CN117274342A - Hydraulic engineering deformation monitoring method based on satellite data - Google Patents
Hydraulic engineering deformation monitoring method based on satellite data Download PDFInfo
- Publication number
- CN117274342A CN117274342A CN202311549722.8A CN202311549722A CN117274342A CN 117274342 A CN117274342 A CN 117274342A CN 202311549722 A CN202311549722 A CN 202311549722A CN 117274342 A CN117274342 A CN 117274342A
- Authority
- CN
- China
- Prior art keywords
- deformation
- image
- map
- matrix
- sar
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 title claims abstract description 33
- 239000011159 matrix material Substances 0.000 claims abstract description 70
- 238000000547 structure data Methods 0.000 claims abstract description 37
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 11
- 238000010586 diagram Methods 0.000 claims description 10
- 238000005070 sampling Methods 0.000 claims description 8
- 230000004913 activation Effects 0.000 claims description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 3
- 238000003325 tomography Methods 0.000 description 3
- 230000009466 transformation Effects 0.000 description 2
- 235000014435 Mentha Nutrition 0.000 description 1
- 241001072983 Mentha Species 0.000 description 1
- 238000012952 Resampling Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000005305 interferometry Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- LWYJUZBXGAFFLP-OCNCTQISSA-N menogaril Chemical compound O1[C@@]2(C)[C@H](O)[C@@H](N(C)C)[C@H](O)[C@@H]1OC1=C3C(=O)C(C=C4C[C@@](C)(O)C[C@H](C4=C4O)OC)=C4C(=O)C3=C(O)C=C12 LWYJUZBXGAFFLP-OCNCTQISSA-N 0.000 description 1
- 235000014569 mints Nutrition 0.000 description 1
- 230000000149 penetrating effect Effects 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
- G06T7/337—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving reference images or patches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
- G06T7/344—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
- G06T2207/10044—Radar image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention relates to the technical field of image processing, and discloses a hydraulic engineering deformation monitoring method based on satellite data, which comprises the following steps: carrying out graph structuring on the registered SAR image to obtain graph structure data; inputting the image structure data and the registered SAR image into an image processing model, and outputting a deformation reference pixel matrix; generating a first elevation map and a second elevation map based on the SAR image shot at the current time and the SAR image shot at the deformation reference time point, obtaining an elevation map by differentiating the first elevation map and the second elevation map, obtaining a deformation map by performing point multiplication on the elevation map and a deformation reference pixel matrix, and then calculating the average value of pixel values in a monitoring area in the deformation map as the deformation of the monitoring area; according to the invention, the image processing model is trained through a large amount of data, the pixels for calculating the deformation can be automatically selected from the SAR image, and a good result can be obtained when the image processing model is applied to deformation monitoring of hydraulic engineering.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a hydraulic engineering deformation monitoring method based on satellite data.
Background
The space-borne synthetic aperture radar interferometry has the characteristics of all weather, high precision, certain ground penetrating power and the like, and the space information of continuous surface displacement can be obtained by utilizing the phase information sent by the satellite radar. Over the years, multi-baseline InSAR technology has evolved to create a variety of techniques such as interferogram Stacking (Stacking), permanent scatterer method (PermanentScatterer, PS), least Square (LS), small baseline subset (SmallBAselineSubset, SBAS), differential SAR tomography or differential TomoSAR (DifferentialSARtomography, DTomoSAR), multi-scale InSAR time series analysis (MultiscaleInSARTimeSeries, MInTS), and the like. The principles based on the technologies are characterized, the solved problems are focused, and many successful applications are achieved, and particularly, the differential SAR tomography technology realizes the differential SAR tomography deformation monitoring of the permanent scatterer with high resolution.
The method for selecting the scatterer target with stable phase by utilizing the coherence coefficient threshold value is the most direct and simple method, and the area where the hydraulic engineering is located generally covers vegetation, has complex terrain and accompanies a large area of water area. If the coherence coefficient threshold is adjusted to consider the detection probability of the permanent scatterers, namely, as many real permanent scatterers as possible are successfully selected, a large number of unreliable permanent scatterers are selected at the same time; if the coherence coefficient threshold is adjusted to ensure the duty ratio of the real permanent scatterer, the selected result only contains a small amount of coherence loss targets, and the accuracy of deformation monitoring is difficult to ensure in any way.
Disclosure of Invention
The invention provides a hydraulic engineering deformation monitoring method based on satellite data, which solves the technical problem that in the related art, accuracy of deformation monitoring is difficult to guarantee by selecting a scatterer target with a stable phase by using a coherence coefficient threshold method.
The invention provides a hydraulic engineering deformation monitoring method based on satellite data, which comprises the following steps:
step 101, extracting SAR images of N previous target areas and the current time;
step 102, selecting one from N SAR images as a main image, and registering other SAR images to a main image space;
step 103, carrying out graph structuring on the registered SAR image to obtain graph structure data, wherein nodes of the graph structure data are mapped with pixels of the SAR image one by one, and if the pixels mapped by the two nodes are adjacent in the SAR image, edges exist between the two nodes; each node has an initial feature vector, one component of which corresponds to the value of one dimension of the pixel;
step 104, inputting the graph structure data and the registered SAR image into an image processing model, wherein the image processing model comprises: the system comprises a first convolution module, a stability pixel extraction module and a first logic layer, wherein the first convolution module is used for respectively inputting N SAR images and outputting N first area diagrams, and the size of the first area diagrams is consistent with that of the SAR images; the SAR image is input into a stability pixel extraction module, and a stability pixel matrix is output; the first logic layer inputs the first region map and the stability pixel matrix, and outputs a deformation reference pixel matrix;
the calculation formula of the stability pixel extraction module is as follows:
;
;
;
;
;
;
;
;
;
;
;
wherein the method comprises the steps ofAnd->The input feature matrix input by the first time step and the t time step are respectively represented, and the ith row vector of the input feature matrix of the t time step is the initial feature vector of the ith node of the t-th graph structure data; />Representing the sum of the adjacency matrix and the identity matrix of the diagram structure data, +.>Representation->Degree matrix of->Representing an initial feature matrix>、、/>Output feature matrices respectively representing the t-th, t-1 th and N-th time steps; />A first gating matrix representing the t-th time step,>a second gating matrix representing the t-th time step,>represents dot product->、/>、Representing a first coding feature matrix, a second coding feature matrix and a third coding feature matrix, respectively,/->、、/>、/>、/>Respectively representing a first weight parameter, a second weight parameter, a third weight parameter, a fourth weight parameter and a fifth weight parameter, +.>、/>、/>Respectively representing a first bias parameter, a second bias parameter and a third bias parameter; />Representing a stability pel vector>Indicating that a value of more than 0.5 is changed to 1, a value of not more than 0.5 is changed to 0,/and a value of not more than 0.5 is changed to 0->Representing hyperbolic tangent function, ">Representing an activation function->Representation->E represents a matrix of stable pixels;
step 105, generating a first elevation map and a second elevation map based on the SAR image shot at the current time and the SAR image shot at the deformation reference time point, obtaining an elevation map by differentiating the first elevation map and the second elevation map, obtaining a deformation map by dot multiplying the elevation map and the deformation reference pixel matrix, and then calculating the average value of pixel values in a monitoring area in the deformation map as the deformation of the monitoring area.
Further, the target area contains the hydraulic engineering for which the deformation is to be monitored.
Further, the SAR images of the N target areas include SAR images photographed at deformation reference time points.
Further, each SAR image map is structured as a map structure data.
Further, the first convolution module comprises a convolution layer and an up-sampling layer, the convolution layer inputs the SAR image, the convolution layer outputs the feature map, and the up-sampling layer up-samples the feature map to obtain a first region map.
Further, the calculation formula of the first logic layer is as follows:
wherein B represents a deformed reference pixel matrix,indicating that a value greater than 0.5 is changed to 1 and a value not greater than 0.5 is changed to 0, E indicates a stability matrix of picture elements, < >>Representing the v first region map, < >>Represents dot product->Representing a hyperbolic tangent function.
Further, the monitoring area is a circular area centered on the monitoring point, and the radius of the circular area is a parameter set by people.
The invention provides a hydraulic engineering deformation monitoring system based on satellite data, which comprises:
an image acquisition module for extracting SAR images of the current time and the previous N target areas;
an image registration module for selecting one from the N SAR images as a main image and registering the other SAR images to a main image space;
the image structure data generation module is used for carrying out image structuring on the registered SAR image to obtain image structure data, nodes of the image structure data are mapped with pixels of the SAR image one by one, and if the pixels mapped by the two nodes are adjacent in the SAR image, edges exist between the two nodes; each node has an initial feature vector, one component of which corresponds to the value of one dimension of the pixel;
the deformation reference generation module is used for inputting the image structure data and the registered SAR image into the image processing model and outputting a deformation reference pixel matrix;
the deformation amount calculation module generates a first elevation map and a second elevation map based on the SAR image shot at the current time and the SAR image shot at the deformation reference time point, obtains the elevation map by differentiating the first elevation map and the second elevation map, obtains the deformation amount map by performing point multiplication on the elevation map and the deformation reference pixel matrix, and then calculates the average value of pixel values in a monitoring area in the deformation amount map as the deformation amount of the monitoring area.
The present invention provides a storage medium storing non-transitory computer readable instructions that, when executed by a computer, are capable of performing the steps of a satellite data based hydraulic engineering deformation monitoring method as described above.
The invention has the beneficial effects that: according to the invention, the image processing model is trained through a large amount of data, the pixels for calculating the deformation can be automatically selected from the SAR image, the pixel is applied to covering vegetation in an area, and a good result can be obtained when the deformation of the hydraulic engineering with complex terrain is monitored.
Drawings
FIG. 1 is a flow chart of a hydraulic engineering deformation monitoring method based on satellite data;
fig. 2 is a schematic block diagram of a hydraulic engineering deformation monitoring system based on satellite data.
In the figure: an image acquisition module 201, an image registration module 202, a graph structure data generation module 203, a deformation reference generation module 204, and a deformation amount calculation module 205.
Detailed Description
The subject matter described herein will now be discussed with reference to example embodiments. It is to be understood that these embodiments are merely discussed so that those skilled in the art may better understand and implement the subject matter described herein and that changes may be made in the function and arrangement of the elements discussed without departing from the scope of the disclosure herein. Various examples may omit, replace, or add various procedures or components as desired. In addition, features described with respect to some examples may be combined in other examples as well.
As shown in fig. 1, a hydraulic engineering deformation monitoring method based on satellite data includes the following steps:
step 101, extracting SAR images of N previous target areas and the current time;
in an embodiment of the invention, the target area comprises hydraulic engineering to be monitored for deformation, which includes dams, bridges, river channels, reservoirs, etc.
The pels of the SAR image include dimensions of phase, amplitude, etc.
The SAR images of the N target areas comprise SAR images shot at deformation reference time points.
The deformation amount monitored is the deformation amount relative to the deformation reference time point.
Step 102, selecting one from N SAR images as a main image, and registering other SAR images to a main image space;
the photographing time of the selected main image is not earliest nor latest, and is preferably one centered in the photographing times among the N SAR images.
The method for registering the SAR image into the main image space can adopt an image resampling method or a registration mathematical model method.
Step 103, carrying out graph structuring on the registered SAR image to obtain graph structure data, wherein nodes of the graph structure data are mapped with pixels of the SAR image one by one, and if the pixels mapped by the two nodes are adjacent in the SAR image, edges exist between the two nodes; each node has an initial feature vector, one component of which corresponds to the value of one dimension of the pixel;
each SAR image map is structured into map structure data;
step 104, inputting the graph structure data and the registered SAR image into an image processing model, wherein the image processing model comprises: the system comprises a first convolution module, a stability pixel extraction module and a first logic layer, wherein the first convolution module is used for respectively inputting N SAR images and outputting N first area diagrams, and the size of the first area diagrams is consistent with that of the SAR images;
in one embodiment of the invention, the first convolution module comprises a convolution layer and an up-sampling layer, the convolution layer is a ResNet (Residual Network) network, the feature map is output, the up-sampling layer carries out up-sampling on the feature map to obtain a first area map, and the up-sampling layer adopts deconvolution or linear interpolation.
In one embodiment of the present invention, the first convolution module performs independent training, and during the training, uses the difference between the first area image and the truth matrix as a training loss, the SAR image is divided into two categories, namely a water area and a non-water area, the elements of the truth matrix are mapped to the pixels of the SAR image, the values of the elements of the truth matrix in the water area are 0, and the values of the elements of the truth matrix in the non-water area are 0.
The SAR image is input into a stability pixel extraction module, and a stability pixel matrix is output;
the calculation formula of the stability pixel extraction module is as follows:
;
;
;
;
;
;
;
;
;
;
;
wherein the method comprises the steps ofAnd->The input feature matrix input by the first time step and the t time step are respectively represented, and the ith row vector of the input feature matrix of the t time step is the initial feature vector of the ith node of the t-th graph structure data; />Representing the sum of the adjacency matrix and the identity matrix of the diagram structure data, +.>Representation->Degree matrix of->Representing an initial feature matrix>、、/>Output feature matrices respectively representing the t-th, t-1 th and N-th time steps; />A first gating matrix representing the t-th time step,>the t time step is represented byTwo gating matrices>Represents dot product->、/>、Representing a first coding feature matrix, a second coding feature matrix and a third coding feature matrix, respectively,/->、、/>、/>、/>Respectively representing a first weight parameter, a second weight parameter, a third weight parameter, a fourth weight parameter and a fifth weight parameter, +.>、/>、/>Respectively representing a first bias parameter, a second bias parameter and a third bias parameter; />Representing a stability pel vector>Representation ofChanging the value of more than 0.5 to 1, changing the value of not more than 0.5 to 0,/for>Representing hyperbolic tangent function, ">Representing an activation function->Representation->E represents a matrix of stable picture elements.
Tensor operation of (2) is to +.>Which in turn map to values of elements of the stability matrix of picture elements.
In one embodiment of the invention, the ordering rules for the graph structure data are as follows: the earlier the shooting time of the SAR image corresponding to the map structure data, the earlier the ordering of the map structure data.
The ordering rules of the nodes are as follows: and ordering pixels row by row from the upper left corner of the SAR image to the right, wherein the ordering of the nodes is consistent with the ordering of the pixels mapped by the nodes.
In one embodiment of the present invention,representing a singmod function.
The adjacency matrix represents the connection relation between nodes, and the elements of the ith row and the jth column of the adjacency matrix are expressed asIf->There is an edge between the i-th node and the j-th node,if->There is no edge between the i-th node and the j-th node.
The first logic layer inputs the first region map and the stability pixel matrix, and outputs a deformation reference pixel matrix;
the calculation formula of the first logic layer is as follows:;
wherein B represents a deformed reference pixel matrix,indicating that a value greater than 0.5 is changed to 1 and a value not greater than 0.5 is changed to 0, E indicates a stability matrix of picture elements, < >>Representing the v first region map, < >>Represents dot product->Representing a hyperbolic tangent function;
in one embodiment of the invention, the stability pixel extraction module is combined with the first logic layer and the first convolution module to perform independent training, the difference between the deformation reference pixel matrix and the artificially marked pixel matrix is taken as a loss during training, the pixels with stable phases are selected for marking by a coherence coefficient threshold method (taking more selected pixels as a standard) during artificial marking, and then the marks in the water area and the marks irrelevant to calculation of the hydraulic engineering deformation are manually removed.
The parameters of the first convolution module are not updated when the stability pixel extraction module is trained.
In one embodiment of the present invention, the image processing model further includes a dimension transformation layer that inputs an initial feature vector of the graph structure data and upscales the initial feature vector.
The calculation formula of the dimension transformation layer is as follows:
;
for the up-right weight parameter, +.>Representing the initial feature vector before the dimension increase of the ith node,/th node>Representing the initial feature vector after the upstroke of the i-th node.
Step 105, generating a first elevation map and a second elevation map based on the SAR image shot at the current time and the SAR image shot at the deformation reference time point, obtaining an elevation map by differentiating the first elevation map and the second elevation map, obtaining a deformation map by dot multiplying the elevation map and the deformation reference pixel matrix, and then calculating the average value of pixel values in a monitoring area in the deformation map as the deformation of the monitoring area.
In one embodiment of the invention, the monitoring area is an area within the contour of the hydraulic engineering.
In one embodiment of the invention, the monitoring area is a circular area centered on a monitoring point on the hydraulic engineering, and the radius of the circular area is an artificially set parameter.
In one embodiment of the invention, a hydraulic engineering deformation monitoring method based on satellite data is provided, and the method further comprises the step of early warning based on the deformation amount of a monitoring area, wherein a threshold value of a set variable is manually set, and if the deformation amount of the monitoring area is larger than the threshold value, early warning is sent to hydraulic engineering management staff so that the management staff can find the danger of the hydraulic engineering in time.
As shown in fig. 2, in one embodiment of the present invention, a hydraulic engineering deformation monitoring system based on satellite data is provided, including:
an image acquisition module 201, configured to extract SAR images of the current time and N previous target areas;
an image registration module 202 for selecting one from the N SAR images as a main image and registering the other SAR images to a main image space;
the image structure data generating module 203 is configured to perform image structuring on the registered SAR image to obtain image structure data, where nodes of the image structure data are mapped with pixels of the SAR image one by one, and if pixels mapped by two nodes are adjacent in the SAR image, an edge exists between the two nodes; each node has an initial feature vector, one component of which corresponds to the value of one dimension of the pixel;
a deformation reference generation module 204, configured to input the map structure data and the registered SAR image into an image processing model, and output a deformation reference pixel matrix;
the deformation amount calculation module 205 generates a first elevation map and a second elevation map based on the SAR image captured at the current time and the SAR image captured at the deformation reference time point, performs difference between the first elevation map and the second elevation map to obtain an elevation difference map, performs dot multiplication on the elevation difference map and the deformation reference pixel matrix to obtain a deformation amount map, and then calculates the average value of pixel values in the monitoring area in the deformation amount map as the deformation amount of the monitoring area.
The present invention provides a storage medium storing non-transitory computer readable instructions that, when executed by a computer, are capable of performing the steps of a satellite data based hydraulic engineering deformation monitoring method as described above.
The embodiment has been described above with reference to the embodiment, but the embodiment is not limited to the above-described specific implementation, which is only illustrative and not restrictive, and many forms can be made by those of ordinary skill in the art, given the benefit of this disclosure, are within the scope of this embodiment.
Claims (9)
1. The hydraulic engineering deformation monitoring method based on satellite data is characterized by comprising the following steps of:
step 101, extracting SAR images of N previous target areas and the current time;
step 102, selecting one from N SAR images as a main image, and registering other SAR images to a main image space;
step 103, carrying out graph structuring on the registered SAR image to obtain graph structure data, wherein nodes of the graph structure data are mapped with pixels of the SAR image one by one, and if the pixels mapped by the two nodes are adjacent in the SAR image, edges exist between the two nodes; each node has an initial feature vector, one component of which corresponds to the value of one dimension of the pixel;
step 104, inputting the graph structure data and the registered SAR image into an image processing model, wherein the image processing model comprises: the system comprises a first convolution module, a stability pixel extraction module and a first logic layer, wherein the first convolution module is used for respectively inputting N SAR images and outputting N first area diagrams, and the size of the first area diagrams is consistent with that of the SAR images; the SAR image is input into a stability pixel extraction module, and a stability pixel matrix is output; the first logic layer inputs the first region map and the stability pixel matrix, and outputs a deformation reference pixel matrix;
the calculation formula of the stability pixel extraction module is as follows:
;
;
;
;
;
;
;
;
;
;
;
wherein the method comprises the steps ofAnd->The input feature matrix input by the first time step and the t time step are respectively represented, and the ith row vector of the input feature matrix of the t time step is the initial feature vector of the ith node of the t-th graph structure data; />Representing the sum of the adjacency matrix and the identity matrix of the diagram structure data, +.>Representation->Degree matrix of->Representing an initial feature matrix>、、/>Output feature matrices respectively representing the t-th, t-1 th and N-th time steps; />A first gating matrix representing the t-th time step,>a second gating matrix representing the t-th time step,>represents dot product->、/>、Representing a first coding feature matrix, a second coding feature matrix and a third coding feature matrix, respectively,/->、、/>、/>、/>Respectively representing a first weight parameter, a second weight parameter, a third weight parameter, a fourth weight parameter and a fifth weight parameter, +.>、/>、/>Respectively representing a first bias parameter, a second bias parameter and a third bias parameter; />Representing a stability pel vector>Indicating that a value of more than 0.5 is changed to 1, a value of not more than 0.5 is changed to 0,/and a value of not more than 0.5 is changed to 0->Representing hyperbolic tangent function, ">Representing an activation function->Representation->E represents a matrix of stable pixels;
step 105, generating a first elevation map and a second elevation map based on the SAR image shot at the current time and the SAR image shot at the deformation reference time point, obtaining an elevation map by differentiating the first elevation map and the second elevation map, obtaining a deformation map by dot multiplying the elevation map and the deformation reference pixel matrix, and then calculating the average value of pixel values in a monitoring area in the deformation map as the deformation of the monitoring area.
2. The method of claim 1, wherein the target area comprises hydraulic engineering to be monitored for deformation.
3. The method for monitoring deformation of hydraulic engineering based on satellite data according to claim 1, wherein the SAR images of the N target areas include SAR images photographed at deformation reference time points.
4. The method for monitoring the deformation of hydraulic engineering based on satellite data according to claim 1, wherein each SAR image map is structured into a map structure data.
5. The hydraulic engineering deformation monitoring method based on satellite data according to claim 1, wherein the first convolution module comprises a convolution layer and an up-sampling layer, the convolution layer inputs SAR images, the convolution layer outputs feature images, and the up-sampling layer up-samples the feature images to obtain a first region image.
6. The hydraulic engineering deformation monitoring method based on satellite data according to claim 1, wherein the calculation formula of the first logic layer is as follows:
;
wherein B represents a deformed reference pixel matrix,indicating that a value greater than 0.5 is changed to 1 and a value not greater than 0.5 is changed to 0, E indicates a stability matrix of picture elements, < >>Representing the v first region map, < >>Represents dot product->Representing a hyperbolic tangent function.
7. The method for monitoring the deformation of hydraulic engineering based on satellite data according to claim 1, wherein the monitoring area is a circular area centered on the monitoring point, and the radius of the circular area is a parameter set by people.
8. Hydraulic engineering deformation monitoring system based on satellite data, characterized by comprising:
an image acquisition module for extracting SAR images of the current time and the previous N target areas;
an image registration module for selecting one from the N SAR images as a main image and registering the other SAR images to a main image space;
the image structure data generation module is used for carrying out image structuring on the registered SAR image to obtain image structure data, nodes of the image structure data are mapped with pixels of the SAR image one by one, and if the pixels mapped by the two nodes are adjacent in the SAR image, edges exist between the two nodes; each node has an initial feature vector, one component of which corresponds to the value of one dimension of the pixel;
the deformation reference generation module is used for inputting the image structure data and the registered SAR image into the image processing model and outputting a deformation reference pixel matrix;
the deformation amount calculation module generates a first elevation map and a second elevation map based on the SAR image shot at the current time and the SAR image shot at the deformation reference time point, obtains the elevation map by differentiating the first elevation map and the second elevation map, obtains the deformation amount map by performing point multiplication on the elevation map and the deformation reference pixel matrix, and then calculates the average value of pixel values in a monitoring area in the deformation amount map as the deformation amount of the monitoring area.
9. A storage medium storing non-transitory computer readable instructions which, when executed by a computer, are capable of performing the steps of a satellite data based hydraulic engineering deformation monitoring method according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311549722.8A CN117274342B (en) | 2023-11-21 | 2023-11-21 | Hydraulic engineering deformation monitoring method based on satellite data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311549722.8A CN117274342B (en) | 2023-11-21 | 2023-11-21 | Hydraulic engineering deformation monitoring method based on satellite data |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117274342A true CN117274342A (en) | 2023-12-22 |
CN117274342B CN117274342B (en) | 2024-02-13 |
Family
ID=89202959
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311549722.8A Active CN117274342B (en) | 2023-11-21 | 2023-11-21 | Hydraulic engineering deformation monitoring method based on satellite data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117274342B (en) |
Citations (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110163911A1 (en) * | 2008-07-04 | 2011-07-07 | Telespazio S.P.A. | Identification and Analysis of Persistent Scatterers In Series of SAR Images |
CN103675790A (en) * | 2013-12-23 | 2014-03-26 | 中国国土资源航空物探遥感中心 | Method for improving earth surface shape change monitoring precision of InSAR (Interferometric Synthetic Aperture Radar) technology based on high-precision DEM (Digital Elevation Model) |
US8774470B1 (en) * | 2012-04-03 | 2014-07-08 | Google Inc. | Processing a satellite image of a region to determine a terrain boundary included therein |
CN109212528A (en) * | 2018-10-30 | 2019-01-15 | 中国科学院遥感与数字地球研究所 | A kind of multidimensional chromatography historic site SAR ruins deformation monitoring method |
CN111474544A (en) * | 2020-03-04 | 2020-07-31 | 广东明源勘测设计有限公司 | Landslide deformation monitoring and early warning method based on SAR data |
US20210132214A1 (en) * | 2018-05-08 | 2021-05-06 | Nec Corporation | Synthetic aperture radar image analysis system, synthetic aperture radar image analysis method, and synthetic aperture radar image analysis program |
CN113064188A (en) * | 2020-08-07 | 2021-07-02 | 国网浙江省电力有限公司 | Transformer substation geological deformation monitoring method based on SAR satellite and Beidou satellite |
US11238307B1 (en) * | 2020-09-24 | 2022-02-01 | Eagle Technology, Llc | System for performing change detection within a 3D geospatial model based upon semantic change detection using deep learning and related methods |
CN114067152A (en) * | 2022-01-14 | 2022-02-18 | 南湖实验室 | Refined flood inundated area extraction method based on satellite-borne SAR image |
CN114966685A (en) * | 2022-05-24 | 2022-08-30 | 中国水利水电科学研究院 | Dam deformation monitoring and predicting method based on InSAR and deep learning |
WO2022214114A2 (en) * | 2021-08-10 | 2022-10-13 | 中咨数据有限公司 | Bridge deformation monitoring method fusing gnss data and insar technology |
CN115184935A (en) * | 2022-07-18 | 2022-10-14 | 云南电网有限责任公司电力科学研究院 | Method, device, equipment and medium for monitoring deformation of power transmission corridor |
EP4113158A1 (en) * | 2021-06-28 | 2023-01-04 | Planetek Italia S.r.l. | System for determining the priority of inspection of satellite remotely sensed objects and method thereof |
US20230065519A1 (en) * | 2021-08-27 | 2023-03-02 | Si Analytics Co., Ltd | Method for scheduling of shooting satellite images based on deep learning |
CN116106904A (en) * | 2023-02-22 | 2023-05-12 | 深圳大学 | Facility deformation monitoring method and facility deformation monitoring equipment for object MT-InSAR |
CN116299438A (en) * | 2023-01-13 | 2023-06-23 | 中国南方电网有限责任公司超高压输电公司昆明局 | Ground surface deformation monitoring method and related equipment based on interference radar |
CN116383992A (en) * | 2023-03-16 | 2023-07-04 | 安徽省交通规划设计研究总院股份有限公司 | InSAR-based highway goaf settlement prediction method through deep and loose soil layers |
US20230236311A1 (en) * | 2022-01-26 | 2023-07-27 | Sun Yat-Sen University | OPTIMIZATION METHOD AND APPARATUS FOR INTERFEROMETRIC SYNTHETIC APERTURE RADAR (InSAR) TIME-SERIES PHASE |
CN116778104A (en) * | 2023-08-16 | 2023-09-19 | 江西省国土资源测绘工程总院有限公司 | Mapping method and system for dynamic remote sensing monitoring |
-
2023
- 2023-11-21 CN CN202311549722.8A patent/CN117274342B/en active Active
Patent Citations (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110163911A1 (en) * | 2008-07-04 | 2011-07-07 | Telespazio S.P.A. | Identification and Analysis of Persistent Scatterers In Series of SAR Images |
US8774470B1 (en) * | 2012-04-03 | 2014-07-08 | Google Inc. | Processing a satellite image of a region to determine a terrain boundary included therein |
CN103675790A (en) * | 2013-12-23 | 2014-03-26 | 中国国土资源航空物探遥感中心 | Method for improving earth surface shape change monitoring precision of InSAR (Interferometric Synthetic Aperture Radar) technology based on high-precision DEM (Digital Elevation Model) |
US20210132214A1 (en) * | 2018-05-08 | 2021-05-06 | Nec Corporation | Synthetic aperture radar image analysis system, synthetic aperture radar image analysis method, and synthetic aperture radar image analysis program |
CN109212528A (en) * | 2018-10-30 | 2019-01-15 | 中国科学院遥感与数字地球研究所 | A kind of multidimensional chromatography historic site SAR ruins deformation monitoring method |
CN111474544A (en) * | 2020-03-04 | 2020-07-31 | 广东明源勘测设计有限公司 | Landslide deformation monitoring and early warning method based on SAR data |
CN113064188A (en) * | 2020-08-07 | 2021-07-02 | 国网浙江省电力有限公司 | Transformer substation geological deformation monitoring method based on SAR satellite and Beidou satellite |
US11238307B1 (en) * | 2020-09-24 | 2022-02-01 | Eagle Technology, Llc | System for performing change detection within a 3D geospatial model based upon semantic change detection using deep learning and related methods |
EP4113158A1 (en) * | 2021-06-28 | 2023-01-04 | Planetek Italia S.r.l. | System for determining the priority of inspection of satellite remotely sensed objects and method thereof |
WO2022214114A2 (en) * | 2021-08-10 | 2022-10-13 | 中咨数据有限公司 | Bridge deformation monitoring method fusing gnss data and insar technology |
US20230065519A1 (en) * | 2021-08-27 | 2023-03-02 | Si Analytics Co., Ltd | Method for scheduling of shooting satellite images based on deep learning |
CN114067152A (en) * | 2022-01-14 | 2022-02-18 | 南湖实验室 | Refined flood inundated area extraction method based on satellite-borne SAR image |
US20230236311A1 (en) * | 2022-01-26 | 2023-07-27 | Sun Yat-Sen University | OPTIMIZATION METHOD AND APPARATUS FOR INTERFEROMETRIC SYNTHETIC APERTURE RADAR (InSAR) TIME-SERIES PHASE |
CN114966685A (en) * | 2022-05-24 | 2022-08-30 | 中国水利水电科学研究院 | Dam deformation monitoring and predicting method based on InSAR and deep learning |
CN115184935A (en) * | 2022-07-18 | 2022-10-14 | 云南电网有限责任公司电力科学研究院 | Method, device, equipment and medium for monitoring deformation of power transmission corridor |
CN116299438A (en) * | 2023-01-13 | 2023-06-23 | 中国南方电网有限责任公司超高压输电公司昆明局 | Ground surface deformation monitoring method and related equipment based on interference radar |
CN116106904A (en) * | 2023-02-22 | 2023-05-12 | 深圳大学 | Facility deformation monitoring method and facility deformation monitoring equipment for object MT-InSAR |
CN116383992A (en) * | 2023-03-16 | 2023-07-04 | 安徽省交通规划设计研究总院股份有限公司 | InSAR-based highway goaf settlement prediction method through deep and loose soil layers |
CN116778104A (en) * | 2023-08-16 | 2023-09-19 | 江西省国土资源测绘工程总院有限公司 | Mapping method and system for dynamic remote sensing monitoring |
Non-Patent Citations (7)
Title |
---|
SUN, J等: "Automatic Detection of Volcanic Surface Deformation Using Deep Learning", JOURNAL OF GEOPHYSICAL RESEARCH -SOLID EARTH * |
朱洪波: "基于时序InSAR技术的输电走廊地形形变提取方法研究", 中国优秀硕士学位论文全文数据库 信息科技辑 * |
毛亚纯;曹旺;赵占国;徐茂林;: "基于多元回归模型的GB-SAR监测误差改正及形变分析", 东北大学学报(自然科学版), no. 01 * |
焦明连;蒋廷臣;: "基于InSAR技术矿区地表形变的监测", 淮海工学院学报(自然科学版), no. 02 * |
王成;方璐;: "基于COSMO-SkyMed雷达影像的山地城市地表形变监测研究", 电子测量技术, no. 09 * |
苗放;梁军;叶成名;杨智翔;: "用InSAR技术提取数字高程模型的研究", 物探化探计算技术, no. 02 * |
葛大庆;王艳;郭小方;刘圣伟;范景辉;: "基于相干点目标的多基线D-InSAR技术与地表形变监测", 遥感学报, no. 04 * |
Also Published As
Publication number | Publication date |
---|---|
CN117274342B (en) | 2024-02-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Li et al. | A SAR-to-optical image translation method based on conditional generation adversarial network (cGAN) | |
US7983474B2 (en) | Geospatial modeling system and related method using multiple sources of geographic information | |
CN108596108B (en) | Aerial remote sensing image change detection method based on triple semantic relation learning | |
CN111340844B (en) | Multi-scale characteristic optical flow learning calculation method based on self-attention mechanism | |
Lizarazo et al. | Automatic mapping of land surface elevation changes from UAV-based imagery | |
CN109767454B (en) | Unmanned aerial vehicle aerial video moving target detection method based on time-space-frequency significance | |
CN113111758B (en) | SAR image ship target recognition method based on impulse neural network | |
CN111027497B (en) | Weak and small target rapid detection method based on high-resolution optical remote sensing image | |
CN104715474B (en) | High resolution synthetic aperture radar linearity building object detecting method based on Based On Method of Labeling Watershed Algorithm | |
CN112016478B (en) | Complex scene recognition method and system based on multispectral image fusion | |
Zhu et al. | Robust registration of aerial images and LiDAR data using spatial constraints and Gabor structural features | |
Sariturk et al. | Feature extraction from satellite images using segnet and fully convolutional networks (FCN) | |
CN112785499A (en) | Super-resolution reconstruction model training method and computer equipment | |
CN113822352A (en) | Infrared dim target detection method based on multi-feature fusion | |
Xiang et al. | Denseppmunet-a: A robust deep learning network for segmenting water bodies from aerial images | |
CN117788296B (en) | Infrared remote sensing image super-resolution reconstruction method based on heterogeneous combined depth network | |
Wang et al. | A deep deformable residual learning network for SAR image segmentation | |
Parmehr et al. | Automatic registration of optical imagery with 3d lidar data using local combined mutual information | |
Lu et al. | Multi-scale enhanced deep network for road detection | |
CN117274342B (en) | Hydraulic engineering deformation monitoring method based on satellite data | |
CN116958434A (en) | Multi-view three-dimensional reconstruction method, measurement method and system | |
Bull et al. | Perceptual improvements for super-resolution of satellite imagery | |
CN116843906A (en) | Target multi-angle intrinsic feature mining method based on Laplace feature mapping | |
Shi et al. | Land Cover Semantic Segmentation of High-Resolution Gaofen-3 SAR Image | |
Koo et al. | A bayesian based unrolling approach to single-photon lidar imaging through obscurants |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |