Ground subsidence monitoring method and system

Abstract

The invention discloses a ground subsidence monitoring method and system, and relates to the technical field of geological monitoring, and the method comprises the steps: unifying the ground deformation time sequences of an airport on a plurality of satellite platforms to the same reference; selecting an SAR surface image of one satellite platform, and searching a homonymous pixel point corresponding to each pixel point in the SAR surface image of the other platform according to a preset search radius by taking each pixel point as a center; and according to a search result, fusing the earth surface deformation time sequence of each pixel point and the earth surface deformation time sequence of the homonymous pixel points searched corresponding to the pixel point through singular value decomposition (SVD) to obtain a long-time-span high-time-resolution deformation time sequence capable of reflecting real land subsidence. According to the invention, real earth surface deformation characteristics with long monitoring period and high time resolution can be obtained.

Description

Technical Field

[0001] The present invention relates to the field of geological monitoring technology, and in particular to a method and system for monitoring ground subsidence. Background Art

[0002] Ground subsidence is a phenomenon in which the ground surface deforms downward due to natural or human activities. Especially in airport areas, ground subsidence, cracks and collapses are likely to occur due to construction and aircraft loads on runways, affecting the normal operation of the airport. Therefore, it is necessary to conduct full coverage and high-precision surface deformation monitoring of the airport and surrounding areas, timely and accurately obtain the temporal and spatial evolution of airport deformation, and ensure the safety of airport infrastructure and aircraft flight safety.

[0003] In the existing technology, conventional ground subsidence monitoring methods (such as leveling and GPS, etc.) have high measurement accuracy, but can only obtain some discrete airport point deformation information, and cannot fully monitor the entire airport and its surrounding areas, nor can it obtain the historical deformation characteristics of the airport. Synthetic aperture radar interferometry InSAR technology has the characteristics of all-day, all-weather and non-contact, and is a monitoring method that is very suitable for airport areas. However, due to the limitations of the life of a single satellite platform, revisit cycle and satellite attitude, it is often impossible to obtain high-temporal resolution real surface deformation over a long span of time.

[0004] In summary, how to obtain the real surface deformation characteristics with high temporal resolution over a long monitoring period is an important issue that needs to be solved urgently. Summary of the invention

[0005] The embodiments of the present invention provide a method and system for monitoring ground subsidence, which can solve the problem in the prior art that it is impossible to obtain the real surface deformation characteristics with high time resolution over a long monitoring period.

[0006] An embodiment of the present invention provides a method for monitoring land subsidence, comprising the following steps: Obtain SAR surface images of the airport from multiple satellite platforms, and use InSAR technology to obtain the surface deformation time series from the SAR surface images of each satellite platform; Unify the geocoding and LOS deformation in the surface deformation time series of multiple satellite platforms into the same reference datum; Select a SAR surface image of a satellite platform, and search for pixels with the same name corresponding to each pixel point in a SAR surface image of another satellite platform according to a preset search radius, with each pixel point as the center; wherein the pixels with the same name represent pixels at the same geographical location; The time series of surface deformation of each pixel is fused with the time series of surface deformation of the corresponding homologous pixel searched for that pixel through singular value decomposition (SVD) to obtain a high-time-resolution deformation time series with a long time span that can reflect the true ground settlement.

[0007] Further, the specific steps of the time series of surface deformation of the searched homologous pixel include: If there is one corresponding homologous pixel within the search radius, the time series of surface deformation of this point is used as the search result; If there are two or more homologous pixels within the search radius, the average value of the time series of surface deformation of all homologous pixels is taken as the search result.

[0008] Further, the specific steps of fusing the time series of surface deformation of each pixel with the time series of surface deformation of the corresponding homologous pixel searched for that pixel through singular value decomposition (SVD) include: Record the cumulative deformation of each pixel of the first platform in the i scene image as: S i ( i = 1, 2, 3,..., m ), in the second platform, the cumulative deformation of the homologous pixel corresponding to each pixel of the first platform in the j scene image is T j ( j = 1, 2, 3,..., n ); where m represents the total number of scenes of each pixel of the first platform, n represents the total number of medium scenes of the homologous pixel of the second platform; Let ST k ( k = 1, 2, 3,..., m + n ) represent the cumulative deformation of the fusion time series in the k scene image; t k and v k represent the time difference and deformation rate corresponding to the k scene and the k +1 scene image after fusion; Establish the relationship between rate and displacement: where , are respectively the first type of data at thei The index in the fusion time series ST corresponding to a deformation and the j index in the fusion time series ST corresponding to the second type of data; Establish a matrix equation according to the relationship between the rate and the displacement: where the matrix t is a matrix of order ( m + n - 2)*( m + n - 1); v represents the rate vector, S represents the deformation time series of the first platform, T represents the deformation time series of the second platform; Use singular value decomposition SVD for the matrix equation to obtain the rate vector v ; According to the time and deformation rate of each scene image, obtain the fused cumulative deformation of the surface deformation time series of each pixel point and the surface deformation time series of the corresponding homologous pixel point searched for by this pixel point.

[0009] Further, the obtaining of the surface deformation time series in the SAR surface image of each satellite platform by the InSAR technology specifically includes the following steps: Preprocess the SAR images of the airport surface on multiple satellite platforms to obtain differential interferograms; Use the InSAR technology to perform deformation time series inversion on the SAR images of each satellite platform to obtain the surface deformation time series of the airport area.

[0010] Further, the unifying of the geocoding and the LOS - direction deformation in the surface deformation time series of multiple satellite platforms to the same reference datum specifically includes the following steps: When performing deformation time series inversion, keep the longitude and latitude of the deformation reference points consistent; After the deformation time series inversion is completed, unify the geocoding of the deformation time series inversion results of multiple satellite platforms to the same geographic coordinate system; project the LOS - direction deformation of the deformation time series inversion of multiple satellite platforms onto the vertical direction.

[0011] An embodiment of the present invention provides a ground settlement monitoring system, including: A deformation time series acquisition module, configured to acquire the SAR surface images of the airport on multiple satellite platforms, and acquire the surface deformation time series in the SAR surface image of each satellite platform by the InSAR technology; A reference datum unifying module, configured to unify the geocoding and the LOS - direction deformation in the surface deformation time series of multiple satellite platforms to the same reference datum; The homonymous pixel search module is used to select the SAR surface image of one satellite platform, and with each pixel as the center, search for the corresponding homonymous pixel of each pixel in the SAR surface image of another satellite platform according to the preset search radius; The fusion module is used to fuse the surface deformation time series of each pixel with the surface deformation time series of the corresponding homonymous pixel searched for by this pixel through singular value decomposition (SVD) to obtain a high temporal resolution deformation time series with a long time span that can reflect the real ground settlement.

[0012] The embodiment of the present invention provides a method and system for monitoring ground settlement. Compared with the prior art, its beneficial effects are as follows: By searching for the corresponding homonymous pixels of the pixels of one satellite platform on another satellite platform, the pixels that need to be fused are determined. Furthermore, the surface deformation time series of each pixel is fused with the surface deformation time series of the corresponding homonymous pixel searched for by this pixel through singular value decomposition (SVD), obtaining the long-term surface deformation time series of the pixels at the same geographical location, increasing the resolution of the monitoring, and finally obtaining the long-term high-resolution surface deformation time series, truly reflecting the deformation of the ground settlement. Description of the Drawings

[0013] Figure 1 It is a flowchart of the SAR data fusion of different platforms for the ground settlement monitoring method provided by the embodiment of the present invention; Figure 2 It is the determination of homonymous points of different data for the ground settlement monitoring method provided by the embodiment of the present invention. Among them, (a) means there is no TerraSAR-X point within the search radius R; (b) means there is 1 TerraSAR-X point within the search radius; (c) means there are two or more TerraSAR-X points within the search radius; Figure 3 It is a schematic diagram of the time series fusion of the ground settlement monitoring method provided by the embodiment of the present invention; Figure 4 It is the fused InSAR deformation map of the ground settlement monitoring method provided by the embodiment of the present invention. Among them, (a) is the vertical cumulative displacement of the first period in October 2019, (b) is the vertical cumulative displacement of the first period in October 2020, (c) is the vertical cumulative displacement of the first period in October 2021, (d) is the vertical cumulative displacement of the first period in October 2022, (e) is the vertical cumulative displacement of the first period in October 2023, and (f) is the vertical cumulative displacement of the first period in October 2024; Figure 5The characteristic point time series fusion result of a ground settlement monitoring method provided by an embodiment of the present invention, where (a) is the displacement time series of the airport maintenance area P1, (b) is the displacement time series of the freight area P2, and (c) is the displacement time series of the terminal building P3; Figure 6 The comparison and verification of InSAR and GNSS results at each site of a ground settlement monitoring method provided by an embodiment of the present invention, where (a) is site G1, (b) is site G2, (c) is site G3, (d) is site G4, (e) is site G5, and (f) is site G6. Detailed implementation manners

[0014] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following describes the detailed implementation manners of the present invention with reference to the accompanying drawings. Many specific details are set forth in the following description to fully understand the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0015] Refer to Figure 1 , an embodiment of the present invention provides a ground settlement monitoring method, including the following steps: Step 1: Obtain SAR surface images of the airport on multiple satellite platforms, and obtain surface deformation time series in the SAR surface images of each satellite platform through InSAR technology.

[0016] Step 2: Unify the geocoding and LOS-direction deformation in the surface deformation time series of multiple satellite platforms to the same reference datum.

[0017] Step 3: Select the SAR surface image of one satellite platform, and search for the corresponding homologous pixel points of each pixel point in the SAR surface image of another satellite platform with a preset search radius centered on each pixel point.

[0018] Step 4: Fuse the surface deformation time series of each pixel point with the surface deformation time series of the corresponding homologous pixel point searched for by the pixel point through singular value decomposition (SVD) to obtain a high-time-resolution deformation time series with a long time span that can reflect the true ground settlement.

[0019] The detailed steps are introduced as follows: 1. Obtain the surface deformation of the airport. First, preprocess various SAR data to obtain their respective differential interferograms, then use time series InSAR technology to invert the deformation time series, and finally obtain the surface deformation time series of the airport area.

[0020] 2. Unify the reference benchmark. To unify the reference benchmark for data from different platforms, first, ensure that the deformation reference points are consistent when solving different data; second, after inversely calculating the deformations of different data, geocode them all into a unified geographic coordinate system; finally, since the LOS directions of different data are inconsistent and the airport area is generally flat, project the LOS-direction deformations of different data onto the vertical direction. In this way, the unification of the reference benchmark is completed.

[0021] 3. Identify homologous points. Since the deformation pixel points calculated from data of different platforms do not completely coincide, it is necessary to determine the homologous points of different data before data fusion. The identification strategy is as follows: taking the pixel point of the first platform's data as the center, set an appropriate search radius to find the homologous pixel point of the other platform corresponding to this point. There are three search situations in total. The first is that there is no point of the second platform within the search radius, then this point does not participate in the fusion; the second is that there is 1 point of the second platform within the search radius, then this point is regarded as a homologous point, and this group of homologous points is fused in the time series; the third is that there are 2 or more points of the second platform within the search radius, then take the average value of these points as the homologous point and then participate in the fusion, as Figure 2 shown.

[0022] 4. SVD fusion of deformation time series. Using SVD to fuse the time series of different satellite platforms is essentially a process of transforming the fusion of two time series into solving a matrix equation. For example: Let the cumulative deformation of the first platform's data in the i th scene image be: ( i = 1, 2, 3,..., m ), and the cumulative deformation of the second platform's data in the j th scene image be ( j = 1, 2, 3,..., n ). ( k = 1, 2, 3,..., m + n ) represents the cumulative deformation of the fusion time series in the k th scene image, and represent the time difference and deformation rate corresponding to the k th scene and the k + 1th scene images after fusion, is a known quantity, is an unknown quantity. Establish the relationship between the rate and displacement:

[0023] (1).

[0024] (2).

[0025] In the formula, , The first data are i The deformation corresponds to the index in the fusion time series ST and the second data j Each deformation corresponds to the index in the fused time series ST. The above equations can be combined to establish the following matrix equation:

[0026] (3).

[0027] In the formula, the matrix t for( m + n -2)*( m + n -1) order matrix, due to its rank deficiency, cannot be solved directly, so the generalized inverse matrix is ​​obtained through SVD decomposition to obtain the rate vector v Finally, the corresponding cumulative deformation is solved according to each moment and rate. At this point, the deformation time series fusion of data from different platforms is completed, such as Figure 4 As shown in Figure 2. The fused time series has a longer time span than a single time series, has more monitoring nodes, and has better continuity of the time series, which can more accurately reflect the real spatiotemporal evolution of the airport's surface deformation, as shown in Figure 2. Figure 3 As shown in the figure, the deformation time series of Sentinel-1 is S, the deformation time series of TerraSAR-X is T, and the fused time series is ST.

[0028] And, if Figure 5 As shown in the figure, the dark blue dots are the vertical displacement of TerraSAR-X, the red dots are the vertical displacement of Sentinel-1, and the black circles are the fused long-time displacement. Fusion of time series can not only unify the two time series to the same benchmark and obtain a long-time deformation series containing two time periods, but also encrypt the monitoring nodes and improve the monitoring time resolution. Figure 6 As shown in the figure, in the data overlap segment, the RMSE of the InSAR and GNSS displacements of the six stations is less than 5 mm except for G2 and G4, which have an E of 5.16 mm and 6.48 mm respectively. The results show that the InSAR data solution and different data fusion have high accuracy.

[0029] The effects of the present invention are as follows: 1. Through the fusion of data from different platforms, the influence of geometric distortion during side-view imaging of a single SAR satellite on the accuracy of airport deformation monitoring can be effectively reduced.

[0030] 2. Through geocoding, operations such as converting LOS deformation to vertical deformation are performed to unify InSAR data from different platforms to the same reference datum. Also, by taking the pixel points of one satellite data as the search center and setting an appropriate search radius, the corresponding homologous pixel points of another satellite data are found, improving the accuracy of data fusion.

[0031] 3. Using the SVD method, the InSAR deformation results of different platforms are fused to obtain a high temporal resolution deformation time series with a long time span for the airport, more accurately reflecting the spatio-temporal evolution characteristics of airport deformation.

[0032] An embodiment of the present invention provides a ground settlement monitoring system, including: A deformation time series acquisition module, configured to acquire SAR surface images of the airport on multiple satellite platforms, and obtain surface deformation time series in the SAR surface images of each satellite platform through InSAR technology.

[0033] A reference datum unification module, configured to unify the geocoding and LOS deformation in the surface deformation time series of multiple satellite platforms to the same reference datum.

[0034] A homologous pixel point search module, configured to select the SAR surface image of one satellite platform, and search for the corresponding homologous pixel points of each pixel point in the SAR surface image of another satellite platform with a preset search radius centered on each pixel point.

[0035] A fusion module, configured to fuse the surface deformation time series of each pixel point with the surface deformation time series of the homologous pixel points corresponding to the pixel points through singular value decomposition SVD, to obtain a high temporal resolution deformation time series with a long time span that can reflect the true ground settlement.

[0036] A specific embodiment is as follows: The present invention discloses a ground settlement monitoring method, and the specific steps are as follows: S1. Obtain the surface deformation of the airport on multiple platforms.

[0037] S2. Unify the data of different platforms to a reference datum.

[0038] S3. It is necessary to determine the homologous points of different data before data fusion.

[0039] S4. Use SVD to fuse the time series of different satellite platforms.

[0040] Finally, the effects of the present invention can be summarized as: The purpose of accurately monitoring the airport area by fusing SAR data from different platforms using the SVD method is achieved. The problem that a single satellite platform cannot obtain the surface deformation of the airport with a high time resolution over a long time span is solved, and more accurate spatio-temporal evolution characteristics of the airport surface deformation are obtained. This method is applicable to other airports covered by multi-source SAR data and has high universality.

[0041] The above-described embodiments merely represent several implementation manners of the present invention, and their descriptions are relatively specific and detailed, but should not be construed as limiting the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention patent shall be subject to the appended claims.

Claims

1. A method for monitoring land subsidence, characterized in that: The following steps are involved: Obtain SAR surface images of the airport from multiple satellite platforms, and use InSAR technology to obtain the surface deformation time series from the SAR surface images of each satellite platform; Unify the geocoding and LOS deformation in the surface deformation time series of multiple satellite platforms into the same reference datum; Select a SAR surface image of a satellite platform, and search for pixels with the same name corresponding to each pixel point in a SAR surface image of another satellite platform according to a preset search radius, with each pixel point as the center; wherein the pixels with the same name represent pixels at the same geographical location; The surface deformation time series of each pixel point is fused with the surface deformation time series of the pixel point with the same name searched corresponding to the pixel point through singular value decomposition (SVD) to obtain a high-temporal-resolution deformation time series with a long time span that can reflect the actual ground subsidence.

2. A method for monitoring land subsidence as claimed in claim 1, characterized in that: The surface deformation time series of the searched pixel points with the same name specifically comprises the following steps: There is a corresponding pixel point with the same name within the search radius, and the surface deformation time series of this point is used as the search result; If there are two or more pixels with the same name within the search radius, the average of the surface deformation time series of all pixels with the same name is taken as the search result.

3. A method for monitoring land subsidence as claimed in claim 1, characterized in that: The surface deformation time series of each pixel point is fused with the surface deformation time series of the pixel point with the same name searched corresponding to the pixel point through singular value decomposition SVD, and the specific steps include: Record each pixel point of the first platform in the i The accumulated deformation of the scene image is: S i ( i =1, 2, 3, ..., m ), in the second platform, each pixel point of the first platform corresponds to the pixel point with the same name in the j The cumulative deformation of the scene image is T j ( j =1, 2, 3, ..., n );in, m Indicates the total number of scenes for each pixel of the first platform, n Indicates the total number of scenes with the same pixel name on the second platform; Will ST k ( k =1, 2, 3, ..., m + n ) indicates that the fused time series is k Cumulative deformation of the scene image; t k and v k After fusion k Scenery and k +1 The time difference and deformation rate corresponding to the scene image; Establish the relationship between velocity and displacement: in, , The first platform i The deformation of the scene image corresponds to the index in the fusion time series ST and the index in the second platform j The deformation of the scene image corresponds to the index in the fused time series ST; The matrix equation is established based on the relationship between velocity and displacement: Among them, the matrix t for( m + n -2)*( m + n -1) order matrix; v represents the velocity vector, S represents the deformation time sequence of the first platform, T represents the deformation time sequence of the second platform; Apply singular value decomposition SVD to the matrix equation to obtain the rate vector v ; According to the time and deformation rate of each scene image, the fused cumulative deformation of the surface deformation time series of each pixel point and the surface deformation time series of the pixel point with the same name searched corresponding to the pixel point is obtained.

4. A method for monitoring land subsidence as claimed in claim 1, characterized in that: The specific steps of obtaining the surface deformation time series from the SAR surface image of each satellite platform by using the InSAR technology include: Preprocess the SAR images of the airport surface from multiple satellite platforms to obtain differential interferograms; The InSAR technology is used to perform deformation time series inversion on the SAR images of each satellite platform to obtain the surface deformation time series of the airport area.

5. A method for monitoring land subsidence as claimed in claim 1, characterized in that: The specific steps of unifying the geocoding and LOS deformation in the surface deformation time series of multiple satellite platforms onto the same reference datum include: When performing deformation time series inversion, the longitude and latitude of the deformation reference point remain consistent; After the deformation time series inversion is completed, the geocoding of the deformation time series inversion results of multiple satellite platforms is unified into the same geographic coordinate system; the LOS deformation of the deformation time series inversion of multiple satellite platforms is projected into the vertical direction.

6. A ground subsidence monitoring system, characterized in that: include: The deformation time series acquisition module is used to obtain the SAR surface images of the airport on multiple satellite platforms, and obtain the surface deformation time series in the SAR surface images of each satellite platform through InSAR technology; The reference datum unification module is used to unify the geocoding and LOS deformation in the surface deformation time series of multiple satellite platforms into the same reference datum; The same-name pixel search module is used to select a SAR surface image of a satellite platform, and search for the same-name pixel corresponding to each pixel in the SAR surface image of another satellite platform according to a preset search radius with each pixel as the center; The fusion module is used to fuse the surface deformation time series of each pixel with the surface deformation time series of the pixel with the same name searched corresponding to the pixel through singular value decomposition (SVD), so as to obtain a high-time resolution deformation time series with a long time span that can reflect the actual ground subsidence.

Images