves the correction accuracy and meets the urgent needs of existing GB-SAR atmospheric correction technology for accurate correction of complex atmospheres in deformation monitoring.

[0005] However, when selecting stable PS points through thresholds, since the threshold settings are often fixed values ​​set based on personnel experience, and the distribution of PS points obtained under different terrains is different, the use of fixed values ​​to screen PS points is not accurate enough, resulting in low accuracy of deformation inversion. Summary of the invention

[0006] The present invention provides a method and system for correcting atmospheric disturbances of a ground-based synthetic aperture slope radar, aiming to solve the problem that in related technologies, threshold settings are often fixed values ​​set based on personnel experience, while the distribution of PS points obtained under different terrains is different, so the use of fixed values ​​to screen PS points is not accurate enough, resulting in low accuracy of deformation inversion.

[0007] In a first aspect, the present invention provides a method for correcting atmospheric disturbances of a ground-based synthetic aperture slope radar, comprising: acquiring radar imaging, performing phase unwrapping on the radar imaging, and obtaining PS points; eliminating the PS points using an improved morphological opening operation to obtain a stable PS point distribution, and completing deformation monitoring and realizing deformation inversion based on the stable PS point distribution; the improved morphological opening operation comprises: obtaining an outlier value of a PS point using an outlier detection algorithm, correcting the outlier value by the distance value between the PS point and its nearest PS point, and obtaining a corrected outlier value corresponding to the PS point, wherein the corrected outlier value is negatively correlated with the distance value; clustering the corrected outlier values ​​of each PS point to obtain multiple categories, and calculating the core radius of the PS point in the jth category when performing the opening operation. , the calculation formula is: ; In the formula, is the average distance value of all PS points in the jth category, is the scaling factor of the jth category; wherein the scaling factor is equal to the product of the local density of PS points in the jth category and the stretching factor, and the stretching factor is the ratio of the minimum value of all corrected outliers in the jth category in the histogram to the corrected outlier corresponding to the upper and lower difference critical points; the upper and lower difference critical points are the distance values ​​on the horizontal axis between the corrected outliers corresponding to the inflection points of the fitting function in the histogram and their minimum slope values, and the inflection point corresponding to the minimum distance value is selected as the upper and lower difference critical points.

[0008] The effect is that the PS points are eliminated by using the improved morphological opening operation, a stable PS point distribution is obtained, and the accuracy of deformation inversion is improved.

[0009] Furthermore, the method for obtaining the local density of PS points in the jth category includes: calculating the absolute value of the difference between the upper peak coefficient and the lower peak coefficient in the histogram corresponding to the jth category as the upper and lower difference coefficients; and using the upper and lower difference coefficients of the jth category to characterize the local density of PS points in the jth category.

[0010] The effect is that the histogram method reflects the function distribution of the number of times the data occurs, and the local density of the PS point can be accurately calculated.

[0011] Further, calculate the scaling factor of the jth category The calculation formula includes: ; In the formula, for The stretch coefficient relative to the upper and lower difference critical points, The normalized value of the minimum value of the outlier correction for all PS points in the jth category, is the upper and lower difference coefficient of the jth category, represents a step function, Represents a phase function.

[0012] Furthermore, the corrected outlier value of the PS point is calculated using the following formula: ; In the formula, is the corrected outlier value of the i-th PS point, is the distance between the ith PS point and its nearest PS point, is the outlier value of the i-th PS point, and exp() is the normalization function.

[0013] The effect is that for any PS point, the distance value between the PS point and the nearest PS point is used to correct the outlier value of the PS point. This is because the larger the distance value is, the farther the nearest PS point is in the vicinity of the PS point, and the lower the possibility that it is a stable PS point. Conversely, the smaller the distance value is, the higher the possibility that it is a stable PS point. Therefore, the calculation can be performed more accurately to obtain an accurate corrected outlier value.

[0014] Furthermore, calculating the distance value between the PS point and its nearest PS point includes: calculating the distance value between the PS point and its nearest PS point using a Euclidean distance calculation method.

[0015] The effect is that the distance value can be calculated conveniently and quickly using the Euclidean distance calculation method.

[0016] Furthermore, the method for determining the upper peak coefficient includes: for any statistical result, obtaining the difference between the statistical result and the maximum value, calculating the ratio of the absolute value of the difference to the maximum value, and subtracting the ratio from 1 to obtain the weight of the statistical result; calculating the ratio of the weighted sum of all statistical results and corresponding weights to the number of all statistical results to obtain the upper peak coefficient.

[0017] Furthermore, the method for determining the lower-peak coefficient includes: obtaining the lower-peak coefficient, including: for any statistical result, obtaining the difference between the statistical result and the minimum value, calculating the ratio of the absolute value of the difference to the minimum value, and subtracting the ratio from 1 to obtain the weight of the statistical result; calculating the ratio of the weighted sum of all statistical results and corresponding weights to the number of all statistical results to obtain the lower-peak coefficient.

[0018] Furthermore, the inflection point corresponding to the minimum distance value is selected as the upper and lower difference critical points, including: using the least squares method to perform polynomial fitting on the function distribution obtained by histogram statistics, and calculating the inflection point of the fitting function obtained by fitting to obtain the inflection point of the fitting function; calculating the slope value of the normalized corrected outlier value on the fitting function, and obtaining the corrected outlier value corresponding to the minimum slope value on the horizontal axis, calculating the distance value between the corrected outlier value corresponding to each inflection point and the corrected outlier value corresponding to the minimum slope value on the horizontal axis, and selecting the inflection point with the minimum distance value as the upper and lower difference critical points.

[0019] Further, calculating the scaling factor of the j-th category also includes: obtaining the minimum value of the corrected outliers corresponding to all PS points in the j-th category, and using the minimum value as the scaling factor of the j-th category, wherein the scaling factor is proportional to the core radius. Negatively correlated.

[0020] The effect is that the method for calculating the scaling factor of the j-th category in this embodiment has a simple calculation process, a small amount of data calculation, and improved calculation efficiency.

[0021] In a second aspect of the present invention, a ground-based synthetic aperture slope radar atmospheric disturbance correction system is also provided, comprising a processor and a memory, wherein the memory stores a computer program, and the processor executes the computer program to implement any of the above-described ground-based synthetic aperture slope radar atmospheric disturbance correction methods.

[0022] Beneficial effects: By analyzing the data distribution of PS points, the core radius of morphological processing corresponding to different PS points is obtained, adaptive morphological processing is realized, isolated PS points are eliminated, more stable PS points are obtained, and the accuracy of deformation inversion is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The following detailed description is read with reference to the accompanying drawings, which illustrate several embodiments of the present invention in an exemplary and non-limiting manner, and in which like or corresponding reference numerals represent like or corresponding parts, wherein:

[0024] Figure 1 is a flowchart schematically illustrating calculation of core radius according to an embodiment of the present invention;

[0025] Figure 2 FIG. 1 is a schematic diagram schematically showing a system structure according to an embodiment of the present invention.

[0026] Reference numerals

[0027] 1. Processor; 2. Communication bus; 3. Memory; 4. Communication interface. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0029] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0030] like Figure 1 As shown, S101: collecting radar images.

[0031] In one embodiment, a synthetic aperture radar (GB-SAR) system is selected by personnel with relevant experience according to the monitoring target (such as terrain deformation, geological structure, etc.) and the monitoring environment (such as climate, vegetation coverage, etc.). The GB-SAR system can actively transmit microwaves and receive echoes to form high-resolution radar images.

[0032] In one embodiment, the raw data obtained by the aperture radar (GB-SAR) can be preprocessed, including radiation calibration, geometric correction, etc., to eliminate instrument errors and atmospheric interference and improve image quality. The preprocessing is an existing technology and will not be described in detail in this solution. It can be adjusted by the implementer according to the specific implementation scenario.

[0033] In one embodiment, two or more radar images collected in the same area at different times are registered and resampled to have the same spatial resolution and geometric position. Image registration and resampling are prior art and can be adjusted by implementers according to specific implementation scenarios. Then, an interferogram is generated by conjugate multiplication. The phase difference in the interferogram reflects the changes in the surface between the two observations.

[0034] S102: performing phase unwrapping on the radar imaging to obtain PS points.

[0035] In one embodiment, since the phase value in the interference graph is wrapped in the range of [-π,π), there will be a phase ambiguity problem. Phase unwrapping can recover the phase value from the wrapped phase through a series of algorithms (such as the minimum cost flow algorithm, branch cutting method, regional growth method, etc.) to obtain a continuous phase value. In the interference graph after phase unwrapping, those points that show strong and stable characteristics in backscattering (i.e., permanent scatterer points, PS points) are identified, such as building corners, bridges, exposed rocks, etc. These points usually have high coherence and can maintain stable phase information in multiple observations. Then, the phase change information of the PS point in different interference graphs can be used, combined with known DEM (digital elevation model) data and other auxiliary information (such as atmospheric correction, orbit error, etc.), to perform differential interference processing to remove the phase changes caused by terrain and other non-deformation factors. Finally, the deformation rate and displacement of the PS point can be calculated to obtain accurate measurement results of the surface deformation.

[0036] It should be noted that when the terrain is complex, there will still be some outlier PS points, which will lead to greater interference in the selection of stable PS points. Therefore, it is necessary to eliminate outlier PS points to obtain a stable PS point distribution.

[0037] In one embodiment, when selecting stable PS points, some isolated PS points can be eliminated through morphological opening operation, that is, corrosion first and then expansion. However, if the radius of the morphological core is too large when using the morphological opening operation to eliminate isolated PS points, some relatively dense PS points will be eliminated. If the core radius is too small, the elimination effect of some PS points will be poor. Therefore, when using the morphological opening operation to eliminate PS points, the selection of the radius length of the morphological opening operation is more important. Due to the different complexities of different terrains, the same morphological core radius has different effects on the elimination of PS points in different terrains. Therefore, this scheme chooses to improve the morphological opening operation so that it can adapt to the morphological core radius to ensure that PS points are eliminated under different complex terrains.

[0038] S103: using an improved morphological opening operation to eliminate the PS points to obtain a stable PS point distribution.

[0039] In one embodiment, when selecting stable PS points, the LOF outlier detection algorithm is used to obtain the outlier value of each PS point. It should be noted that the LOF outlier detection algorithm is used because LOF is a density-based outlier detection method. The outlier value of each point can be obtained through the LOF algorithm, which represents its distribution density. If the density is lower, the outlier value is larger, and the corresponding PS point is more likely to be an isolated point. However, when there are locally densely distributed points, there will also be relatively large outlier values.

[0040] In one embodiment, after obtaining each PS outlier, the Euclidean distance value A between each PS point and its nearest PS point is obtained. In other embodiments, multiple adjacent PS points may be selected to obtain the average of the Euclidean distance values. However, the number of adjacent PS points may not be accurately defined, resulting in the selected PS point being too far away. Then, the PS point with the minimum Euclidean distance value to the PS point is selected. Then, a circular window is established with the PS points, where the radius of the window is X, and the calculation of the first The Euclidean distance value between the PS point and the PS point in its window is used to adjust the outlier value. If there are no other PS points in the window of a PS point, the PS point can be assumed to be an isolated point and can be directly eliminated without further morphological considerations. Then the corrected outlier value is obtained.

[0041] S104: Correct the outlier value according to the distance value between the PS point and its nearest PS point to obtain a corrected outlier value of the PS point.

[0042] In one embodiment, an outlier detection algorithm is used to obtain an outlier value of a PS point, and the outlier value is corrected by the distance value between the PS point and its nearest PS point to obtain a corrected outlier value of the PS point, and the corrected outlier value is negatively correlated with the distance value, wherein the distance value between the PS point and its nearest PS point can be obtained by a Euclidean distance calculation method.

[0043] In one embodiment, the formula for outlier correction is: , where is the corrected outlier value of the i-th PS point, is the distance between the ith PS point and its nearest PS point, is the outlier value of the i-th PS point, and exp() plays the role of normalization and negative correlation.

[0044] in, The larger the value is, the farther the nearest PS point is in the vicinity of the i-th PS point, and the lower the possibility that it is a stable PS point. The smaller the value, the The closer the nearest PS point is within the vicinity of a PS point, the higher the possibility that it is a stable PS point. The distance value is then used to further correct the LOF value. The larger the value of, the larger the outlier value. The more PS points are at the edge of the distribution of all PSs, the lower the possibility that they are stable PS points. However, if there are multiple PS points locally, they may be local stable PS points far away from the overall distribution. Negative correlation with Multiply them and correct the LOF value to get the corrected outlier value. That is, the larger the corrected outlier value is, the more likely it is to be a stable PS point, and thus it should be retained after the morphological operation.

[0045] S105: Calculate the core radius of each PS point when performing the opening operation.

[0046] In one embodiment, after obtaining the corrected outlier value corresponding to the i-th PS point, the corrected outlier values ​​corresponding to each PS point are classified. Since the distance interval is related to the density, this scheme chooses to use the density mean algorithm to obtain the horizontal coordinates and corrected outliers corresponding to each PS point, and then uses the mean shift or DBSCAN and other density clustering algorithms to classify the coordinates and corrected outliers corresponding to each PS point, and classify those with similar coordinates and corrected outliers into one category, thereby obtaining multiple categories.

[0047] In one embodiment, after clustering the corrected outliers of each PS point to obtain multiple categories, the core radius of the PS point in the jth category when performing the opening operation is calculated. , the calculation formula is: ; In the formula, is the average distance value of all PS points in the jth category, is the scaling factor for the jth category.

[0048] In the formula, is the average distance value of all PS points in the jth category. The reason for using this distance value is that this point represents the minimum distance value to the remaining PS points. In the morphological opening operation, it is processed according to the size of the kernel, that is, the neighborhood range data is processed. Therefore, when When it is directly used as the kernel size of the opening operation, some PS points in the j-th category will inevitably be eliminated and some PS points will be retained, but this may cause the stable PS points to be mistakenly deleted and the outlier PS points to be eliminated poorly. Because when most of the PS points in the category are stable PS points, the kernel of the opening operation should be as small as possible to retain more PS points. When most of the PS points in the category are not stable PS points, the kernel of the opening operation should be as large as possible to eliminate more PS points.

[0049] In one embodiment, calculating the scaling factor of the jth category includes: obtaining the minimum value of the corrected outliers corresponding to all PS points in the jth category , taking the minimum value as the scaling factor of the jth category, where the scaling factor is related to the core radius Among them, the minimum value of the corrected outliers corresponding to all PS points in the jth category is used, which means that the smallest corrected outlier value of the category is selected as the standard, because the minimum value indicates that the category belongs to the lower limit of unstable PS points, and the smaller the minimum value, the smaller the possibility of belonging to stable PS points, and the more it needs to be eliminated. The smaller the minimum value of the corrected outliers corresponding to all PS points in the jth category, the smaller the core radius of the PS points in the jth category in the open operation. The larger the value of .

[0050] In another embodiment, after obtaining the minimum value of the corrected outliers corresponding to all PS points in the jth category, if the negative correlation of the minimum value is directly Multiply the values ​​of However, since the minimum value is an adjustment to the corrected outlier, its essence is still to correct the outlier, and the corrected outlier may be negative or positive, and may be large and uncertain, resulting in The adjustment is not stable, and the corresponding Normalize and then calculate the scaling factor .

[0051] It should be noted that in this embodiment, the scaling factor is calculated , it is necessary to normalize the corrected outliers of all PS points, wherein normalization can make all corrected outliers between 0-1, so that the value range is unified when analyzing the corrected outliers, which is conducive to the calculation of the adaptive core radius. Among them, the normalization method chooses to use the maximum and minimum value normalization method in this scheme, and other normalization methods cannot be used, because the maximum and minimum value normalization method can make the corrected outliers between 0-1. However, after normalization, no matter how large the value is, there will inevitably be 0 and 1 values, resulting in distortion, and then obtaining The normalized value is recorded as . After normalization The value data is statistically analyzed by histogram, where the histogram statistical result reflects the function distribution of the number of occurrences of the data.

[0052] The greater the difference between the upper and lower parts of the overall function distribution, the more it is considered that the PS points in the jth category are locally dense, otherwise it is considered to be sparse, and then its density distribution cutoff point is obtained, where the density cutoff point represents the scaling and stretching critical point. If it is less than the density cutoff point, it is more likely to be a stable point and does not need to be eliminated, and its radius should be small. If it is greater than the density cutoff point, it is more likely to be an unstable point and needs to be eliminated, and its radius should be large. At this point, the scaling coefficient for calculating the jth category can be established. The calculation formula for .

[0053] In one embodiment, a scaling factor is calculated, where the scaling factor is equal to the product of the local density of PS points in the jth category and the stretching factor, where the stretching factor is the ratio of the minimum value of all corrected outliers in the jth category in the histogram to the corrected outlier corresponding to the upper and lower difference critical points; the upper and lower difference critical points are the distance values ​​on the horizontal axis between the corrected outliers corresponding to the inflection points of the fitting function in the histogram and their minimum slope values, and the inflection point corresponding to the minimum distance value is selected as the upper and lower difference critical points.

[0054] In one embodiment, the larger the upper and lower difference coefficients are, the larger the upper and lower differences in the overall function distribution are, and it is considered that there is local density of PS points within the jth category, and the upper and lower difference critical points are valid, and then the scaling is performed according to the corresponding differences of the upper and lower difference critical points. If the upper and lower differences are smaller, it means that they are relatively sparse, and they need to be eliminated. At this time, the upper and lower difference critical points are invalid and are in an inconspicuous state.

[0055] In one embodiment, the inflection point corresponding to the minimum distance value is selected as the upper and lower difference critical points, including: performing polynomial fitting on the function distribution obtained by histogram statistics using the least squares method, performing inflection point calculation on the fitted function, and obtaining the inflection point of the fitted function; calculating the slope value of the normalized corrected outlier value on the fitting function, and obtaining the corrected outlier value corresponding to the minimum slope value on the horizontal axis, calculating the distance value between the corrected outlier value corresponding to each inflection point and the corrected outlier value corresponding to the minimum slope value on the horizontal axis, and selecting the inflection point with the minimum distance value as the upper and lower difference critical points.

[0056] In one embodiment, obtaining the upper peak coefficient includes: for any statistical result, obtaining the difference between the statistical result and the maximum value, calculating the ratio of the absolute value of the difference to the maximum value, and subtracting the ratio from 1 to obtain the weight of the statistical result; calculating the ratio of the value after weighted sum of all statistical results and corresponding weights to the number of all statistical results to obtain the upper peak coefficient. Obtaining the lower peak coefficient includes: for any statistical result, obtaining the difference between the statistical result and the minimum value, calculating the ratio of the absolute value of the difference to the minimum value, and subtracting the ratio from 1 to obtain the weight of the statistical result; calculating the ratio of the value after weighted sum of all statistical results and corresponding weights to the number of all statistical results to obtain the upper peak coefficient.

[0057] In one embodiment, the scaling factor for the jth category is calculated The calculation formula includes: ; In the formula, for The stretch factor relative to the upper and lower difference critical points, that is, the stretch factor is the ratio of the minimum value of all corrected outliers in the jth category in the histogram to the corrected outliers corresponding to the upper and lower difference critical points. is the upper and lower difference coefficient of the jth category; represents a step function, Represents a phase function.

[0058] Among them, when When the corresponding corrected outlier value in the histogram is less than the corrected outlier value corresponding to the upper and lower difference critical points, then A value less than 1 indicates shrinking. When the corresponding corrected outlier value in the histogram is greater than or equal to the corrected outlier value corresponding to the upper and lower difference critical points, then A value greater than or equal to 1 indicates stretching. The larger the value of is, the more it is considered that the PS points in the jth category are locally dense, and the core adaptive length is required, and then the upper and lower difference critical points are used to adjust the PS points. Stretch, where the stretching effect is based on On the contrary, if The smaller the value of is, the more it is considered that the PS points in the jth category are not locally dense and are relatively sparse, so we directly It can be used as the adaptive core length, and then the threshold of W is set to 0.3, where 0.3 can be adjusted by the implementer according to the specific implementation scenario.

[0059] Among them, if When the value of is less than or equal to 0, The value of is 0; if When the value of is greater than 0, The value of is 1. If the value of is less than or equal to 0, then The value of is 1; if Greater than 0, then The value is equal to .

[0060] S106: According to the stable PS point distribution, deformation monitoring is completed to achieve deformation inversion.

[0061] In one embodiment, after obtaining the core radius of the opening operation corresponding to the above classification result, the determined core radius is used to perform an opening operation on the PS points in the corresponding category to screen the PS points and obtain stable PS points, where the core is a square matrix of all ones. Then, deformation monitoring is completed based on the stable PS point distribution to achieve deformation inversion.

[0062] Through the above steps, by analyzing the data distribution of PS points, the core radius of morphological processing corresponding to different PS points is obtained, adaptive morphological processing is realized, isolated PS points are eliminated, more stable PS points are obtained, and the accuracy of deformation inversion is improved.

[0063] The present invention also provides a ground-based synthetic aperture slope radar atmospheric disturbance correction system. Figure 2 As shown, the system includes a processor 1 and a memory 3, wherein the memory 3 stores computer program instructions. When the computer program instructions are executed by the processor 1, a method for correcting atmospheric disturbances of a ground-based synthetic aperture slope radar according to the first aspect of the present invention is implemented.

[0064] The system also includes other components well known to those skilled in the art, such as a communication bus 2 and a communication interface 4. The configuration and functions of these components are known in the art and thus will not be described in detail herein.

[0065] In the present invention, the aforementioned memory 3 may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in the present invention may be implemented using computer-readable / executable instructions that may be stored or otherwise maintained by such a computer-readable medium.

[0066] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0067] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0068] The above-mentioned embodiments only express several implementation methods of the present invention, and the description is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent application. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A method for correcting atmospheric disturbances of ground-based synthetic aperture slope radar, characterized in that: include: Acquire radar images, perform phase unwrapping on the radar images, and obtain PS points; The PS points are eliminated by using an improved morphological opening operation to obtain a stable PS point distribution, and deformation monitoring is completed based on the stable PS point distribution to achieve deformation inversion; The improved morphological opening operation includes: obtaining an outlier value of a PS point by using an outlier detection algorithm, correcting the outlier value by using a distance value between the PS point and its nearest PS point to obtain a corrected outlier value of the PS point, wherein the corrected outlier value is negatively correlated with the distance value; Cluster the modified outliers corresponding to each PS point to obtain multiple categories, and calculate the core radius of the PS point in the jth category when performing the opening operation , the calculation formula is: ; In the formula, is the average distance value of all PS points in the jth category, is the scaling factor of the jth category; wherein the scaling factor is equal to the product of the local density of PS points in the jth category and the stretching factor, and the stretching factor is the ratio of the minimum value of all corrected outliers in the jth category in the histogram to the corrected outlier corresponding to the upper and lower difference critical points; the upper and lower difference critical points are the distance values ​​on the horizontal axis between the corrected outliers corresponding to the inflection points of the fitting function in the histogram and their minimum slope values, and the inflection point corresponding to the minimum distance value is selected as the upper and lower difference critical points.

2. The atmospheric disturbance correction method for ground-based synthetic aperture slope radar according to claim 1, characterized in that: The method for obtaining the local density of PS points in the jth category includes: Calculate the absolute value of the difference between the upper peak coefficient and the lower peak coefficient in the histogram corresponding to the jth category as the upper and lower difference coefficient; The upper and lower difference coefficients of the jth category are used to characterize the local density of PS points in the jth category.

3. The atmospheric disturbance correction method for ground-based synthetic aperture slope radar according to claim 1, characterized in that: Calculate the scaling factor for the jth category The calculation formula is: include: ； In the formula, for The stretch coefficient relative to the upper and lower difference critical points, The normalized value of the minimum value of the outlier correction for all PS points in the jth category, is the upper and lower difference coefficient of the jth category, represents a step function, Represents a phase function.

4. The atmospheric disturbance correction method for ground-based synthetic aperture slope radar according to claim 1, characterized in that: Calculate the corrected outlier value of the PS point using the following formula: ； In the formula, is the corrected outlier value of the i-th PS point, is the distance between the ith PS point and its nearest PS point, is the outlier value of the i-th PS point, and exp() is the normalization function.

5. The method for correcting atmospheric disturbances of ground-based synthetic aperture slope radar according to claim 4, characterized in that: Calculate the distance between a PS point and its nearest PS point, including: The distance between a PS point and its nearest PS point is calculated using the Euclidean distance calculation method.

6. The atmospheric disturbance correction method for ground-based synthetic aperture slope radar according to claim 2, characterized in that: The method for determining the peak coefficient includes: Obtaining the upper peak coefficient includes: for any statistical result, obtaining the difference between the statistical result and the maximum value, calculating the ratio of the absolute value of the difference to the maximum value, and subtracting the ratio from 1 to obtain the weight of the statistical result; The peak coefficient is obtained by calculating the ratio of the weighted sum of all statistical results and their corresponding weights to the number of all statistical results.

7. The method for correcting atmospheric disturbances of ground-based synthetic aperture slope radar according to claim 2, characterized in that: The method for determining the lower peak coefficient includes: Obtaining a lower peak coefficient includes: for any statistical result, obtaining a difference between the statistical result and a minimum value, calculating a ratio of an absolute value of the difference to the minimum value, and subtracting the ratio from 1 to obtain a weight of the statistical result; The ratio of the weighted sum of all statistical results and corresponding weights to the number of all statistical results is calculated to obtain the lower peak coefficient.

8. The atmospheric disturbance correction method for ground-based synthetic aperture slope radar according to claim 1, characterized in that: The inflection point corresponding to the minimum distance value is selected as the upper and lower difference critical points, including: The function distribution obtained by histogram statistics is fitted with a polynomial using the least square method, and the inflection point of the fitted function is calculated to obtain the inflection point of the fitted function; Calculate the slope value of the normalized corrected outlier on the fitting function, and obtain the corrected outlier corresponding to the minimum slope value on the horizontal axis, calculate the distance value between the corrected outlier corresponding to each inflection point and the corrected outlier corresponding to the minimum slope value on the horizontal axis, and select the inflection point with the minimum distance value as the upper and lower difference critical points.

9. The atmospheric disturbance correction method for ground-based synthetic aperture slope radar according to claim 1, characterized in that: Calculate the scaling factor for the jth category, also including: Get the minimum value of the corrected outliers corresponding to all PS points in the jth category, and use the minimum value as the scaling factor of the jth category, where the scaling factor is proportional to the core radius. Negatively correlated.

10. A ground-based synthetic aperture slope radar atmospheric disturbance correction system, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program, and the processor executes the computer program to implement the atmospheric disturbance correction method for a ground-based synthetic aperture slope radar as described in any one of claims 1 to 9.

Images