CN116148855A - Method and system for removing atmospheric phase and resolving deformation of time sequence InSAR - Google Patents
Method and system for removing atmospheric phase and resolving deformation of time sequence InSAR Download PDFInfo
- Publication number
- CN116148855A CN116148855A CN202310349555.6A CN202310349555A CN116148855A CN 116148855 A CN116148855 A CN 116148855A CN 202310349555 A CN202310349555 A CN 202310349555A CN 116148855 A CN116148855 A CN 116148855A
- Authority
- CN
- China
- Prior art keywords
- phase
- differential
- atmospheric
- deformation
- insar
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
- G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
- G01S13/9021—SAR image post-processing techniques
- G01S13/9023—SAR image post-processing techniques combined with interferometric techniques
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B7/00—Measuring arrangements characterised by the use of electric or magnetic techniques
- G01B7/16—Measuring arrangements characterised by the use of electric or magnetic techniques for measuring the deformation in a solid, e.g. by resistance strain gauge
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Computer Networks & Wireless Communication (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method and a system for removing and resolving deformation of a time sequence InSAR atmospheric phase, wherein the method firstly acquires a time sequence SAR image and DEM data of a monitoring area, and carries out preprocessing, differential interference, filtering and phase unwrapping; secondly, constructing a sample library containing differential interferograms of atmospheric phases; generating an antagonistic neural network CGAN based on the conditions, amplifying the sample library, and constructing a complete version of sample library; then constructing an atmosphere phase removal transune network model based on a transune network, and training and testing to remove the atmosphere phase in the differential interferogram; and finally, performing time sequence InSAR deformation calculation based on the differential interference pattern with the atmospheric phase removed so as to acquire the earth surface deformation information of the monitoring area. The invention can break through the technical bottleneck that the atmospheric phase error cannot be completely eliminated in the existing InSAR technology, and simultaneously improves the accuracy of time sequence InSAR deformation calculation.
Description
Technical Field
The invention relates to the technical field of InSAR deformation monitoring, in particular to a method and a system for removing atmospheric phase and resolving deformation of a time sequence InSAR.
Background
Synthetic aperture radar interferometry (InSAR) is widely used for deformation monitoring of the ground. The presence of atmospheric phase errors in the InSAR measurements makes the InSAR technique challenging in estimating accurate surface deformation. The measurement accuracy of a single interferogram is susceptible to atmospheric effects when radar signals propagate in the laminar turbulent atmosphere and ionosphere. Ionospheric effects are due to variations in electron density in the ionosphere, which can lead to phase distortion of radar signals, which can have a significant impact on larger wavelength data such as P-band and L-band SAR data. Tropospheric delays are caused by temperature, pressure and humidity variations and can be divided into turbulence component delays and stratification delays. Turbulence component delays are due to variations in the water vapor distribution in the troposphere over short time resolutions and spatial scales of several kilometers, which are often difficult to model and remove. The layering delay of the atmosphere is terrain dependent and exhibits spatial dependence over a length scale of tens of kilometers. In this context, both 20% relative humidity spatial or temporal variations can introduce atmospheric errors on the order of tens of centimeters to the InSAR interferogram, resulting in misinterpretation of the interference phases in the interferogram and inaccurate extraction of deformations.
A number of methods have been developed in the art of InSAR technology to estimate and eliminate the effect of atmospheric errors on the InSAR interferograms, which are mainly divided into three categories: the first class is based on empirical model methods, i.e. using a high Cheng Xianxing model or a power law model to remove the atmospheric layering delay associated with the terrain, the second class is based on methods to generate an atmospheric delay map based on external data, including meteorological data (ERA 5/WRF/EWMC), global satellite navigation system (GNSS) data, spectral data (MODIS/MERIS/sentel-3), and the third class is phase-based methods, such as linear elevation correction and time series InSAR methods. The invention patent with publication number of CN114624708A provides a ground-based radar atmospheric correction method and system under a complex environment, wherein the method acquires an atmospheric position estimation result by extracting permanent scatterer points and carrying out regional division, triangulation and atmospheric position interpolation on the permanent scatterer points, and removes the permanent scatterer points in accumulated phases to realize atmospheric correction. Deep learning in recent years has shown application potential in the aspects of phase unwrapping, atmospheric phase estimation, DEM super-resolution reconstruction, coherent estimation and phase filtering, deformation detection, prediction and the like in InSAR technology. The invention patent with publication number of CN114280608A provides a DINSAR elevation-related atmospheric effect removal method and system, wherein the method adopts an MLP neural network model to simulate an elevation-related atmospheric phase; the unwrapping phase is subtracted from the simulated elevation dependent atmospheric phase to complete the atmospheric effect removal. However, this method only removes elevation dependent atmospheric stratification delays, which cannot be effectively removed for turbulent atmospheric delays.
The following problems were found by summarizing the above methods: (1) The empirical model may remove terrain-related atmospheric stratification delays, but not atmospheric turbulence components; (2) The method based on external data depends on the space-time resolution of the external data, for example, the method based on GNSS data depends on the density of a GNSS network, local atmospheric turbulence change of a frozen soil area is difficult to capture, the space-time resolution of MODIS/MERIS/Sentinel-3 images is low, cloud conditions cannot be applied, the data acquisition time is inconsistent with SAR data acquisition time, and the local turbulence and non-uniform atmospheric components are ignored based on a combined data (GACOS) method; (3) The time domain filtering algorithm depends on parameter setting, is easily subjected to seasonal change of InSAR time sequence deformation, and cannot remove atmospheric turbulence components. In summary, the existing method for removing the atmospheric phase error in the time sequence InSAR technology depends on external data and algorithm model performance, and cannot effectively remove partial turbulent atmospheric errors.
In order to more effectively restrain and remove the influence of atmospheric errors in the time sequence InSAR technology, a method and a system for time sequence InSAR atmospheric phase removal and deformation calculation are developed, a method for generating InSAR atmospheric phase sample augmentation of an anti-neural network CGAN based on the condition of a self-attention mechanism is designed, and a network framework for atmospheric phase removal based on the time sequence InSAR technology is constructed by introducing a Transunet network.
Disclosure of Invention
The invention aims at overcoming the defects of the prior art and provides a method and a system for removing the atmospheric phase and resolving the deformation of a time sequence InSAR. The invention can effectively separate the time sequence InSAR atmospheric delay error and deformation, and improve the accuracy of InSAR deformation calculation.
The aim of the invention is realized by the following technical scheme: the first aspect of the embodiment of the invention provides a method for removing atmospheric phase and resolving deformation of a time sequence InSAR, which comprises the following steps:
(1) Acquiring time sequence SAR image data and digital elevation model data of a monitoring area, and performing preprocessing, differential interference, filtering and phase unwrapping on the data;
(2) Constructing a first sample library containing differential interferograms of atmospheric phases according to thresholds of the time base line and the space base line;
(3) Performing augmentation and construction on a first sample library containing atmospheric phases based on a conditional generation antagonistic neural network CGAN to obtain a complete version sample library of the differential interferogram;
(4) Constructing an atmosphere phase removal transune network model based on a transune network, and training and testing the model to remove the atmosphere phases in all differential interferograms;
(5) And performing time sequence InSAR deformation calculation based on the differential interference pattern with the atmospheric phase removed so as to acquire the earth surface deformation information of the monitoring area.
Optionally, the step (1) includes the sub-steps of:
(1.1) acquiring time sequence SAR image data of a monitoring area, and extracting digital elevation model data of the monitoring area according to the longitude and latitude range of the monitoring area;
(1.2) preprocessing data: firstly, importing an original time sequence SAR image data file, performing format conversion to generate a single-view complex data set, then updating track parameters by combining a downloaded fine track data file, and finally performing geometric positioning registration by using digital elevation model data and the fine track data file to obtain a registered time sequence SAR image;
(1.3) differential interference: firstly, carrying out interference processing on registered time series SAR images, and carrying out phase difference on a main image and an auxiliary image to obtain an interference image; then calculating the terrain phase and the land phase of each interferogram according to an inverse distance weight interpolation algorithm by using digital elevation model data, and subtracting the two phases to obtain a differential interferogram;
(1.4) selecting nonlinear self-adaptive Goldstein spatial filtering to filter the differential interference pattern to obtain a filtered differential interference pattern for suppressing noise influence in the differential interference pattern;
(1.5) phase unwrapping: and carrying out phase unwrapping treatment on the filtered differential interferogram by adopting a minimum cost flow method so as to obtain the differential interferogram after phase unwrapping.
Optionally, the step (2) includes the sub-steps of:
(2.1) selecting a differential interference pattern from the phase unwrapped differential interference patterns according to the threshold values of the time base line and the space base line to generate a differential interference pattern set
And from the differential interference atlas->
Selecting differential interference atlas containing atmospheric phase +.>
；/>
(2.2) pair differential interference atlas
Performing image sample and label making, performing image segmentation processing on each differential interferogram to generate an image sample, andnormalizing the generated image samples to construct a first sample library containing differential interferograms of atmospheric phase and denoted +.>
。
Optionally, the step (3) comprises the following sub-steps:
(3.1) first sample library based on differential interferograms containing atmospheric phase
Training and testing the condition generating antagonistic neural network CGAN to generate an additional second sample library containing atmospheric phase +.>
；
(3.2) storing the second sample library
Is +.>
Combining to generate a third sample library I containing differential interferograms of atmospheric phases; selecting a differential interference pattern which is not affected by the atmospheric phase according to the differential interference pattern after phase unwrapping, and carrying out image segmentation and data set manufacturing to generate a fourth sample library P of the differential interference pattern which does not contain the atmospheric phase; constructing a complete version sample library of the differential interferogram according to the third sample library I and the fourth sample library P and marking the complete version sample library as +.>
。
Optionally, the condition generating antagonistic neural network CGAN comprises a generator and a arbiter.
Optionally, the full version sample library includes an InSAR differential interferogram sample data set which corresponds to the fourth sample library and does not contain an atmospheric phase, and the network corresponding to the second sample library generates the InSAR differential interferogram sample data set which contains the atmospheric phase and the InSAR differential interferogram sample data set which corresponds to the first sample library and contains a true atmospheric phase.
Optionally, the step (4) includes the sub-steps of:
(4.1) constructing an atmosphere phase removal transune network model based on a transune network to reconstruct and output a differential interference diagram after the atmosphere phase removal;
(4.2) library of full version samples
As an input data set of the atmosphere bit removal transune network model, dividing the input data set into a training data set and a test data set according to the proportion input by a user, loading the training data set into the atmosphere bit removal transune network model for training to obtain trained weight parameter information, and loading the trained weight parameter information by using the test data set to obtain a differential interferogram sample with the atmosphere phase removed;
(4.3) performing image slicing and merging on the differential interference pattern samples with the atmospheric phases removed to generate a differential interference pattern with the atmospheric phases removed.
Optionally, the transune network combines two network structures of UNet and transformers, and a U-shaped structure is formed by an encoder and a decoder.
Optionally, the step (5) specifically includes: and (3) performing time sequence InSAR deformation calculation based on the differential interferogram with the atmospheric phase removed, calculating time sequence deformation and accumulated deformation of the monitoring region by adopting a small baseline subset method, and using weighted least square estimation to realize stable calculation of the time sequence deformation so as to acquire the earth surface deformation information of the monitoring region.
The second aspect of the embodiment of the invention provides a system for removing and resolving deformation of a time sequence InSAR atmosphere phase, which is used for realizing the method for removing and resolving deformation of the time sequence InSAR atmosphere phase, and comprises the following steps:
the data preprocessing module is used for carrying out data importing and image registering on the acquired time sequence SAR images;
the data differential interference and filtering and phase unwrapping processing module is used for carrying out differential interference, nonlinear self-adaptive Goldstein spatial filtering and minimum cost flow phase unwrapping processes according to the registered time sequence SAR images;
the differential interferogram sample library construction and amplification module is used for generating samples of differential interferograms containing atmospheric phases, generating amplification of atmospheric phase samples of the anti-neural network CGAN based on conditions and constructing the differential interferogram sample library;
the atmosphere phase removal module is used for removing the atmosphere phase of the constructed differential interferogram sample library based on the TransUNet network; and
the time sequence InSAR deformation calculation module is used for carrying out time sequence InSAR deformation calculation on the generated differential interference diagram with the atmospheric phase removed, and obtaining the deformation information of the earth surface.
The method has the advantages that the method can break through the technical bottleneck that the atmospheric phase error cannot be completely eliminated in the existing InSAR technology, the influence of the atmosphere in the InSAR interferogram can be effectively reduced through the proposed transune network, the great potential of the atmospheric phase elimination method based on deep learning is shown, and the accuracy of time sequence InSAR deformation calculation is improved. The method can also be used in the application field of high-precision deformation extraction based on InSAR technology, is suitable for the deformation extraction application of natural earth surface, and is used for the fields of geographic national condition monitoring of ground subsidence in China, general survey of geological disasters and the like.
Drawings
FIG. 1 is a flow chart of a method for removing atmospheric bits and resolving deformations of a time sequence InSAR of the present invention;
FIG. 2 is a schematic diagram of a condition generation antagonistic neural network CGAN network architecture according to an embodiment of the invention;
FIG. 3 is a schematic illustration of a sample containing atmospheric phase in accordance with an embodiment of the present invention; wherein (a) in fig. 3 is a sample diagram containing an atmospheric phase in reality; fig. 3 (b) is a sample graph containing atmospheric phase generated by CGAN;
fig. 4 is a schematic diagram of a transune network-based atmospheric bit removal structure according to an embodiment of the present invention;
FIG. 5 is a differential interferogram contrast map of an embodiment of the present invention; wherein (a) in fig. 5 is a differential interferogram originally containing an atmospheric phase; FIG. 5 (b) is a differential interferogram after atmospheric phase removal based on ERA5 meteorological data; fig. 5 (c) is a differential interference diagram after the removal of the atmospheric phase based on the TransUNet network;
FIG. 6 is a linear deformation rate comparison plot of a time series InSAR deformation solution of an embodiment of the present invention; wherein (a) in fig. 6 is a linear deformation rate of the SBAS method for removing the atmospheric phase based on time-space domain filtering; FIG. 6 (b) is a linear deformation rate of the SBAS method based on ERA5 weather data to remove atmospheric phase; fig. 6 (c) is a linear deformation rate of SBAS method based on TransUNet network to remove atmospheric phase;
FIG. 7 is a schematic diagram of the system for time-series InSAR atmospheric phase removal and deformation resolution of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to fall within the scope of the invention.
The method for removing the atmospheric phase and resolving the deformation of the time sequence InSAR disclosed by the invention, as shown in figure 1, specifically comprises the following steps of:
(1) Time series SAR (synthetic aperture radar ) image data and DEM (digital elevation model, digital Elevation Model) data of the monitoring area are acquired, and preprocessing, differential interference, filtering and phase unwrapping processing are performed on the data.
And (1.1) acquiring time sequence SAR image data of the monitoring area, and extracting DEM data of the monitoring area according to the longitude and latitude range of the monitoring area.
Specifically, the monitoring area in this embodiment is located near a ZUZUZU lake in the middle of the Tibetan autonomous region, a time sequence SAR image dataset of the monitoring area is firstly obtained, the time sequence SAR image dataset in this embodiment is Sentinel-1A data of TOPS mode of open source of European space, and IW mode VV polarized images with track number of 150 (derailment) and Frame number of 490 are selected to perform time sequence InSAR processing, and the time coverage range is 2017, 03, 16, 2020, 03, 24. The specific parameters of the data are as follows: 120km breadth, derailment mode, range and azimuth resolution are about 2.3m and 13.9m. Meanwhile, SRTM DEM data with resolution of 30 meters are selected according to the longitude and latitude range 89.807 DEG E-92.785 DEG E and 29.654 DEG N-31.776 DEG N of SAR images, and all the DEM data are spliced by adopting ENVI5.3 software to form DEM data of a monitoring area.
(1.2) preprocessing data: firstly, importing an original time sequence SAR image data file, performing format conversion to generate a single-view complex data set, then, updating track parameters by combining a downloaded fine track data file, and finally, performing geometric positioning registration by using DEM data and the fine track data file to obtain a registered time sequence SAR image.
Specifically, preprocessing an acquired time sequence SAR image dataset of a monitoring area, importing a tiff file of original time sequence SAR data, performing format conversion to generate single-view complex data, and simultaneously, combining a downloaded fine track data file to update track parameters for subsequent DEM registration. And then carrying out geometric positioning registration on the SAR image data by using the DEM data and the fine rail data file, and carrying out Enhanced Spectrum Diversity (ESD) registration to resample all auxiliary images to the frame of the main image so as to obtain a registered time sequence SAR image in order to ensure that the accuracy of the alignment of the Sentinel-1 data in the azimuth direction is one thousandth.
(1.3) differential interference: firstly, carrying out interference processing on registered time series SAR images, and carrying out phase difference on a main image and an auxiliary image to obtain an interference image; and then, calculating the terrain phase and the land phase of each interferogram according to an inverse distance weight interpolation algorithm by using DEM data, and subtracting the two phases to obtain a differential interferogram.
It should be understood that in this embodiment, the open source software GMTSAR is used to perform differential interference processing on the registered time-series SAR images.
(1.4) selecting nonlinear self-adaptive Goldstein spatial filtering to filter the differential interference pattern to restrain noise influence in the differential interference pattern, so as to obtain the filtered differential interference pattern.
(1.5) phase unwrapping: and carrying out phase unwrapping treatment on the filtered differential interferogram by adopting a minimum cost flow method so as to obtain the differential interferogram after phase unwrapping.
Specifically, the filtered differential interferograms may be phase unwrapped using a minimum cost flow algorithm. The distance difference of microwaves in the two imaging processes can be obtained according to the phase value, so that the topography, the landform and the tiny change of the surface of a target area can be calculated, the method can be used for digital elevation model establishment, crust deformation detection and the like, and topography elevation data and the like can be obtained from interference fringes.
It should be appreciated that after preprocessing, differential interference, filtering and phase unwrapping of the time series SAR images, a series of time series differential interferograms are formed, so that a phase unwrapped differential interferogram dataset may be ultimately obtained.
(2) And constructing a first sample library containing differential interferograms of the atmospheric phases according to the thresholds of the time base line and the space base line.
(2.1) selecting a differential interference pattern from the phase unwrapped differential interference patterns according to the threshold values of the time base line and the space base line to generate a differential interference pattern set
And from the differential interference atlas->
Selecting differential interference atlas containing atmospheric phase +.>
。
Specifically, the differential interferograms are selected according to the multiple differential interferograms finally generated in the step (1.5), in this embodiment, a small baseline threshold method is adopted for selection, namely, the time baseline is selected to be 50 days, and the space baseline is selected to be 100 meters, so as to preliminarily determine the differential interferogram set
Then from the differential interference atlas +.>
Selecting differential interference atlas containing atmospheric phase by artificial visual interpretation method>
。
(2.2) constructing a sample library containing differential interferograms of atmospheric phase: for differential interference atlas
For image sample and label making, because the original differential interferograms have larger widths, the training samples of the neural network are formed by image blocking processing, that is, the image segmentation processing is performed on each differential interferogram, for example, the training samples in the embodiment have the size of 64×64, the generated image samples are normalized, and finally, a first sample library containing differential interferograms with atmospheric phases is constructed and marked as->
An input dataset against the neural network CGAN is generated as a subsequent condition.
(3) And (3) augmenting and constructing a first sample library containing the atmospheric phase based on the condition generation antagonistic neural network CGAN so as to obtain a complete version sample library of the differential interferogram.
(3.1) first sample library based on differential interferograms containing atmospheric phase
Training and testing the condition generating antagonistic neural network CGAN to generate an additional second sample library containing atmospheric phase +.>
。
To generate more similar samples containing atmospheric phase to train the subsequent neural network, a conditional generation antagonistic neural network CGAN is used to extract the atmospheric phase features and generate samples containing differential interferograms of atmospheric phase.
In this embodiment, the network structure of the condition generation countermeasure neural network CGAN is as shown in FIG. 2, and a first sample library is used
Input into a conditional generation antagonistic neural network CGAN, firstly, the input of the generator is a random noise vector z obeying normal distribution and a characteristic vector containing atmospheric phase +.>
Outputting a pseudo-atmospheric phase sample generated +.>
. Then to the arbiter, the input of the arbiter is the atmospheric phase sample x from the real dataset and the pseudo-atmospheric phase sample generated +.>
The role of the discriminator is to discriminate whether the atmospheric phase sample is true or not, and output as the probability that the discriminator judges whether the atmospheric phase sample is true or not. The self-attention mechanism can obtain the global geometric feature of the image, reduce the dependence on external information and better capture the internal correlation of the atmospheric phase feature. Meanwhile, aiming at the problems of unstable training and easy disappearance/explosion of gradient existing in the GAN, a Wasserstein distance is introduced to replace the traditional JS divergence, and a loss function of the GAN is written as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,
is the distribution of the real atmospheric phase data blocks,/-, for example>
Is to generate pseudo-atmospheric phase samples (+.>
) Distribution of atmospheric phase data blocks, +.>
Is->
And->
A uniform sample distribution along a straight line between two points,
representing true atmospheric phase sample x obeys +.>
Distribution desire->
Representing the probability that the discriminator judges whether the atmospheric phase sample generated by the generator is authentic, < >>
Representation generator model generating an atmospheric phase sample +.>
Obeys->
The desire for a distribution is that,
representing the probability of the arbiter judging whether the real atmospheric phase sample is real, < >>
Representation generator model generating an atmospheric phase sample +.>
Obeys->
Distribution desire->
Representation->
Gradient operator->
Representing a 2-norm. The arbiter network and the generator network compete with each other during the training process so that the generator network can be as close as possible to a distribution that truly contains atmospheric error phase data.
A first sample library containing differential interferograms of atmospheric phase generated based on step (2.2)
The sample library->
Training and testing for conditional generation of antagonistic neural networks CGAN, first sample library at training time>
The training set and the test set are divided in an 8:2 ratio, and the PyTorch framework of a Langchao server (NF 5468M 6) is utilized for training and testing the condition generation antagonistic neural network CGAN, wherein the server runs 128GB of memory and has 8 Nvidia A40 GB of display cards. Iteratively updating neural network weights during training using Adam optimizer, with specific parameters being smoothing constant +.>
Smooth constant->
Learning rate->
. Finally generating an additional second sample bank containing atmospheric phases>
. Illustratively, the condition generation is utilized to combat neural networksThe sample containing the atmospheric phase generated by the CGAN is shown in (b) of fig. 3, and is compared with a real sample graph containing the atmospheric phase shown in (a) of fig. 3, wherein the abscissa represents pixels in a distance direction, the ordinate represents pixels in an azimuth direction, the number of the pixels is shown, and the comparison of the two graphs can show that the capability of modeling the sample containing the atmospheric phase of the CGAN by the anti-neural network can be generated, and the characteristic of the atmospheric phase on the differential interference graph can be better simulated.
(3.2) storing the second sample library
Is +.>
Combining to generate a third sample library I containing differential interferograms of atmospheric phases, selecting the differential interferograms which are not affected by the atmospheric phases according to the differential interferograms after phase unwrapping, performing image segmentation and data set manufacturing to generate a fourth sample library P without the differential interferograms of the atmospheric phases, constructing a complete version sample library of the differential interferograms according to the third sample library I and the fourth sample library P and marking as>
。
Specifically, the second sample library generated in the step (3.1) is used for
And (2.2) generating a first sample library +.>
The third sample library I, which is composed of differential interferograms with atmospheric phases, is combined to increase the diversity of training data of the subsequent atmospheric phase removal network. Then selecting a differential interference pattern which is not affected by the atmospheric phase according to the phase unwrapped differential interference pattern generated in the step (1.5), performing image segmentation and data set manufacturing to generate a fourth sample library P which does not contain the differential interference pattern of the atmospheric phase, and finally constructing a differential according to the third sample library I and the fourth sample library PA complete version of the sample library of interferograms and record it as +.>
And taking the data as an input data set for carrying out atmosphere bit removal on a subsequent TransUNet network.
It should be appreciated that the full version sample library
The method comprises the steps that an InSAR differential interferogram sample data set which corresponds to a fourth sample library and does not contain an atmospheric phase is generated through a network which corresponds to a second sample library, and the InSAR differential interferogram sample data set which contains the atmospheric phase and corresponds to a first sample library is generated.
(4) And constructing an atmosphere phase removal transune network model based on the transune network, and training and testing the model to remove the atmosphere phases in all the differential interferograms.
And (4.1) constructing an atmosphere bit removal transune network model based on the transune network so as to reconstruct and output a differential interference diagram after the atmosphere bit removal.
In this embodiment, the atmosphere phase removal structure based on the TransUNet network is shown in FIG. 4, and the input of the network is a complete version sample library
The main body of the network structure is a transune network, which fully combines two network structures of UNet and tranformators, and a U-shaped structure is formed by an encoder and a decoder. And the encoder structure of the Transformers is applied to the encoder structure, so that the multi-scale characteristics of the atmospheric phase can be learned from the differential interferogram data set. And finally, adding a single residual unit and a convolution layer for reconstructing and outputting the image with the atmospheric phase removed, wherein the residual unit is the difference value between the input differential interference pattern and the differential interference pattern with the atmospheric phase removed. An image containing an atmospheric phase error is input first, and if the image is a single channel, the channel is expanded into three channels by repeating the repeat function twice. Three-channel image is carried out through ResNetV2 network structureDownsampling, encoding the image into a high-level feature representation. A feature list is then created to save each of the downsampled feature maps. And finally, outputting through a ResNetV2 network structure. The feature list contains three size feature maps, respectively [ B, C,112]，[B,4C,56,56]，[B,8C,28,28]. Specific ResNetV2 network architecture: the convolution operation of the convolution kernel (7x7, s=2, p=3) is used for root downsampling to [ B, C,112]Downsampling by the maxpooling layer is then [ B, C,56]. And finally, the feature map is output through downsampling of three blocks. Then the downsampled feature map [ B,16C,14]The image was cut into patches using convolution kernels and added position embedding to get the output feature map. Then using the transform's Encoder, conv2d and decoder outputs as [ B,32,224,224]An image of a size. The decoder is used for doubling the size of the feature map through bilinear upsampling, then concatating the feature map of the shortcut after previous convolution, and finally mapping the feature map to a low-dimensional space through two Conv2 d. And dividing the characteristic diagram output by the network, adding a residual unit at the last layer of the transune network, namely, the difference value between the input interference diagram and the interference diagram without the atmospheric phase, and finally outputting the difference value as a sample for removing the atmospheric phase.
(4.2) library of full version samples
The method comprises the steps of dividing an input data set of an atmospheric phase removal transune network model into a training data set and a test data set according to the proportion input by a user, loading the training data set into the atmospheric phase removal transune network model for training to obtain trained weight parameter information, and loading the trained weight parameter information by using the test data set to obtain a differential interferogram sample with the atmospheric phase removed.
Specifically, the complete version sample library constructed in the step (3.2) is prepared
As the input data set of the atmosphere bit removal TransUNet network model, the input data set is divided into a training data set and a training data set according to the proportion of user inputThe test data set can be divided into a training data set and a test data set according to the proportion of 8:2, the training data set is loaded into an atmospheric bit removal transune network model for training, and a PyTorch framework of a Langchao server (NF 5468M 6) which runs 128GB and has 8 Nvidia A40 GB display cards can be used for training and testing the transune network. Iteratively updating neural network weights during training using Adam optimizer, with specific parameters being smoothing constant +.>
Smoothing constant
Learning rate->
. After training, trained weight parameter information can be obtained, and then the pre-trained weight parameter information is loaded by using a test data set, so that a differential interferogram sample with the atmospheric phase removed is obtained.
(4.3) performing image slicing and merging on the differential interference pattern samples with the atmospheric phases removed to generate a differential interference pattern with the atmospheric phases removed.
Specifically, image slicing and merging are carried out on the differential interferogram sample with the atmospheric phase removed, which is generated in the step (4.2), so that a differential interferogram with the atmospheric phase removed is generated, and in order to further compare the capacity of removing the atmospheric phase in the differential interferogram based on a TransUNet network, ERA5-inter re-analysis data of an ECMWF comprehensive prediction system model are downloaded to obtain weather data corresponding to a monitoring area, and the differential interferogram in a 20170917-201011 time period is subjected to atmospheric phase removal by using ERA5 external weather data. For example, a differential interferogram contrast chart is shown in fig. 5, wherein the abscissa represents pixels in a distance direction, the ordinate represents pixels in an azimuth direction, and the number of pixels is shown, and fig. 5 (a) is an original differential interferogram, which contains a serious atmospheric turbulence delay error and can seriously affect subsequent time sequence InSAR deformation calculation; FIG. 5 (b) is a differential interference diagram after the atmospheric phase is removed based on ERA5 weather data, which shows that partial atmospheric turbulence error phase can be removed based on ERA5 weather data, but not all of the atmospheric turbulence error phase can be removed, and the differential interference diagram still has serious atmospheric error (see the lower right corner of FIG. 5 (b)); in fig. 5, (c) is a differential interference diagram after the removal of the atmospheric phase based on the transune network, it can be seen that the atmospheric phase error of the differential interference diagram can be effectively removed by the atmospheric phase removal transune network model constructed based on the transune network, and the method is used for the subsequent time sequence InSAR deformation calculation, and the method also improves the precision of the subsequent time sequence InSAR deformation calculation.
(5) And performing time sequence InSAR deformation calculation based on the differential interference pattern with the atmospheric phase removed so as to acquire the earth surface deformation information of the monitoring area.
In this embodiment, the time sequence InSAR deformation calculation is performed based on the differential interferogram with the atmospheric phase removed, the time sequence deformation amount and the accumulated deformation amount of the monitoring area are calculated by adopting an SBAS (small baseline subset ) method, and the weighted least square estimation is used to realize the steady calculation of the time sequence deformation so as to obtain the earth surface deformation information of the monitoring area.
Specifically, performing time sequence InSAR deformation calculation according to the differential interferogram with the atmospheric phase removed in the step (4.3), calculating time sequence deformation and accumulated deformation of a monitoring area by adopting an SBAS method, realizing steady calculation of the time sequence deformation by using a solution algorithm through weighted least square estimation, and finally obtaining the earth surface deformation information of the monitoring area. In order to evaluate the effect of deformation calculation of the SBAS method for removing the atmospheric phase based on the TransUNet network, the SBAS method for removing the atmospheric phase based on the time-space domain filtering and the SBAS method for removing the atmospheric phase based on the ERA5 meteorological data are adopted for comparison analysis. As shown in fig. 6, which shows a linear deformation rate diagram of time-series InSAR deformation solution in this embodiment, wherein the abscissa represents pixels in the distance direction and the ordinate represents pixels in the azimuth direction, the number of pixels is shown, from which it can be found that the deformation result obtained by the SBAS method of removing the atmospheric phase based on the transune network is smoother, as shown in fig. 6 (c), and that the deformation result does not include local atmospheric disturbance delay affecting a specific area, the linear deformation rate diagram obtained by the SBAS method of removing the atmospheric phase based on the time-space domain filtering is shown in fig. 6 (a), and that no improvement is seen in the deformation result obtained by the SBAS method of removing the atmospheric phase based on ERA5 weather data, as shown in fig. 6 (b), which is related to the lower spatial resolution of ERA5 weather data.
It should be noted that the embodiment of the invention also provides a system for removing the atmospheric phase and resolving the deformation of the time sequence InSAR, which is used for realizing the method for removing the atmospheric phase and resolving the deformation of the time sequence InSAR.
In this embodiment, the system includes a data preprocessing module, a data differential interference and filtering and phase unwrapping processing module, a differential interferogram sample library constructing and amplifying module, an atmospheric phase removing module, and a time sequence InSAR deformation resolving module, as shown in fig. 7.
In this embodiment, the data preprocessing module is configured to perform data importing and image registering on the acquired time-series SAR image.
In this embodiment, the data differential interference and filtering and phase unwrapping processing module is configured to perform differential interference, nonlinear adaptive Goldstein spatial filtering, and minimum cost flow phase unwrapping according to the registered time-series SAR image. It should be noted that, the user may automatically configure different parameters of the flow algorithm according to the requirement.
In this embodiment, the construction and amplification module of the differential interferogram sample library is configured to generate a sample of a differential interferogram including an atmospheric phase, generate an amplification of an atmospheric phase sample against the neural network CGAN based on a condition, and construct the differential interferogram sample library. It should be noted that, the user may select the dividing ratio of the training data set and the test data set according to the requirement, and modify the network training parameters.
In this embodiment, the atmospheric phase removal module is configured to perform atmospheric phase removal on the constructed differential interferogram sample library based on the transune network. It should be noted that, the user may select the dividing ratio of the training data set and the test data set according to the requirement, and modify the network training parameters.
In this embodiment, the time sequence InSAR deformation calculation module is configured to perform time sequence InSAR deformation calculation on the generated differential interferogram from which the atmospheric phase is removed, and obtain deformation information of the earth surface.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for time-series InSAR atmospheric phase removal and deformation resolution, comprising the steps of:
(1) Acquiring time sequence SAR image data and digital elevation model data of a monitoring area, and performing preprocessing, differential interference, filtering and phase unwrapping on the data;
(2) Constructing a first sample library containing differential interferograms of atmospheric phases according to thresholds of the time base line and the space base line;
(3) Performing augmentation and construction on a first sample library containing atmospheric phases based on a conditional generation antagonistic neural network CGAN to obtain a complete version sample library of the differential interferogram;
(4) Constructing an atmosphere phase removal transune network model based on a transune network, and training and testing the model to remove the atmosphere phases in all differential interferograms;
(5) And performing time sequence InSAR deformation calculation based on the differential interference pattern with the atmospheric phase removed so as to acquire the earth surface deformation information of the monitoring area.
2. The method of time-series InSAR atmospheric phase removal and deformation resolution according to claim 1, wherein said step (1) comprises the sub-steps of:
(1.1) acquiring time sequence SAR image data of a monitoring area, and extracting digital elevation model data of the monitoring area according to the longitude and latitude range of the monitoring area;
(1.2) preprocessing data: firstly, importing an original time sequence SAR image data file, performing format conversion to generate a single-view complex data set, then updating track parameters by combining a downloaded fine track data file, and finally performing geometric positioning registration by using digital elevation model data and the fine track data file to obtain a registered time sequence SAR image;
(1.3) differential interference: firstly, carrying out interference processing on registered time series SAR images, and carrying out phase difference on a main image and an auxiliary image to obtain an interference image; then calculating the terrain phase and the land phase of each interferogram according to an inverse distance weight interpolation algorithm by using digital elevation model data, and subtracting the two phases to obtain a differential interferogram;
(1.4) selecting nonlinear self-adaptive Goldstein spatial filtering to filter the differential interference pattern to obtain a filtered differential interference pattern for suppressing noise influence in the differential interference pattern;
(1.5) phase unwrapping: and carrying out phase unwrapping treatment on the filtered differential interferogram by adopting a minimum cost flow method so as to obtain the differential interferogram after phase unwrapping.
3. The method of time-series InSAR atmospheric phase removal and deformation resolution according to claim 1, wherein said step (2) comprises the sub-steps of:
(2.1) selecting a differential interference pattern from the phase unwrapped differential interference patterns according to the threshold values of the time base line and the space base line to generate a differential interference pattern set
And from the differential interference atlas->
Selecting a differential interference atlas containing atmospheric phase
；
(2.2) pair differential interference atlas
Performing image sample and label making, performing image segmentation processing on each differential interferogram to generate an image sample, and performing normalization processing on the generated image sample to construct a first sample library of differential interferograms containing atmospheric phases and marked as->
。
4. The method of time-series InSAR atmospheric phase removal and deformation resolution according to claim 1, wherein said step (3) comprises the sub-steps of:
(3.1) first sample library based on differential interferograms containing atmospheric phase
Training and testing the condition generating antagonistic neural network CGAN to generate an additional second sample library containing atmospheric phase +.>
；
(3.2) storing the second sample library
Is +.>
Combining to generate a third sample library I containing differential interferograms of atmospheric phases; selecting a differential interference pattern which is not affected by the atmospheric phase according to the differential interference pattern after phase unwrapping, and carrying out image segmentation and data set manufacturing to generate a fourth sample library P of the differential interference pattern which does not contain the atmospheric phase; constructing a complete version sample library of the differential interferogram according to the third sample library I and the fourth sample library P and marking the complete version sample library as +.>
。
5. The method of time series InSAR atmospheric phase removal and deformation solution according to claim 4, wherein the condition generating antagonistic neural network CGAN comprises a generator and a arbiter.
6. The method for removing atmospheric phase and resolving deformation of time sequence InSAR according to claim 4, wherein the complete version sample library comprises an InSAR differential interferogram sample data set which corresponds to a fourth sample library and does not contain atmospheric phase, and the network corresponding to a second sample library generates an InSAR differential interferogram sample data set which contains atmospheric phase and an InSAR differential interferogram sample data set which corresponds to a first sample library and contains real atmospheric phase.
7. The method of time-series InSAR atmospheric phase removal and deformation resolution according to claim 1, wherein said step (4) comprises the sub-steps of:
(4.1) constructing an atmosphere phase removal transune network model based on a transune network to reconstruct and output a differential interference diagram after the atmosphere phase removal;
(4.2) library of full version samples
As an input data set of the atmosphere bit removal transune network model, dividing the input data set into a training data set and a test data set according to the proportion input by a user, loading the training data set into the atmosphere bit removal transune network model for training to obtain trained weight parameter information, and loading the trained weight parameter information by using the test data set to obtain a differential interferogram sample with the atmosphere phase removed;
(4.3) performing image slicing and merging on the differential interference pattern samples with the atmospheric phases removed to generate a differential interference pattern with the atmospheric phases removed.
8. The method for removing bits and resolving deformation of time sequence InSAR atmosphere according to claim 7, wherein the transune network combines two network structures of UNet and tranforms, and a U-shaped structure is formed by an encoder and a decoder.
9. The method of time-series InSAR atmospheric phase removal and deformation resolution according to claim 1, wherein the step (5) specifically comprises: and (3) performing time sequence InSAR deformation calculation based on the differential interferogram with the atmospheric phase removed, calculating time sequence deformation and accumulated deformation of the monitoring region by adopting a small baseline subset method, and using weighted least square estimation to realize stable calculation of the time sequence deformation so as to acquire the earth surface deformation information of the monitoring region.
10. A system of time series InSAR atmospheric phase removal and deformation solution for implementing the method of time series InSAR atmospheric phase removal and deformation solution of any one of claims 1-9, comprising:
the data preprocessing module is used for carrying out data importing and image registering on the acquired time sequence SAR images;
the data differential interference and filtering and phase unwrapping processing module is used for carrying out differential interference, nonlinear self-adaptive Goldstein spatial filtering and minimum cost flow phase unwrapping processes according to the registered time sequence SAR images;
the differential interferogram sample library construction and amplification module is used for generating samples of differential interferograms containing atmospheric phases, generating amplification of atmospheric phase samples of the anti-neural network CGAN based on conditions and constructing the differential interferogram sample library;
the atmosphere phase removal module is used for removing the atmosphere phase of the constructed differential interferogram sample library based on the TransUNet network; and
the time sequence InSAR deformation calculation module is used for carrying out time sequence InSAR deformation calculation on the generated differential interference diagram with the atmospheric phase removed, and obtaining the deformation information of the earth surface.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310349555.6A CN116148855B (en) | 2023-04-04 | 2023-04-04 | Method and system for removing atmospheric phase and resolving deformation of time sequence InSAR |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310349555.6A CN116148855B (en) | 2023-04-04 | 2023-04-04 | Method and system for removing atmospheric phase and resolving deformation of time sequence InSAR |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116148855A true CN116148855A (en) | 2023-05-23 |
| CN116148855B CN116148855B (en) | 2023-07-21 |
Family
ID=86339099
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310349555.6A Active CN116148855B (en) | 2023-04-04 | 2023-04-04 | Method and system for removing atmospheric phase and resolving deformation of time sequence InSAR |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116148855B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106772342A (en) * | 2017-01-11 | 2017-05-31 | 西南石油大学 | A kind of Timing Difference radar interference method suitable for big gradient surface subsidence monitoring |
| 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 |
| US20200394780A1 (en) * | 2017-06-15 | 2020-12-17 | The University Of Nottingham | Land deformation measurement |
| CN112816983A (en) * | 2021-01-06 | 2021-05-18 | 中南大学 | Time sequence InSAR turbulence atmospheric delay correction method based on optimized interferogram set |
| CN113340191A (en) * | 2021-04-08 | 2021-09-03 | 河北省高速公路延崇筹建处 | Time series interference SAR deformation quantity measuring method and SAR system |
| CN216411556U (en) * | 2021-04-08 | 2022-04-29 | 民航机场规划设计研究总院有限公司 | Synthetic aperture radar system based on time series interference deformation measurement |
| CN114814839A (en) * | 2022-03-22 | 2022-07-29 | 武汉大学 | InSAR phase gradient stacking-based wide-area earth surface deformation detection method and system |
| 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 |
| CN114966692A (en) * | 2022-07-19 | 2022-08-30 | 之江实验室 | Transformer-based InSAR technology frozen soil area multivariable time sequence deformation prediction method and device |
| CN115113203A (en) * | 2022-07-20 | 2022-09-27 | 江苏省水利科学研究院 | Method for removing InSAR atmospheric phase |
| CN115856889A (en) * | 2023-02-03 | 2023-03-28 | 北京数慧时空信息技术有限公司 | InSAR time sequence deformation monitoring method capable of automatically correcting errors |
-
2023
- 2023-04-04 CN CN202310349555.6A patent/CN116148855B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106772342A (en) * | 2017-01-11 | 2017-05-31 | 西南石油大学 | A kind of Timing Difference radar interference method suitable for big gradient surface subsidence monitoring |
| US20200394780A1 (en) * | 2017-06-15 | 2020-12-17 | The University Of Nottingham | Land deformation measurement |
| 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 |
| CN112816983A (en) * | 2021-01-06 | 2021-05-18 | 中南大学 | Time sequence InSAR turbulence atmospheric delay correction method based on optimized interferogram set |
| CN113340191A (en) * | 2021-04-08 | 2021-09-03 | 河北省高速公路延崇筹建处 | Time series interference SAR deformation quantity measuring method and SAR system |
| CN216411556U (en) * | 2021-04-08 | 2022-04-29 | 民航机场规划设计研究总院有限公司 | Synthetic aperture radar system based on time series interference deformation measurement |
| CN114814839A (en) * | 2022-03-22 | 2022-07-29 | 武汉大学 | InSAR phase gradient stacking-based wide-area earth surface deformation detection method and system |
| 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 |
| CN114966692A (en) * | 2022-07-19 | 2022-08-30 | 之江实验室 | Transformer-based InSAR technology frozen soil area multivariable time sequence deformation prediction method and device |
| CN115113203A (en) * | 2022-07-20 | 2022-09-27 | 江苏省水利科学研究院 | Method for removing InSAR atmospheric phase |
| CN115856889A (en) * | 2023-02-03 | 2023-03-28 | 北京数慧时空信息技术有限公司 | InSAR time sequence deformation monitoring method capable of automatically correcting errors |
Non-Patent Citations (2)
| Title |
|---|
| 李永生;: "高级时序InSAR地面形变监测及地震同震震后形变反演", 国际地震动态, no. 04, pages 42 - 43 * |
| 陈志达等: "基于时序InSAR技术的矿区形变监测与分析", 地理空间信息, vol. 20, no. 10, pages 5 - 10 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116148855B (en) | 2023-07-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yan et al. | A review of low‐cost space‐borne data for flood modelling: topography, flood extent and water level | |
| Wongchuig-Correa et al. | Assimilation of future SWOT-based river elevations, surface extent observations and discharge estimations into uncertain global hydrological models | |
| Karvonen | A sea ice concentration estimation algorithm utilizing radiometer and SAR data | |
| Chen et al. | ARU-net: Reduction of atmospheric phase screen in SAR interferometry using attention-based deep residual U-net | |
| Samsonov et al. | Multidimensional Small Baseline Subset (MSBAS) for volcano monitoring in two dimensions: Opportunities and challenges. Case study Piton de la Fournaise volcano | |
| Hostache et al. | A drifting GPS buoy for retrieving effective riverbed bathymetry | |
| CN109696152B (en) | Method for estimating ground settlement in low coherence region | |
| CN113281749A (en) | Time sequence InSAR high-coherence point selection method considering homogeneity | |
| Yun et al. | Mitigating atmospheric effects in InSAR measurements through high-resolution data assimilation and numerical simulations with a weather prediction model | |
| CN107037428A (en) | It is a kind of to improve the method that spaceborne dual station difference InSAR extracts deformation precision | |
| Tziavos et al. | The contribution of local gravimetric geoid models to the calibration of satellite altimetry data and an outlook of the latest GOCE GGM performance in GAVDOS | |
| Hamden et al. | Regional mean sea surface and mean dynamic topography models around Malaysian seas developed from 27 years of along-track multi-mission satellite altimetry data | |
| Dasgupta et al. | Earth observation and hydraulic data assimilation for improved flood inundation forecasting | |
| Wang et al. | Deep learning augmented data assimilation: Reconstructing missing information with convolutional autoencoders | |
| Conroy et al. | Probabilistic estimation of InSAR displacement phase guided by contextual information and artificial intelligence | |
| Fayad et al. | Hy-TeC: a hybrid vision transformer model for high-resolution and large-scale mapping of canopy height | |
| Medeiros et al. | Locality of topographic ground truth data for salt marsh Lidar DEM elevation bias mitigation | |
| Rexer et al. | Evaluation of the third-and fourth-generation GOCE Earth gravity field models with Australian terrestrial gravity data in spherical harmonics | |
| CN116068511B (en) | Deep learning-based InSAR large-scale system error correction method | |
| Elfadaly et al. | Revealing the paleolandscape features around the archaeological sites in the northern Nile Delta of Egypt using radar satellite imagery and GEE platform | |
| CN116148855B (en) | Method and system for removing atmospheric phase and resolving deformation of time sequence InSAR | |
| Klos et al. | Introducing the idea of classifying sets of permanent gnss stations as benchmarks for hydrogeodesy | |
| CN111611929A (en) | River flood risk point identification method, device, server and storage medium based on LiDAR and InSAR technologies | |
| Cortese et al. | Coupling numerical models of deltaic wetlands with AirSWOT, UAVSAR, and AVIRIS-NG remote sensing data | |
| Yue et al. | A global seamless DEM based on multi-source data fusion (GSDEM-30): Product generation and evaluation |
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 |