InSAR co-seismic deformation down-sampling method and system based on image saliency

Abstract

The invention discloses an InSAR co-seismic deformation downsampling method and system based on image saliency, and belongs to the field of surface deformation monitoring, and the method comprises the steps: carrying out the average saliency extraction, binarization and expansion of an InSAR deformation image, and then generating a mask file; performing size expansion processing on the near-field region and the multi-scale average saliency map to generate a masked multi-scale average saliency map; performing quadtree sampling on the near-field region to generate a first sampling result, and performing uniform sampling on the far-field region based on a constant window to generate a second sampling result; and splicing the first sampling result and the second sampling result to generate a target sampling result. According to the method, details of a main deformation area can be reserved, interference of far-field or local deformation can be suppressed, reliable co-seismic deformation sampling data can be obtained, the reliability of researches such as subsequent focus parameter inversion and the like is improved, prior fault information and pre-inversion are not needed, the complexity of the sampling process is reduced, and the applicability of the method is enhanced.

Description

Technical Field

[0001] This application belongs to the field of surface deformation monitoring, and specifically relates to an InSAR co-seismic deformation downsampling method and system based on image saliency. Background Art

[0002] Synthetic Aperture Radar Interferometry (InSAR) is an advanced radar remote sensing technology for earth observation. One of the most important applications of this technology is to obtain the deformation that occurs on the earth's surface during the period between two observations through repeated observations of the same area. As a direct manifestation of crustal stress release, surface deformation is a key observation data for analyzing the evolution law of disasters. InSAR technology has been widely used in the field of surface deformation monitoring. Compared with point-like observation means such as the Global Navigation Satellite System (GNSS), InSAR technology can provide more comprehensive surface deformation information, thus supporting more detailed and in-depth research on the causes and impacts of disasters. Co-seismic deformation refers to the surface displacement generated at the moment of an earthquake. The co-seismic deformation field obtained by InSAR has the advantages of wide coverage and high spatial resolution. There are millions of observation points in a conventional InSAR image. Although more observation points are helpful for obtaining more deformation detail information, they will also bring huge computational costs to parameter inversion.

[0003] Before using the InSAR deformation field for fault parameter inversion, data downsampling will be performed. Traditional InSAR co-seismic deformation downsampling methods include uniform sampling, resolution-based downsampling, and quadtree sampling. However, the above sampling methods have the following deficiencies: First, uniform sampling is achieved by averaging the deformation points in a given-sized window. The method is relatively simple and crude. Although it can reduce the amount of data, it will also blur the details of the deformation center.

[0004] Second, the resolution-based downsampling method introduces the initial fault geometric parameters as prior information into the sampling process, and judges the near and far fields according to the distance between the deformation points and the fault, so as to reduce the interference of local deformation in the far field. However, due to the need for prior information, the applicability of this method is limited.

[0005] Thirdly, the quadtree sampling algorithm uses the deformation gradient to divide the window for sampling. Without using prior information, this algorithm can greatly reduce the data volume and effectively retain the main deformation features. Compared with the resolution-based downsampling method, quadtree sampling is more widely used in InSAR earthquake research. However, the coseismic deformation field of InSAR is usually affected by temporal decorrelation, atmospheric delay, etc., resulting in local deformation, so that high deformation gradient areas appear in some regions far from the deformation center. The quadtree sampling method using the deformation gradient as the window division index may sample many unnecessary points in these regions, thus reducing the downsampling efficiency and the accuracy of subsequent parameter inversion. Summary of the Invention

[0006] The purpose of the embodiments of the present application is to provide an InSAR coseismic deformation downsampling method and system based on image saliency, which can solve the problems of high complexity and low applicability of existing sampling techniques.

[0007] To solve the above technical problems, the present application is implemented as follows: In a first aspect, the embodiments of the present application provide an InSAR coseismic deformation downsampling method based on image saliency. The method includes: extracting the average saliency of a preset InSAR deformation map to obtain a multi-scale average saliency map; performing binarization and dilation processing on the multi-scale average saliency map in sequence to generate a mask file. The mask region with the largest area in the mask file is the near-field region, and other regions in the mask file except the near-field region are the far-field regions; respectively performing size expansion processing on the mask file and the multi-scale average saliency map of the near-field region to generate a masked multi-scale average saliency map; performing quadtree sampling on the near-field region based on the masked multi-scale average saliency map and the size-expanded InSAR deformation map to generate a first sampling result, and performing uniform sampling on the far-field region based on a constant window to generate a second sampling result; splicing the first sampling result and the second sampling result to generate a target sampling result.

[0008] As an optional implementation manner of the first aspect of the present application, the process of extracting the average saliency of a preset InSAR deformation map to obtain a multi-scale average saliency map includes: obtaining a two-dimensional matrix image of the InSAR coseismic deformation to be sampled; setting the sizes of the sliding window and the search window, where the size of the sliding window is a sequence and the size of the search window is a fixed constant; expanding the width of the normalized InSAR deformation map outward and filling it with null values; calculating the average value of the non-null points in the sliding window and the search window; calculating the absolute value of the difference between the average values, and assigning the absolute value to all pixel points in the sliding window as the saliency of each pixel point in the sliding window; moving the sliding window and the search window to traverse the image to generate the saliency of the image in the deformation field. Change the size of the sliding window and repeat the above steps to generate significance maps of different sizes of the deformation field; Calculate the average significance of the image based on the significance map of each image to generate a multi-scale average significance map.

[0009] As an alternative implementation of the first aspect of the present application, obtaining the first sampling result includes: performing size expansion processing on a preset InSAR deformation map to generate an expanded InSAR deformation map; performing quadtree sampling on the expanded InSAR deformation map, the multi-scale average significance map after masking, a set minimum window, a set maximum window, and a segmentation threshold, and after dividing the image into four equal parts, obtaining four sub-windows; calculating the standard deviation of each sub-window respectively, and performing iterative quadtree division on each sub-window that satisfies the standard deviation being greater than the set threshold until the standard deviation of the sub-window is less than the set threshold or the size of the sub-window is equal to the minimum window to stop the division, generating multiple segmented sub-windows; calculating the longitude, latitude, and average deformation of the sampling points of each segmented sub-window; generating the first sampling result based on the longitude, latitude, and average deformation.

[0010] As an alternative implementation of the first aspect of the present application, the first sampling result is a matrix with N rows and 3 columns, the first column is the longitude, the second column is the latitude, and the third column is the average deformation, where N represents the number of sampling points.

[0011] As an alternative implementation of the first aspect of the present application, the process of uniformly sampling the far-field region based on a constant window to generate the second sampling result includes: setting a constant window based on the maximum window set by quadtree sampling; dividing the preset InSAR deformation map based on the constant window to obtain multiple segmented far-field regions; calculating the average deformation of each window in each segmented far-field region; obtaining the second sampling result based on the average deformation.

[0012] In a second aspect, an embodiment of the present application provides an InSAR co-seismic deformation downsampling system based on image significance, and the system includes: A significance extraction module, configured to perform average significance extraction on a preset InSAR deformation map to obtain a multi-scale average significance map; A mask file generation module, configured to perform binarization and dilation processing on the multi-scale average significance map in sequence to generate a mask file, where the mask region with the largest area in the mask file is the near-field region, and the mask region with the smallest area in the mask file is the far-field region; A size expansion module, configured to perform size expansion processing on the mask file and the multi-scale average significance map of the near-field region respectively to generate a multi-scale average significance map after masking; A sampling module, configured to perform quadtree sampling on the near-field region based on the masked multi-scale average saliency map and the size-expanded InSAR deformation map to generate a first sampling result, and perform uniform sampling on the far-field region based on a constant window to generate a second sampling result; A target sampling result generation module, configured to splice the first sampling result and the second sampling result to generate a target sampling result.

[0013] In a third aspect, an embodiment of the present application provides an electronic device, which includes a processor, a memory, and a program or instruction stored on the memory and executable on the processor. When the program or instruction is executed by the processor, the steps of the method in the first aspect are implemented.

[0014] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by the processor, the steps of the method in the first aspect are implemented. Description of the Drawings

[0015] Figure 1 is a flowchart of a method for downsampling InSAR co-seismic deformation based on image saliency provided by the first embodiment of the present application; Figure 2 is an overall detailed flowchart of a method for downsampling InSAR co-seismic deformation based on image saliency provided by the first embodiment of the present application; Figure 3 is a schematic diagram of obtaining an image saliency map by a multi-scale filtering method provided by the first embodiment of the present application; Figure 4 is a satellite line-of-sight co-seismic deformation map and saliency maps at different scales provided by the first embodiment of the present application; Figure 5 is an internal structure diagram of a system for downsampling InSAR co-seismic deformation based on image saliency provided by the second embodiment of the present application. Detailed Embodiments

[0016] Next, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in the present application belong to the scope of protection of the present application.

[0017] The terms "first", "second", etc. in the description and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of this application can be implemented in an order other than those illustrated or described herein. In addition, "and / or" in the description and claims means at least one of the connected objects, and the character " / " generally indicates an "or" relationship between the associated objects before and after.

[0018] The following will, with reference to the accompanying drawings, through specific embodiments and their application scenarios, elaborate in detail on a method and system for InSAR co-seismic deformation downsampling based on image saliency provided by the embodiments of this application.

[0019] Embodiment 1 Please refer to Figure 1 , which shows a flowchart of a method for InSAR co-seismic deformation downsampling based on image saliency provided by the present invention, including steps S1 to S5.

[0020] Step S1: Extract the average saliency of a preset InSAR deformation map to obtain a multi-scale average saliency map.

[0021] Specifically, the process of extracting the average saliency of a preset InSAR (Interferometric Synthetic Aperture Radar) deformation map to obtain a multi-scale average saliency map includes: obtaining a two-dimensional matrix image of the InSAR co-seismic deformation to be sampled; setting the sizes of the sliding window and the search window, where the size of the sliding window is a sequence and the size of the search window is a fixed constant; expanding the width of the normalized InSAR deformation map and filling the expanded part with null values; calculating the average value of the non-null points of the sliding window and the search window and ; calculating the saliency of each pixel point in the sliding window based on the average value ; assigning the absolute value to all pixel points in the sliding window as their saliency values; moving the sliding window and the search window to traverse the image to generate the saliency of each image in the deformation field; calculating the average saliency of the image based on the saliency of each image to generate a multi-scale average saliency map.

[0022] The detailed process of obtaining the multi-scale average saliency map in the above steps is as follows: S11. Obtain the InSAR coseismic deformation image to be sampled.

[0023] Specifically, in step S11, the deformation image is two-dimensional matrix data, and the two-dimensional matrix includes multiple deformation images.

[0024] S12. Use the multi-scale filtering method to obtain the significance of each pixel in the deformation field.

[0025] Please refer to Figure 4 , which shows the coseismic deformation map in the satellite line-of-sight direction and the significance maps at different scales. Figure 4 It shows the significance and its average value obtained using different values. Increasing the size of the sliding window will reflect the significance at a larger scale, but the resolution of the significance map will gradually decrease. The size of the search window is positively correlated with the integrity of the significant region. Specifically, step S12 specifically includes: S121. Set the size of the sliding window to be , and the size of the search window to be , where represents the sequence , , , represents the sequence of length.

[0026] S122. Calculate the average values of the non-empty value points in the sliding window and the search window and , and then calculate the significance value. The mathematical expression of the significance value is: Assign the significance of all points in the sliding window to be .

[0027] S123. Move the two windows to traverse the entire image to obtain the significance value of each deformation image, where , move the sliding window and the search window until the entire deformation map is traversed to obtain the significance value .

[0028] S124. Change the size of the sliding window , and repeat steps S121 to S123 to obtain multiple significance values.

[0029] S125. Calculate the mean of multiple significances in calculation step S124, and the mathematical expression for the average significance of the image is: Wherein, represents the average significance of the image, represents the sequence length of, represents the significance value of each deformed image.

[0030] Step S2: After successively performing binarization and dilation processing on the multi-scale average significance map, generate a mask file. The mask region with the largest area in the mask file is the near-field region, and other regions in the mask file except the near-field region are the far-field regions.

[0031] Specifically, the process of successively performing binarization on the multi-scale average significance map includes: setting a threshold according to the average significance , generally a value that can better separate the main deformation region. The purpose of binarization is to establish a mask file; dilation processing is used to fill data gaps and make the main deformation region form a whole. The mask file established through binarization and dilation processing can distinguish the near-field, far-field, and local deformations, and the mask region with the largest area is the near-field region.

[0032] Step S3: Respectively perform size expansion processing on the mask file of the near-field region and the multi-scale average significance map to generate a multi-scale average significance map after masking.

[0033] Specifically, please refer to Figure 2 , on the one hand, perform size expansion on the multi-scale average significance map to obtain an expanded multi-scale average significance map, and its size is . On the other hand, perform binarization processing and dilation processing on the multi-scale average significance map in sequence, select the largest block (the mask region with the largest area) as the near-field region, and after performing size expansion on the whole image (the expanded size is also ), perform fusion processing with the expanded multi-scale average significance map to generate a multi-scale average significance map after masking. Figure 3 represents a schematic diagram of obtaining an image significance map by a multi-scale filtering method.

[0034] Step S4: Perform quadtree sampling on the near-field region based on the multi-scale average significance map after masking and the InSAR deformation map after size expansion to generate a first sampling result, and perform uniform sampling on the far-field region based on a constant window to generate a second sampling result.

[0035] Further, the specific process of step S4 for generating the first sampling result by performing quadtree sampling on the near-field region based on the masked multi-scale average saliency map includes: S41. Expand the multi-scale average saliency map and the original deformation map into a multi-scale average saliency map with both length and width of 2 n pixels based on the original multi-scale average saliency map for quadrisection.

[0036] S42. Use the standard deviation of the saliency in the initial window as the reference index for near-field quadtree sampling, where the mathematical expression of this reference index is: where represents the standard deviation, represents the number of points at the center of the window, represents the th point in the window, represents the mean of the saliencies of all non-empty points in the window.

[0037] S43. When the of the window is greater than the set threshold, divide the window into four sub-windows of equal size on average.

[0038] S44. Calculate the annotation deviation corresponding to each of the four divided sub-windows respectively , if in the sub-window is greater than the set threshold, perform iterative division in sequence, if in the sub-window is less than the set threshold or the window is smaller than the set minimum window size, stop the division.

[0039] S45. Use the divided panes to obtain the average deformation value in each pane and use it as the deformation value of the center point of the pane, and generate the first sampling result based on the deformation value.

[0040] Further, please refer to Figure 2, the process of generating the first sampling result is as follows: First, perform size expansion processing on the preset InSAR deformation map to generate an expanded InSAR deformation map; perform quadtree sampling based on the expanded InSAR deformation map, the masked multi-scale average saliency map, the set minimum window, maximum window, and segmentation threshold. After dividing the image into four equal parts, four sub-windows are obtained; calculate the standard deviation of each sub-window respectively, and perform iterative quadtree segmentation on each sub-window that satisfies the standard deviation being greater than the set threshold until the standard deviation of the sub-window is less than the set threshold or the sub-window size is equal to the minimum window, generating multiple segmented sub-windows; calculate the longitude, latitude, and average deformation of the sampling points in each segmented sub-window; based on the longitude, latitude, and average deformation, generate the first sampling result.

[0041] Further, the process of step S4 for uniformly sampling the far-field region based on a constant window to generate the second sampling result is as follows: Set a constant window based on the maximum window set by quadtree sampling; segment the preset InSAR deformation map based on the constant window to obtain multiple segmented far-field regions; calculate the average deformation of each window in each segmented far-field region; obtain the second sampling result based on the average deformation, where the second sampling result is a matrix of N rows and 3 columns. Set the longitude of the average deformation as the first column of the second sampling result, the latitude of the average deformation as the second column of the second sampling result, and set the average deformation as the third column of the second sampling result, where N represents the number of sampling points.

[0042] Step S5: Stitch the first sampling result and the second sampling result to generate the target sampling result.

[0043] Specifically, after obtaining the first sampling result and the second sampling result, since both the first sampling result and the second sampling result are in matrix form, stitch the first sampling result and the second sampling result to generate the target sampling result in matrix form.

[0044] The present invention creates a mask file based on saliency. The specific process is as follows: When establishing the mask file, it is necessary to perform binarization and dilation processing on the average saliency. The purpose of binarization is to establish the mask file; the dilation algorithm is to fill data gaps and make the main deformation areas form a whole. The mask file established through binarization and dilation processing can distinguish near-field, far-field, and local deformations, and the mask area with the largest area is the near-field region.

[0045] The present invention also performs quadtree sampling of saliency. The specific process is as follows: Perform quadtree sampling on the near-field using the masked average saliency map, and uniformly sample the far-field using a constant large window. During the process of performing quadtree sampling of saliency, use the standard deviation of saliency in the window As a reference index for near-field quadtree sampling. When the is greater than the set threshold, the window is equally divided into four small windows and iteratively segmented until the in the partition window is less than the threshold or the window is smaller than the set minimum window size. Similar to the traditional quadtree algorithm, the image also needs to be cropped or extended to a length and width of pixels (n = 0, 1, 2...) before sampling for quaternary segmentation.

[0046] In summary, the beneficial effects of an InSAR co-seismic deformation downsampling method based on image saliency provided by the present invention are as follows. First, the sampling method proposed by the present invention is simple and effective, which can retain the details of the main deformation area, suppress the interference of far-field or local deformation, thereby obtaining reliable co-seismic deformation sampling data and improving the reliability of subsequent research such as focal mechanism inversion.

[0047] Second, different from the downsampling methods based on resolution and pre-inversion, the sampling method proposed by the present invention does not require prior fault information and pre-inversion, reduces the complexity of the sampling process, and enhances the applicability of the method.

[0048] Embodiment 2 Please refer to Figure 5 , which shows the internal structure diagram of an InSAR co-seismic deformation downsampling system provided by the present invention, including: A saliency extraction module 100 for extracting the average saliency of a preset InSAR deformation map to obtain a multi-scale average saliency map; A mask file generation module 200 for binarizing and dilating the multi-scale average saliency map in sequence to generate a mask file. The mask area with the largest area in the mask file is the near-field area, and other areas except the near-field area in the mask file are the far-field areas.

[0049] A size expansion module 300 for respectively performing size expansion processing on the mask file of the near-field area and the multi-scale average saliency map to generate a masked multi-scale average saliency map; A sampling module 400 for performing quadtree sampling on the near-field area based on the masked multi-scale average saliency map to generate a first sampling result, and performing uniform sampling on the far-field area based on a constant window to generate a second sampling result; A target sampling result generation module 500 for splicing the first sampling result and the second sampling result to generate a target sampling result.

[0050] The beneficial effects of an InSAR co-seismic deformation downsampling system based on image saliency provided by the present invention are as follows. This system is simple and effective, can retain the details of the main deformation area, suppress the interference of far-field or local deformation, thereby obtaining reliable co-seismic deformation sampling data, and improve the reliability of subsequent research such as focal mechanism inversion. In addition, different from the downsampling systems based on resolution and pre-inversion, the sampling system proposed in the present invention does not require prior fault information and pre-inversion, reduces the complexity of the sampling process, and enhances the applicability of the system.

[0051] An InSAR co-seismic deformation downsampling system based on image saliency in the embodiments of the present application can be a device, or a component, integrated circuit, or chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. Exemplarily, the mobile electronic device can be a mobile phone, a tablet computer, a laptop computer, a handheld computer, a vehicle-mounted electronic device, a wearable device, an ultra-mobile personal computer (UMPC), a netbook, or a personal digital assistant (PDA), etc., and the non-mobile electronic device can be a server, a Network Attached Storage (NAS), a personal computer (PC), a television (TV), a teller machine, or a self-service machine, etc. The embodiments of the present application do not make specific limitations.

[0052] An InSAR co-seismic deformation downsampling system based on image saliency in the embodiments of the present application can be a device with an operating system. The operating system can be the Android operating system, the iOS operating system, or other possible operating systems. The embodiments of the present application do not make specific limitations.

[0053] An InSAR co-seismic deformation downsampling system based on image saliency provided in the embodiments of the present application can implement Figures 1 to 4 each process implemented by an InSAR co-seismic deformation downsampling system in the method embodiments. To avoid repetition, it will not be elaborated here.

[0054] Optionally, the embodiments of the present application further provide an electronic device, including a processor, a memory, and a program or instruction stored on the memory and executable on the processor. When the program or instruction is executed by the processor, it implements each process of the above-mentioned method embodiment of an InSAR co-seismic deformation downsampling method based on image saliency, and can achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0055] The embodiments of the present application also provide a readable storage medium, on which a program or instructions are stored. When the program or instructions are executed by a processor, the various processes of the above embodiments of a method for InSAR coseismic deformation downsampling based on image saliency are implemented, and the same technical effects can be achieved. To avoid repetition, details are not described herein again.

[0056] Among them, the processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs, etc.

[0057] It should be noted that in this article, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in a reverse order according to the functions involved. For example, the described methods may be performed in an order different from that described, and various steps may be added, omitted, or combined. Additionally, the features described with reference to certain examples may be combined in other examples.

[0058] Through the description of the above embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc) and includes several instructions for causing a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present application.

[0059] The embodiments of the present application have been described above in conjunction with the accompanying drawings. However, the present application is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present application, those of ordinary skill in the art can also make many forms without departing from the purpose of the present application and the scope protected by the claims, and all of them belong to the protection scope of the present application.

Claims

1. An InSAR co-seismic deformation downsampling method based on image saliency, characterized in that Including: Performing average significance extraction on a preset InSAR deformation map to obtain a multi-scale average significance map; After successively performing binarization and dilation processing on the multi-scale average significance map, generating a mask file, where the mask region with the largest area in the mask file is the near-field region, and other regions in the mask file except the near-field region are far-field regions; Performing size expansion processing on the mask file of the near-field region and the multi-scale average significance map respectively to generate a masked multi-scale average significance map; Generating a first sampling result by performing quadtree sampling on the near-field region based on the masked multi-scale average significance map and the size-expanded InSAR deformation map, and generating a second sampling result by performing uniform sampling on the far-field region based on a constant window; Stitching the first sampling result and the second sampling result to generate a target sampling result.

2. The method for downsampling InSAR co-seismic deformation based on image saliency according to claim 1, wherein The process of performing average significance extraction on a preset InSAR deformation map to obtain a multi-scale average significance map includes: Obtaining a two-dimensional matrix image of the co-seismic deformation of the InSAR to be sampled; Setting the sizes of a sliding window and a search window, where the size of the sliding window is a sequence and the size of the search window is a fixed constant; Expanding the width of the normalized InSAR deformation map outward and filling it with null values; Calculating the average value of the non-null points of the sliding window and the search window; Calculating the absolute value of the difference between the average values, and assigning the absolute value to all pixel points in the sliding window as the significance of each pixel point in the sliding window; Moving the sliding window and the search window to traverse the image to generate the significance of the image in the deformation field; Changing the size of the sliding window and repeating the above steps to generate significance maps of different scales in multiple deformation fields; Calculating the average significance of the image based on the significance maps of different scales in multiple deformation fields to generate a multi-scale average significance map.

3. A method for downsampling InSAR co-seismic deformation based on image saliency according to claim 1, characterized in that, Obtaining the first sampling result includes: Performing size expansion processing on the preset InSAR deformation map to generate an expanded InSAR deformation map; Performing quadtree sampling according to the expanded InSAR deformation map, the masked multi-scale average significance map, a set minimum window, a maximum window, and a segmentation threshold, and after dividing the image into four equal parts, obtaining four sub-windows; Calculating the standard deviation of each sub-window respectively, and performing iterative quadtree segmentation on each sub-window that satisfies the standard deviation being greater than the set threshold until the standard deviation of the sub-window is less than the set threshold or the size of the sub-window is equal to the minimum window, generating multiple segmented sub-windows; Calculating the longitude, latitude, and deformation average value of the sampling points of each segmented sub-window; Generating a first sampling result based on the longitude, latitude, and deformation average value.

4. A method for downsampling InSAR co-seismic deformation based on image saliency according to claim 3, characterized in that The first sampling result is a matrix with N rows and 3 columns, the first column is the longitude, the second column is the latitude, and the third column is the deformation average value, where N represents the number of sampling points.

5. A method for downsampling InSAR co-seismic deformation based on image saliency according to claim 1, characterized in that The process of uniformly sampling the far-field region based on a constant window to generate a second sampling result includes: Setting a constant window based on the maximum window set by the quadtree sampling; Segmenting a preset InSAR deformation map based on the constant window to obtain multiple segmented far-field regions; Calculating the average deformation value of each window in each of the segmented far-field regions; Obtaining a second sampling result based on the average deformation value.

6. An InSAR co-seismic deformation downsampling system based on image saliency, characterized in that, Including: A significance extraction module for performing average significance extraction on a preset InSAR deformation map to obtain a multi-scale average significance map; A mask file generation module for successively performing binarization and dilation processing on the multi-scale average significance map to generate a mask file. The mask region with the largest area in the mask file is the near-field region, and the other regions in the mask file except the near-field region are the far-field regions; A size expansion module for respectively performing size expansion processing on the near-field region and the multi-scale average significance map to generate a masked multi-scale average significance map; A sampling module for performing quadtree sampling on the near-field region based on the masked multi-scale average significance map and the InSAR deformation map after size expansion to generate a first sampling result, and uniformly sampling the far-field region based on a constant window to generate a second sampling result; A target sampling result generation module for splicing the first sampling result and the second sampling result to generate a target sampling result.

7. An electronic device, characterized in that, Including a processor, a memory, and a program or instruction stored on the memory and executable on the processor. When the program or instruction is executed by the processor, the steps of a method for downsampling InSAR co-seismic deformation based on image significance as described in any one of claims 1-5 are implemented.

8. A readable storage medium, characterized in that, A program or instruction is stored on the readable storage medium. When the program or instruction is executed by the processor, the steps of a method for downsampling InSAR co-seismic deformation based on image significance as described in any one of claims 1-5 are implemented.

Images