Large-area PSInSAR processing method, device and equipment based on partitioning strategy

Abstract

The invention discloses a large-area PSInSAR processing method, device and equipment based on a blocking strategy, and the method comprises the steps: carrying out the automatic data block division through employing a regular grid division method with an overlapping degree according to the number of rows and columns of original data and sub-block parameters, and obtaining a plurality of sub-blocks; block PS point selection, block networking and block parameter calculation processing are carried out on each sub-block; and adjusting parameters among the sub-blocks by utilizing overlapping degree information and a communication relation among the sub-blocks so as to realize smooth results and accurate splicing. According to the method, data block division is carried out firstly, and then block processing is carried out, so that memory occupation in links such as point selection, networking and parameter calculation during PSInSAR processing is improved, and parallel processing efficiency is improved; under the current hardware condition and under the condition that the calculation efficiency is not reduced, the number of PS points and the number of networking redundant arc segments can be increased, and the precision, the density and the coverage area of the result can be improved; sub-blocks can be spliced smoothly and accurately by considering information of homonymy points of sub-block overlapping degrees, and propagation of global errors is inhibited.

Description

Technical Field

[0001] The present invention relates to the technical field of time-series InSAR deformation monitoring, and in particular to a large-area PSInSAR processing method, device and equipment based on a block strategy. Background Art

[0002] Time-series InSAR technology is a technology developed on the basis of InSAR that uses multi-scene SAR images to achieve high-precision and high-density surface deformation monitoring. It suppresses the influence of temporal decoherence and atmospheric delay by extracting time-series coherent targets. Persistent scatterer InSAR technology (PSInSAR) is one of the most classic methods. In a scattering unit, there is a dominant ground scatterer, such as artificial structures, exposed rocks, etc., which can maintain phase stability for a long time and is called a permanent scatterer. Therefore, PSInSAR is suitable for areas with many artificial objects, especially urban ground subsidence and infrastructure deformation monitoring. With the launch of new high-resolution SAR satellites (such as TerraSAR-X, Fucheng No. 1, Lutan No. 1, etc.), we can obtain SAR data with higher spatial resolution, shorter revisit period, and wider coverage area. How to use PSInSAR technology to process these massive SAR data areas for rapid, accurate and detailed surface deformation monitoring is an urgent problem to be solved.

[0003] For large-scale PSInSAR deformation monitoring, it is necessary to consider the influence of hardware configuration, calculation speed, result density and result accuracy. The hierarchical method or the block method is a common method for large-scale PSInSAR processing. The hierarchical method is to construct a two-layer network, using the idea of ​​first control and then rag. The first layer of the network is processed with a stricter point selection threshold, redundant network construction method and parameter solution threshold, and then the second layer of the network is used to encrypt the observation points. The disadvantage is that it is necessary to comprehensively consider the global connectivity of the first layer of the network and the result accuracy of the parameter solution. If the first layer of the network cannot be globally connected or the calculation result accuracy is low, it will affect the coverage, density and calculation accuracy of the final deformation monitoring points. The block method is to perform PSInSAR processing and splicing on the blocks first. Considering the complex distribution characteristics of the actual objects, it is also necessary to perform complex block division according to the distribution characteristics of the PS points, which has a great impact on the automatic processing and the accurate and effective splicing of the results. Moreover, these methods are all divided into blocks after point selection, without considering the memory usage and processing efficiency of PS point selection.

[0004] At present, the layering method or the blocking method is a common method for large-area PSInSAR processing. However, these methods lack comprehensive consideration of memory usage, PSInSAR processing efficiency, and the accuracy and density of the results, which will also affect the overall results of PSInSAR processing. Summary of the invention

[0005] The purpose of the present invention is to overcome the deficiencies of the above-mentioned prior art and to provide a large-area PSInSAR processing method, device and equipment based on a blocking strategy, so as to provide an effective solution to the problems of excessive memory usage, low processing efficiency, low density and coverage of results, and suppression of result errors during large-area PSInSAR processing.

[0006] To achieve the above object, the technical solution of the present invention is:

[0007] In a first aspect, the present invention provides a large-area PSInSAR processing method based on a block strategy, comprising:

[0008] According to the number of rows and columns of the original data and the sub-block parameters, the data blocks are automatically divided using a regular grid division method with overlapping degrees to obtain several sub-blocks;

[0009] Perform block PS point selection, block network construction and block parameter solution processing on each sub-block;

[0010] The overlap information and connectivity between sub-blocks are used to adjust the parameters between sub-blocks to achieve smooth results and accurate splicing.

[0011] Optionally, the regular grid division method with overlapping degree is:

[0012] The size and overlap of the sub-blocks are set according to the configuration of the computer used, and the step length of each sub-block is calculated; the position and boundary range of each block are determined by moving in the matrix of the original data size by the step length, and the size of the sub-block is adjusted to adapt to the boundary of the matrix at the last row and the last column, ensuring that all sub-blocks are divided according to the specified size and overlap requirements, and the last sub-block can adapt to the actual size of the matrix.

[0013] Optionally, asynchronous parallel computing, multi-threaded processing or distributed system is used to perform block PS point selection, block network construction and block parameter solution processing on each sub-block.

[0014] Optionally, the block PS point selection includes:

[0015] The intensity value sequence of each pixel in the sub-block is read in sequence, relative radiation correction is performed and the amplitude deviation value is calculated, the coherence value sequence of each pixel is read in sequence to calculate the average coherence, and the PS point of the sub-block is selected using the amplitude deviation threshold and the average coherence threshold;

[0016] Take the union of the PS points in the overlapping area, retain all the PS points in the overlapping area, and update the PS points of each sub-block.

[0017] Optionally, the block networking includes:

[0018] Determine the number of PS points in the sub-block. If it is less than the specified number, skip the sub-block. Otherwise, read the position information, phase sequence, time and space baseline, incident angle and slant distance information of the PS point set in the sub-block, build a local Delaunay triangulation, and use the distance threshold to delete arcs with too large distances; solve the arc parameters and retain arcs greater than the overall coherence threshold; eliminate isolated arcs and points to generate a maximum connected network.

[0019] Optionally, the overall coherence γ is calculated using the formula:

[0020]

[0021] Among them, ΔΦ i The phase difference between the two end points of the arc segment is is the vertical baseline, T i time baseline, λ wavelength, R slant distance, θ incident angle, N is the total number of images, Δε and Δv are the elevation residual increment and deformation rate increment of the two end points of the arc segment.

[0022] Optionally, the block parameter solving includes:

[0023] In the sub-block, according to the arc parameter solution results and the maximum connected network, the point with the smallest amplitude deviation in the sub-block is selected as the reference point, and the elevation residual and deformation rate parameters of the PS point and the nonlinear deformation decomposition processing are calculated using the weighted least squares adjustment method.

[0024] Optionally, the adjusting of parameters between sub-blocks by using the overlap information and connectivity relationship between sub-blocks includes:

[0025] According to the block strategy, the points with the same name in the overlapping areas of two sub-blocks are searched in turn. If the number of points with the same name is greater than the specified threshold, the difference between the deformation rate and the elevation residual of each point with the same name is calculated;

[0026] The outliers are eliminated by using triple mean error, and the average value of the difference between the deformation rate and elevation residual of the remaining points with the same name after eliminating the outliers is calculated as the adjustment value between sub-blocks;

[0027] According to the calculated adjustment values ​​between sub-blocks and the corresponding sub-block relationships, the coefficient matrix B is established. m×n and the observation matrix L m×1 , m represents the number of adjustment values, n represents the total number of sub-blocks, where the coefficient matrix B m×n There are only three elements 1, -1 and 0, 1 and -1 represent the positions of two sub-blocks of the overlapping block with adjustment values; L m×1 Indicates the adjustment value of the elevation residual or deformation rate of the corresponding overlapping block;

[0028] Find and delete the columns in the coefficient matrix where all the elements are 0, and get the updated coefficient matrix B m×n

[0029] For the updated coefficient matrix B m×n Use the connected component analysis method to find the connectivity relationship of overlapping sub-blocks;

[0030] According to the connectivity relationship, from the coefficient matrix B m×n and the observation matrix L m×1 Extract the coefficient matrix B for each connected component subset mi×ni and the observation matrix L mi×1 , where mi and ni represent the total number of overlapping sub-blocks and the total number of sub-blocks in the i-th connected component;

[0031] The least squares adjustment is performed in each connected component subset, in which the block with the largest number of overlapping sub-blocks is selected as the reference block with zero adjustment value, and other sub-blocks are relatively adjusted.

[0032] In a second aspect, the present invention provides a large-area PSInSAR processing device based on a block strategy, comprising:

[0033] The data block division module is used to automatically divide the data blocks according to the number of rows and columns of the original data and the sub-block parameters by using a regular grid division method with an overlap degree to obtain a number of sub-blocks;

[0034] The block processing module is used to select block PS points, construct block networks and solve block parameters for each sub-block;

[0035] The block splicing module is used to adjust the parameters between sub-blocks by using the overlap information and connectivity relationship between sub-blocks to achieve smooth and accurate splicing results.

[0036] In a third aspect, the present invention provides an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0037] Memory, used to store computer programs;

[0038] The processor is used to implement the steps of the large-area PSInSAR processing method based on the block strategy described in any one of claims 1 to 8 when executing the program stored in the memory.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The large-area PSInSAR processing method based on the block strategy provided in this embodiment first divides the data into blocks and then processes the blocks, which improves the memory usage of the points selection, network construction, parameter solution and other links during PSInSAR processing and improves the parallel processing efficiency; it can increase the number of PS points and the number of redundant arcs in the network construction under the current hardware conditions and without reducing the calculation efficiency, which is conducive to improving the accuracy, density and coverage area of ​​the results; considering the information of the same-name points of the sub-block overlap, the sub-blocks can be smoothly and accurately spliced, which inhibits the propagation of global errors. The entire method process has a clear structure and is fully automated. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is an implementation flow chart of the large-area PSInSAR processing method based on the block strategy provided in the embodiment of the present application.

[0042] Figure 2 It is a graph of the data block partitioning results of the instance data.

[0043] Figure 3 It is the block point selection result diagram of the instance data.

[0044] Figure 4 It is the result graph of deformation rate and elevation residual of block processing of instance data.

[0045] Figure 5 It is a result diagram of the sub-block connectivity and the correction value of each sub-block's deformation rate and elevation residual.

[0046] Figure 6 It is a comparison chart of deformation rate results of instance data splicing and traditional methods.

[0047] Figure 7 It is a comparison chart of the elevation residual results of instance data splicing and traditional methods.

[0048] Figure 8 It is a scatter plot of the deformation rate of instance data splicing and traditional methods.

[0049] Fig. 9 It is a scatter error map of the elevation residuals of the example data splicing and the traditional method.

[0050] Fig.10 It is a schematic diagram of the composition of a large-area PSInSAR processing device based on a block strategy provided in an embodiment of the present application.

[0051] Fig.11 A schematic diagram of the composition of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0052] Example:

[0053] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0054] See also Figure 1 As shown, this embodiment provides a large-area PSInSAR processing method based on a block strategy, which mainly includes the following steps:

[0055] 110. According to the number of rows and columns of the original data and the sub-block parameters, a regular grid division method with overlapping degree is used to automatically divide the data blocks to obtain a number of sub-blocks; this step is the data block division step.

[0056] 120. Perform block PS point selection, block meshing and block parameter calculation on each sub-block; this step is the block processing step.

[0057] In this step, in order to further improve the efficiency, asynchronous parallel computing, multi-threaded processing or distributed system can be used to perform block PS point selection, block network construction and block parameter solution processing on each sub-block.

[0058] 130. Use the overlap information and connectivity between sub-blocks to adjust the parameters between sub-blocks to achieve smooth and accurate splicing results; this step is the block splicing step.

[0059] It can be seen that the large-area PSInSAR processing method based on the block strategy provided in this embodiment first divides the data into blocks and then processes them in blocks, which improves the memory usage of the points selection, network construction, parameter solution and other links during PSInSAR processing and improves the parallel processing efficiency; it can increase the number of PS points and the number of redundant arcs in the network under the current hardware conditions and without reducing the calculation efficiency, which is conducive to improving the accuracy, density and coverage area of ​​the results; considering the information of the same-name points of the sub-block overlap, the sub-blocks can be smoothly and accurately spliced, which suppresses the propagation of global errors. The whole method has a clear flow structure and is fully automated.

[0060] In a specific embodiment, the step 110 includes:

[0061] First, prepare the required files (i.e., raw data), obtain the SAR data and external DEM of the study area, use SAR processing software to generate a single-view complex image set, select one image as the main image, and perform registration and interference processing on other images to generate the files required by PSInSAR: spatiotemporal baseline file, differential interference atlas after removing the reference ellipsoid and DEM, coherence atlas, intensity atlas, slant range and incidence angle files;

[0062] Then, according to the configuration of the computer used and the row and column data of the original data, the overlap is set and regular grid division is performed. According to the actual configuration of the computer used, the size and overlap of the sub-blocks are set, and the step length of each block movement step=block_size-overlap is calculated, which represents the number of rows or columns of each block movement. By moving in the matrix of the original data size by the step length, the position and boundary range (rows and columns) of each block are determined, and the size of the block is adjusted to fit the boundary of the matrix at the last row and the last column, ensuring that all blocks are divided according to the specified size and overlap requirements, and the last block can fit the actual size of the matrix, and will not be omitted or exceed the range of the matrix due to the size of the block.

[0063] In a specific embodiment, the block PS point selection includes:

[0064] The intensity value sequence of each pixel in the sub-block is read in turn, relative radiation correction is performed and the amplitude deviation value is calculated, the coherence value sequence of each pixel is read in turn to calculate the average coherence, and the PS point of the sub-block is selected using the amplitude deviation threshold and the average coherence threshold.

[0065] The PS points in the overlapping areas of the sub-blocks are taken as the union to update the PS points of each sub-block. Since relative radiation correction is performed on the sub-blocks separately, the point selection results in the overlapping areas of different sub-blocks using a unified amplitude deviation point selection threshold are not exactly the same. After the block point selection is completed, the PS points in the overlapping areas are taken as the union to retain all the PS points in the overlapping areas and update the PS points of each sub-block.

[0066] In a specific embodiment, the block networking includes:

[0067] Determine the number of PS points in the sub-block. If it is less than the specified number, skip the sub-block. Otherwise, read the position information, phase sequence, time and space baseline, incident angle and slant distance information of the PS point set in the sub-block, build a local Delaunay triangulation, and use the distance threshold to delete arcs with too large distances; calculate the arc parameters and retain arcs greater than the overall coherence threshold; remove isolated arcs and points to generate the maximum connected network. The overall coherence γ calculation formula is:

[0068]

[0069] Among them, ΔΦ i The phase difference between the two end points of the arc segment is is the vertical baseline, T i Time baseline, λ wavelength, R slope distance, θ incident angle, N is the total number of images, Δε and Δv are the elevation residual increment and deformation rate increment of the two end points of the arc segment.

[0070] In a specific embodiment, the block parameter solution includes:

[0071] In the sub-block, according to the arc parameter solution results and the maximum connected network, the point with the smallest amplitude deviation threshold in the sub-block is selected as the reference point, and the weighted least squares adjustment method is used to calculate the elevation residual and deformation rate parameters of the PS point and the nonlinear deformation decomposition and other processing.

[0072] In a specific embodiment, the step 130 includes:

[0073] According to the initial block strategy, the points with the same name in the overlapping areas of two sub-blocks are searched in turn. If the number of points with the same name is greater than the specified threshold, the difference between the deformation rate and the elevation residual of each point with the same name is calculated:

[0074] Δvv k =v i -v j

[0075] Δhh k =h i -h j

[0076] Among them, i and j represent the index numbers of the two sub-blocks, Δvv k and Δhh k It represents the difference between the deformation rate and elevation residual of the k-th points of the same name in two sub-blocks.

[0077] The triple mean error is used to eliminate outliers. The following calculation formula takes the calculation of the deformation rate inter-block adjustment value as an example. The calculation method of the elevation residual inter-block adjustment value is the same:

[0078] Calculate the average

[0079]

[0080] Compute the difference between an observation and the mean:

[0081]

[0082] Get the mean error:

[0083]

[0084] Determine whether it exceeds the limit:

[0085]

[0086] The average value of the difference in deformation rate of the remaining points with the same name after removing the outliers is calculated and used as the adjustment value between sub-blocks.

[0087] According to the calculated adjustment values ​​between sub-blocks, the coefficient matrix B is established m×n and the observation matrix L m×1, m represents the number of adjustment values, n represents the total number of sub-blocks, where the coefficient matrix B m×n There are only three elements 1, -1 and 0, 1 and -1 represent the positions of two sub-blocks of the overlapping block with adjustment values. m×1 Indicates the adjustment value of the elevation residual or deformation rate of the corresponding overlapping block.

[0088] Find and delete the columns in the coefficient matrix whose elements are all 0. These column indices represent sub-blocks without PS points and independent sub-blocks without overlapping areas, and do not need to be spliced.

[0089] For the updated B m×n Connected Component Analysis (CCA) is used to find the connectivity of overlapping sub-blocks. Connected Component Analysis refers to a method used in graph theory to identify interconnected subsets in a graph. The non-zero values ​​of the coefficient matrix are converted into edges of the graph by constructing an adjacency matrix, so that the columns in the matrix are regarded as nodes, and the non-zero values ​​in the rows represent the connectivity between the nodes.

[0090] According to the connectivity relationship, from the matrix B m×n and the observation matrix L m×1 Extract the coefficient matrix B for each connected component subset mi×ni and the observation matrix L mi×1 , where mi and ni represent the total number of overlapping sub-blocks and the total number of sub-blocks in the i-th connected component; the least squares adjustment is performed in each connected component subset, and the block with the largest number of overlapping sub-blocks with other sub-blocks is selected as the reference block (the sub-block with zero deformation and elevation residual adjustment value), which is the coefficient matrix B mi×ni The column with the most non-zero elements in the , and other sub-blocks are adjusted relatively.

[0091] X=(B T PB) -1 B T PL

[0092] Among them, B is the coefficient matrix, L is the observation matrix, and P is the weight matrix. The unit matrix can be used as the weight, or the ratio of the number of points with the same name in the overlapping sub-blocks to the total number of points with the same name in all overlapping sub-blocks can be used as the weight.

[0093] The method of the present invention is further described below with reference to an application scenario example:

[0094] (1) Data block division

[0095] In this example, the SAR data is 23 scenes of TerraSAR-X strip mode (SM) data with a resolution of 3m, and the external DEM is CopDEM GLO-30. After clipping, registration and interferometry, the number of rows and columns of the data is 13500*11500. Figure 2 As shown, using the regular grid division method with overlap, the regular grid is divided into 3000*3000 sub-blocks, the overlap is 500, divided into 20 sub-blocks, the number of columns of the rightmost column sub-block is 4000 and the number of rows of the lowest row sub-block is 3500, adaptive to the boundary of the original data.

[0096] (2) Block processing

[0097] First, according to the above data block division method, the intensity value and coherence value of each pixel in the sub-block are read in blocks, the intensity value is corrected for relative radiation, the amplitude deviation value and the average coherence value are calculated, and the pixels with amplitude deviation threshold less than 0.35 and average coherence greater than 0.6 are selected as PS points. Finally, the union of the selected points of the sub-blocks is taken, as shown in Figure 3 As shown, the total number of PS points of the sub-blocks is 4.5 million, of which 1.1 million are overlapped PS points, and the number of PS points after merging is 3.4 million.

[0098] In each block, read the location information, deformation sequence, incident angle, slant distance and other information of each candidate point. Construct a local Delaunay triangulation network and delete arcs greater than the distance threshold. It is generally believed that the atmospheric effects of two points within a 1 km range are similar. The solution space search method is used to calculate the parameters such as the elevation residual increment and deformation rate increment of the two end points of the arc segment when the overall coherence of the arc segment is the largest. The overall coherence γ calculation formula is:

[0099]

[0100] Among them, ΔΦ i The phase difference between the two end points of the arc segment is is the vertical baseline, T i time baseline, λ wavelength, R slant distance, θ incident angle, N is the total number of images, Δε and Δv are the elevation residual increment and deformation rate increment of the two end points of the arc segment.

[0101] Delete arcs with overall coherence less than 0.75, calculate the connectivity of the remaining arcs, delete isolated arcs and sub-networks, and retain only the global maximum connected network. Select the point with the smallest amplitude deviation in the network as the point of zero deformation and zero elevation residual as the reference point, and use the weighted least squares adjustment method to calculate the deformation rate and elevation residual value of each PS point. The results can also be iteratively processed later to separate nonlinear deformation and two-layer network construction. Figure 4As shown in the figure, a) is the result of deformation rate after block solution, and b) is the result of elevation residual after block solution. There are obvious boundaries between different sub-blocks.

[0102] (3) Block stitching

[0103] In order to unify the results of different sub-blocks, splicing processing is required. The deformation rate and elevation residual information of the same-name points in the overlapping area between sub-blocks are used to make overall adjustments to the sub-blocks. First, find the same-name points in different sub-blocks. If the number of same-name points is less than the specified number, the two sub-blocks are considered to be disconnected. Otherwise, calculate the deformation rate difference and elevation residual difference of the same-name points in different sub-blocks, eliminate the same-name points that are greater than three times the mean error, and calculate the average value of the deformation rate difference and elevation residual difference of the remaining same-name points as the overall adjustment value of the deformation rate and the overall adjustment value of the elevation residual between the sub-blocks. Because different sub-blocks have multiple overlapping sub-blocks, the final adjustment amount of each sub-block is calculated using the least squares adjustment method and the redundant observation information of the overlapping sub-blocks.

[0104] Because the distribution of PS points in sub-blocks may be extremely uneven, there are no PS points in the sub-blocks or there are PS points in the sub-blocks but there are no points with the same name in the overlapping areas. Therefore, it is necessary to determine the connectivity of the sub-blocks with PS points and use the Connected Component Analysis (CCA) method to find the connectivity of overlapping sub-blocks. The connected component analysis method refers to a method used in graph theory to identify interconnected subsets in a graph. The non-zero values ​​of the coefficient matrix are converted into edges of the graph by constructing an adjacency matrix, so that the columns in the matrix are regarded as nodes, and the non-zero values ​​in the rows represent the connection relationship between the nodes. Perform the least squares adjustment in each connected component to obtain the adjustment number of each sub-block in each connected component. For example Figure 5 As shown in the figure, a) is the relationship of connected components. There are two connected components in total. There are 19 sub-blocks in ① and one sub-block in ②. There is no sub-block without PS points. b) is the overall adjustment of the elevation residual of each sub-block, and c) is the overall adjustment of the deformation rate of each sub-block.

[0105] Finally, the adjustment amount of each sub-block is used to adjust the result obtained by block processing. Figure 6 As shown in the figure, a) is the final deformation rate result obtained by this method. b) is the result of the traditional method processing the entire image. The results of the two are relatively consistent. Figure 7As shown in the figure, a) is the final elevation residual result obtained by this method. b) is the result of the traditional method processing the entire image. The results of the two are relatively consistent. There are 3.4 million PS points in a) and 700,000 PS points in b). According to the configuration of the workstation used, the traditional method has too many PS points, so sparse sampling of regular grids is performed after point selection before subsequent PSInSAR processing can be performed. The method in this article does not need to consider the problem of excessive memory usage. In addition, due to connectivity issues, the upper left and lower right corners of the image in b) are not covered by PS points in the end. Therefore, this also highlights the advantages of this method. Figure 8 As shown, the scatter plot error of deformation rate of instance data splicing and traditional methods. Fig. 9 As shown in Figure 1, the scatter plot of the elevation residuals of the example data splicing and the traditional method. The results of the two are quite consistent, with a correlation above 0.95.

[0106] In summary, compared with the prior art, the invention has the following technical advantages:

[0107] (1) A block partitioning strategy suitable for large-scale PSInSAR processing is proposed. It does not need to be influenced by other prior information such as the distribution characteristics of PS points. The sub-block size and overlap parameters are set according to the hardware configuration, especially the computer memory size. The operations of first partitioning, then reading, processing and finally splicing are performed.

[0108] (2) A PSInSAR processing method for sub-blocks based on the aforementioned block partitioning strategy is proposed, including methods for point selection, network construction and parameter solution. The processing of each sub-block can also be performed by asynchronous parallel computing, multi-threaded processing or distributed system to improve efficiency.

[0109] (3) A method for splicing sub-block PSInSAR processing results based on the aforementioned block segmentation strategy is proposed. The splicing is performed based on the information of the same-name points in the overlapping areas and the connectivity of each sub-block to ensure the consistency and integrity of the overall results.

[0110] Accordingly, if Fig.10 As shown, this embodiment also provides a large-area PSInSAR processing device based on a block strategy, comprising:

[0111] The data block division module is used to automatically divide the data blocks according to the number of rows and columns of the original data and the sub-block parameters by using a regular grid division method with an overlap degree to obtain a number of sub-blocks;

[0112] The block processing module is used to select block PS points, construct block networks and solve block parameters for each sub-block;

[0113] The block splicing module is used to adjust the parameters between sub-blocks by using the overlap information and connectivity relationship between sub-blocks to achieve smooth and accurate splicing results.

[0114] It should be noted that the large-area PSInSAR processing device based on the blocking strategy provided in the embodiment of the present application can execute the large-area PSInSAR processing method based on the blocking strategy provided in any embodiment of the present application, and has the corresponding functions and beneficial effects of the execution method.

[0115] like Fig.11 As shown, the embodiment of the present application also provides an electronic device, including a processor 111, a communication interface 112, a memory 113 and a communication bus 114, wherein the processor 111, the communication interface 112, and the memory 113 complete mutual communication through the communication bus 114; the memory 113 is used to store computer programs; the processor 111 is used to execute the program stored in the memory 113, and implement the steps of the large-area PSInSAR processing method based on the block strategy provided by any of the above method embodiments. Exemplarily, the steps of the large-area PSInSAR processing method based on the block strategy may include the following steps: according to the number of rows and columns of the original data and the sub-block parameters, the regular grid division method with overlap is used to automatically divide the data blocks to obtain a number of sub-blocks; block PS point selection, block network construction and block parameter solution processing are performed on each sub-block; and the overlap information and connectivity relationship between sub-blocks are used to adjust the parameters between sub-blocks to achieve smooth results and accurate splicing. .

[0116] The embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the large-area PSInSAR processing method based on the blocking strategy provided in any of the aforementioned method embodiments are implemented.

[0117] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0118] The above description is only a specific implementation of the present application, so that those skilled in the art can understand or implement the present application. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features of the present application.

[0119] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable ordinary technicians in the field to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the protection scope of the present invention. Any equivalent changes or modifications made based on the essence of the content of the present invention should be included in the protection scope of the present invention.

Claims

1. A large-area PSInSAR processing method based on a block strategy, characterized in that: include: According to the number of rows and columns of the original data and the sub-block parameters, the data blocks are automatically divided using a regular grid division method with overlapping degrees to obtain several sub-blocks; Perform block PS point selection, block network construction and block parameter solution processing on each sub-block; The overlap information and connectivity between sub-blocks are used to adjust the parameters between sub-blocks to achieve smooth results and accurate splicing.

2. The large-area PSInSAR processing method based on the block strategy as claimed in claim 1, characterized in that: The regular grid division method with overlapping degree is: The size and overlap of the sub-blocks are set according to the configuration of the computer used, and the step length of each sub-block is calculated; By moving in the matrix of the original data size by step size, the position and boundary range of each block are determined, and the size of the sub-block is adjusted to adapt to the boundary of the matrix at the last row and the last column, ensuring that all sub-blocks are divided according to the specified size and overlap requirements, and the last sub-block can adapt to the actual size of the matrix.

3. The large-area PSInSAR processing method based on the block strategy as claimed in claim 1, characterized in that: Asynchronous parallel computing, multi-threaded processing or distributed system is used to select PS points, construct blocks and solve block parameters for each sub-block.

4. The large-area PSInSAR processing method based on the block strategy as described in claim 1 or 3 is characterized in that: The block PS point selection includes: The intensity value sequence of each pixel in the sub-block is read in sequence, relative radiation correction is performed and the amplitude deviation value is calculated, the coherence value sequence of each pixel is read in sequence to calculate the average coherence, and the PS point of the sub-block is selected using the amplitude deviation threshold and the average coherence threshold; Take the union of the PS points in the overlapping area, retain all the PS points in the overlapping area, and update the PS points of each sub-block.

5. The large-area PSInSAR processing method based on the block strategy as claimed in claim 4, characterized in that: The block networking includes: Determine the number of PS points in the sub-block. If it is less than the specified number, skip the sub-block. Otherwise, read the position information, phase sequence, time and space baseline, incident angle and slant distance information of the PS point set in the sub-block, build a local Delaunay triangulation, and use the distance threshold to delete the arc segments with too large distance. Calculate the arc segment parameters and retain the arc segments that are greater than the overall coherence threshold; Eliminate isolated arcs and points to generate a maximum connected network.

6. The large-area PSInSAR processing method based on the block strategy as claimed in claim 5, characterized in that: The overall coherence γ is calculated by: Among them, ΔΦ i The phase difference between the two end points of the arc segment is is the vertical baseline, T i time baseline, λ wavelength, R slant distance, θ incident angle, N is the total number of images, Δε and Δv are the elevation residual increment and deformation rate increment of the two end points of the arc segment.

7. The large-area PSInSAR processing method based on the block strategy as claimed in claim 5, characterized in that: The block parameter solution includes: In the sub-block, according to the arc parameter solution results and the maximum connected network, the point with the smallest amplitude deviation in the sub-block is selected as the reference point, and the elevation residual and deformation rate parameters of the PS point and the nonlinear deformation decomposition processing are calculated using the weighted least squares adjustment method.

8. The large-area PSInSAR processing method based on the block strategy as claimed in claim 1, characterized in that: The adjusting of parameters between sub-blocks by using the overlap information and connectivity relationship between sub-blocks includes: According to the block strategy, the points with the same name in the overlapping areas of two sub-blocks are searched in turn. If the number of points with the same name is greater than the specified threshold, the difference between the deformation rate and the elevation residual of each point with the same name is calculated; The outliers are eliminated by using triple mean error, and the average value of the difference between the deformation rate and elevation residual of the remaining points with the same name after eliminating the outliers is calculated as the adjustment value between sub-blocks; According to the calculated adjustment values ​​between sub-blocks and the corresponding sub-block relationships, the coefficient matrix B is established. m×n and the observation matrix L m×1 , m represents the number of adjustment values, n represents the total number of sub-blocks, where the coefficient matrix B m×n There are only three elements 1, -1 and 0, 1 and -1 represent the positions of two sub-blocks of the overlapping block with adjustment values; L m×1 Indicates the adjustment value of the elevation residual or deformation rate of the corresponding overlapping block; Find and delete the columns in the coefficient matrix where all the elements are 0, and get the updated coefficient matrix B m×n For the updated coefficient matrix B m×n Use the connected component analysis method to find the connectivity relationship of overlapping sub-blocks; According to the connectivity relationship, from the coefficient matrix B m×n and the observation matrix L m×1 Extract the coefficient matrix B for each connected component subset mi×ni and the observation matrix L mi×1 , where mi and ni represent the total number of overlapping sub-blocks and the total number of sub-blocks in the i-th connected component; The least squares adjustment is performed in each connected component subset, in which the block with the largest number of overlapping sub-blocks is selected as the reference block with zero adjustment value, and other sub-blocks are relatively adjusted.

9. A large-area PSInSAR processing device based on a block strategy, characterized in that: include: The data block division module is used to automatically divide the data blocks according to the number of rows and columns of the original data and the sub-block parameters by using a regular grid division method with an overlap degree to obtain a number of sub-blocks; The block processing module is used to select block PS points, construct block networks and solve block parameters for each sub-block; The block splicing module is used to adjust the parameters between sub-blocks by using the overlap information and connectivity relationship between sub-blocks to achieve smooth and accurate splicing results.

10. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor is used to implement the steps of the large-area PSInSAR processing method based on the blocking strategy as described in any one of claims 1 to 8 when executing the program stored in the memory.

Images