A method for evaluating the risk level of river bank collapse

Abstract

This application relates to a method for evaluating the risk level of river bank collapse, including the following steps: collecting remote sensing images and DEM data of the research area; processing SAR satellite remote sensing images using the SBAS technique to obtain the deformation rate map of the research area; after identifying possible surface deformations in the research area, further synchronously conducting optical satellite image observations to obtain more detailed potential bank collapse information; after identifying bank collapse through optical satellite images, conducting emergency monitoring of the river channel topography near the bank collapse and performing shipborne hydrographic and topographic surveys of the bank collapse; constructing a Bayesian model for the risk level of river bank collapse, selecting the various data collected in the above steps as inputs, and evaluating the risk level. This application strengthens the monitoring capabilities and means of river bank collapse, better meets the emergency response and rescue work for bank collapse emergencies, can provide a scientific basis for relevant departments, formulate targeted disaster prevention and mitigation measures, and thus ensure the safety of the river channel and the stability of the surrounding environment.

Description

Technical Field

[0001] This application relates to the technical field of bank collapse monitoring and early warning, and particularly to a method for evaluating the risk level of river bank collapse by integrating SBAS-InSAR, optical images, emergency monitoring and Bayesian statistics. Background Technique

[0002] Bank collapse refers to the physical process in which the river bank slope collapses and retreats under the continuous scouring of the water flow. According to incomplete statistics, from 2003 to 2021, there were about 1044 bank collapse hazards and the bank collapse length was about 747 km in the main stream of the middle and lower reaches of the Yangtze River only.

[0003] In recent years, the state has gradually strengthened the inspection and monitoring of dangerous bank collapses in the middle and lower reaches of the Yangtze River. However, due to the 5046-km-long shoreline of the main stream of the middle and lower reaches of the Yangtze River, and the characteristics of bank collapse disasters such as suddenness, multiplicity, and randomness, the early warning is extremely difficult, which can be said to be one of the most complex problems faced by the middle and lower reaches of the Yangtze River at present. In the past, the observations of bank collapses in the middle and lower reaches of the Yangtze River were all based on the conventional river channel observations and the collection of basic data. The observation ability was conventional, the observation means were limited, the types of results were relatively single, and the resolution and timeliness of the results were difficult to meet the current situation of frequent bank collapses and the needs of disaster prediction, flood control and emergency rescue, and emergency rescue of bank collapses. Summary of the Invention

[0004] The purpose of this application is to provide a method for evaluating the risk level of river bank collapse, which can make scientific decisions on bank collapse disasters, take corresponding engineering measures as early as possible to avoid major bank collapse hazards, better provide technical support for the treatment of bank collapses in the middle and lower reaches of the Yangtze River, further meet the needs of shoreline protection, prevention and control of bank collapse hazards and emergency rescue and disposal, and ensure the safety of the river channel.

[0005] To achieve the above purpose, this application provides the following technical solutions:

[0006] An embodiment of this application provides a method for evaluating the risk level of river bank collapse, including the following steps:

[0007] Step 1, collect remote sensing images and DEM data of the study area, including remote sensing image data of SAR satellites and optical satellites covering the study area, and DEM data of the study area;

[0008] Step 2, use the SBAS technology to process the SAR satellite remote sensing images to obtain the deformation rate map of the study area;

[0009] Step 3, after identifying possible surface deformations in the study area, further synchronously carry out optical satellite image observations to obtain more detailed potential bank collapse information;

[0010] Step 4: After identifying the bank collapse through optical satellite imagery, conduct emergency monitoring of the river channel topography near the bank collapse. Use hardware integration such as a three-dimensional laser scanner, multi-beam sounding system, and GNSS / INS tightly coupled inertial navigation system to perform shipborne land and water three-dimensional measurement of the bank collapse;

[0011] Step 5: Construct a Bayesian model for the risk level of river bank collapse. Select the various data collected in the above steps as inputs to evaluate the risk level.

[0012] The implementation method of the above Step 2 is specifically as follows:

[0013] Preprocessing: Preprocess the SAR and DEM data collected in Step 1, including range correction, radiometric correction, covariance matrix filtering, and DEM registration with the main image;

[0014] Interferometry and differential interferometry processing: Use the preprocessed SAR data for interferometry processing, that is, perform registration and phase difference on SAR images at multiple time points to obtain surface deformation information;

[0015] Phase unwrapping: After interferometry processing, perform phase unwrapping to obtain accurate deformation amounts;

[0016] Temporal deformation analysis: Use the processed SAR data set above for SBAS-InSAR analysis, that is, perform temporal interferometry analysis to obtain the deformation amount of each pixel point;

[0017] Result analysis: Analyze the temporal deformation results, and use statistical methods or model fitting to obtain the deformation rate characteristics of the surface.

[0018] The above Step 3 is specifically as follows:

[0019] Image preprocessing: Preprocess the optical satellite imagery obtained in Step 1, including atmospheric correction, geometric correction, and radiometric correction;

[0020] Image registration: Corresponding to the InSAR data, register the optical satellite imagery with the previously used SAR data to ensure that the image positions corresponding to the surface deformation areas are consistent;

[0021] Image interpretation: Use the registered optical satellite imagery to interpret and analyze the previously identified surface deformation areas;

[0022] Extract change information: During the image interpretation process, identify the water body and river bank parts in the image, extract the water edge coordinates, compare and calculate the difference in water edge coordinates at different times, and analyze to obtain the occurrence and expansion of the bank collapse.

[0023] The shipborne land and water three-dimensional measurement of the bank collapse in the above Step 4 is specifically as follows:

[0024] Shipborne land - water three - dimensional measurement: Mount a three - dimensional laser scanner and a multi - beam sounding system on the same ship to integrally measure the on - shore and underwater topography, so as to improve the measurement efficiency and accuracy;

[0025] Data acquisition: The three - dimensional laser scanner and multi - beam technology continuously obtain a large amount of coordinate point data. At the same time, the GNSS / INS tightly coupled inertial navigation system records the position information of the ship on water and on land, and collects and collates these data for subsequent data processing;

[0026] Data processing and analysis: Use professional geographic information system software to convert the coordinate point data into a three - dimensional terrain model, and analyze and visualize it. By analyzing the three - dimensional point cloud data, understand the terrain changes and potential risk factors at the bank - caving area.

[0027] The implementation method of step 5 is specifically as follows:

[0028] Collect data: Let the historical bank - caving data set that has been collected be D, which includes N sample data. Each sample data includes the degree of surface deformation, the change range of the water edge line, on - site monitoring data, and the corresponding bank - caving risk level;

[0029] Construct a Bayesian model: Assume that the bank - caving risk level is R and the sample data is D, and establish a conditional probability model P(R|D), which represents the conditional probability of the bank - caving risk level given the sample data;

[0030] Establish prior probabilities: Determine based on past experience, expert knowledge, or according to the frequencies of various potential risk levels in the data set D. The prior probability of high risk is P(R = high), the prior probability of medium risk is P(R = medium), and the prior probability of low risk is P(R = low);

[0031] Update probabilities according to the observed data: The sample data observed this time is D_obs, then according to Bayes' theorem, use the observed data to update the probabilities of the potential risk levels;

[0032] Calculate the posterior probabilities: Calculate the posterior probabilities of each risk level according to the probabilities updated by the observed data, and take the level with the largest posterior probability value as the final risk level.

[0033] Compared with the prior art, the beneficial effects of the present invention are:

[0034] The present invention evaluates the risk level of river - bank caving based on multi - source images, emergency monitoring data, and Bayesian statistics, strengthens the monitoring capabilities and means of river - bank caving, better meets the emergency disposal and rescue work of bank - caving emergencies, can provide a scientific basis for relevant departments, formulate targeted disaster prevention and mitigation measures, and thus ensure the safety of the river and the stability of the surrounding environment. Description of the Drawings

[0035] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the accompanying drawings required for use in the embodiments of the present application. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation of the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0036] Figure 1 It is the specific flowchart of the method of the present invention;

[0037] Figure 2 It is the schematic diagram of the research area where the embodiment is located;

[0038] Figure 3 It is the schematic diagram of the collection of optical remote sensing image data;

[0039] Figure 4 It is the schematic diagram of the SBAS-InSAR processing flow;

[0040] Figure 5 It is the scope of the deformation area and the severely deformed area of the embodiment;

[0041] Figure 6 It is the water edge line change map of the embodiment based on optical satellite images;

[0042] Figure 7 It is the three-dimensional point cloud data of the land and water terrain before and after the bank collapse of the embodiment. Specific Embodiments

[0043] The following will describe the technical solutions in the embodiments of the present application in conjunction with the accompanying drawings in the embodiments of the present application. It should be noted that: similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0044] The term "comprising", "including" or any other variant thereof is intended to cover a non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the phrase "comprising a..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0045] The terms "first", "second", etc. are only used to distinguish one entity or operation from another, and should not be construed as indicating or implying relative importance, nor should they be construed as requiring or implying any such actual relationship or order between these entities or operations.

[0046] As Figure 1 shown, the present invention provides a method for river bank collapse risk level and assessment, including the following specific steps:

[0047] S1. Collect data such as remote sensing images and DEM of the study area. This includes remote sensing image data of SAR satellites and optical satellites covering the study area, and DEM data of the study area;

[0048] S2. Use the SBAS technique to process the SAR images to obtain the deformation rate map of the study area;

[0049] S3. After identifying possible surface deformations in the study area through the SBAS-InSAR method in S2, further synchronously conduct optical satellite image observations to obtain more detailed potential bank collapse information;

[0050] S4. After identifying the bank collapse through the optical satellite images in S3, conduct emergency monitoring on the river channel topography near the bank collapse. Use the integration of hardware such as 3D laser scanners, multibeam bathymetric systems, and GNSS / INS tightly coupled inertial navigation systems to perform shipborne land and water three-dimensional surveys of the bank collapse;

[0051] S5. Construct a Bayesian model for the river bank collapse risk level, select the data collected in the above steps as inputs, and conduct risk level assessment.

[0052] The following further specifically describes the technical solution of the present invention through embodiments and in conjunction with the drawings:

[0053] Step 1, collect data such as remote sensing images and DEM of the study area. This includes remote sensing image data of SAR satellites and optical satellites covering the study area, and DEM data of the study area. The data collection belongs to the conventional technology in this technical field.

[0054] Taking the Xiaopan section of the Yangtze River as an example, this study area is located at the administrative boundary between the northern part of Jiayu County, Xianning City, Hubei Province and the southern part of Honghu City, Jingzhou City. This location is bounded by the Yangtze River, and there are mostly farmlands on both sides of the river bank. The study area ranges from 114.07°E to 114.11°E and 30.11°N to 30.14°N, as shown in Figure 2 shown.

[0055] Step 1.1: Obtain SAR satellite remote sensing image data covering the study area. In this embodiment, a total of 15 scenes of Sentinel-1A IW SAR ascending orbit images from October 2021 to March 2022 (the basic information of the images is shown in Table 1) are collected as the basic data for SBAS-InSAR deformation monitoring.

[0056] Table 1 Sentinel-1 SAR image data

[0057]

[0058] Step 1.2: Obtain optical remote sensing image data covering the study area. In this embodiment, based on the data services of the Hubei Data and Application Center of the High-Resolution Earth Observation System, a total of 30 scenes of images from October 2021 to March 2022 of domestic Gaofen-1 and Gaofen-2 satellites are collected, as follows Figure 3 shown, including panchromatic band data and 4-band multispectral data, and the 4 bands include red, green, blue, and near-infrared bands.

[0059] Step 1.3: Collect DEM data. The terrain phase simulation data mainly uses ALOS World 3D DEM data. The spatial resolution of ALOS World 3D DEM data is 30m. During the InSAR deformation monitoring process, ALOS World 3D has relatively rich terrain information and can better reduce the influence of terrain phase on the SBAS-InSAR monitoring results.

[0060] Step 2: Use the SBAS (Small Baseline Subsets, abbreviated as SBAS) technology to process the SAR images to obtain the deformation rate map of the study area.

[0061] The process is as follows Figure 4 shown, and the steps are introduced as follows:

[0062] (1) Pretreatment: Pretreat the SAR and DEM data collected in Step 1, including steps such as range correction, radiometric correction, covariance matrix filtering, and DEM registration with the main image, to improve data quality and reduce noise.

[0063] (2) Interferometry and differential interferometry processing: Use the preprocessed SAR data for interferometry processing, that is, register and perform phase difference on SAR images at multiple time points to obtain surface deformation information.

[0064] (3) Phase unwrapping: After interferometry processing, there may be a problem of discontinuous phase, and phase unwrapping is required to obtain accurate deformation amounts.

[0065] (4) Temporal deformation analysis: Use the processed SAR dataset above for SBAS-InSAR analysis, that is, perform temporal interferometric analysis to obtain the deformation amount of each pixel point.

[0066] (5) Result analysis: Analyze the temporal deformation results, which can adopt statistical methods or model fitting, etc., to obtain the deformation rate characteristics of the ground surface.

[0067] For the embodiment, SBAS-InSAR temporal deformation solution is adopted to obtain the deformation rate results of the study area as shown in Figure 5 The figure shows the ground surface deformation rate characteristics from October 12, 2021 to December 11, 2021. Negative values in the figure represent that the deformation direction is consistent with the satellite line-of-sight (LOS) direction (displacement occurs away from the LOS direction, that is, subsidence), and positive values indicate that the deformation direction is opposite to the LOS direction (displacement occurs towards the LOS direction, that is, uplift).

[0068] Most of the ground surface deformations in the study area show relative stability. According to the deformation rate, 7 obvious deformation areas are identified, mainly distributed near both banks of the Yangtze River. Among them, one deformation area intersects with the embankment, which is the key area of concern for potential bank collapse. It will be intensively observed by optical satellite remote sensing in the next step. The remaining deformation areas are all located in farmland.

[0069] Step 3: After identifying the possible ground surface deformation in the study area by the SBAS-InSAR method in Step 2, further synchronously carry out intensive observation of optical satellite images to obtain more detailed information on potential bank collapse. The steps are introduced as follows:

[0070] (1) Image preprocessing: Preprocess the optical satellite images obtained in Step 1, including atmospheric correction, geometric correction, radiometric correction, etc., to ensure the accuracy and comparability of the data.

[0071] (2) Image registration: Corresponding to the InSAR data, it is necessary to register the optical satellite images with the previously used SAR data to ensure that the image positions corresponding to the ground surface deformation areas are consistent.

[0072] (3) Image interpretation: Use the registered optical satellite images to interpret and analyze the previously identified ground surface deformation areas. The method adopted is DeepLabv3+. Semantic segmentation is a very important field in computer vision. It refers to identifying images at the pixel level, that is, labeling the object category to which each pixel in the image belongs. DeepLabv3+ is a typical semantic segmentation algorithm based on deep neural networks, and it has shown good image segmentation performance in multiple studies.

[0073] (4) Extract change information: During the image interpretation process, the water bodies and riverbank parts in the image can be determined by the DeepLabv3+ method, and the waterline coordinates can be extracted. By comparing and calculating the differences in waterline coordinates at different times, the occurrence and expansion of bank collapse can be analyzed.

[0074] For the embodiment, the results are as Figure 6 shown. By comparing the waterlines extracted from multiple satellite imagery, the occurrence and expansion of bank collapse can be clearly observed. This result proves that the technical route of automatically observing and analyzing the changes of the shoreline through optical satellite imagery at regular intervals and identifying the further expansion of bank collapse is feasible.

[0075] Step 4: After identifying the occurrence of bank collapse through the optical satellite imagery in Step 3, conduct emergency monitoring on the river channel terrain near the bank collapse. Use the integration of hardware such as 3D laser scanners, multibeam bathymetric systems, and GNSS / INS tightly coupled inertial navigation systems to perform shipborne land and water three-dimensional surveys of the bank collapse. Finally, obtain the three-dimensional monitoring data of the land and water terrain under the target coordinates.

[0076] The steps are introduced as follows:

[0077] (1) Shipborne land and water three-dimensional survey: Mount a 3D laser scanner and a multibeam bathymetric system on the same ship to integrally measure the onshore and underwater terrains to improve the measurement efficiency and accuracy. (2) Data acquisition: The 3D laser scanner and multibeam technology continuously obtain a large amount of coordinate point data. At the same time, the GNSS / INS tightly coupled inertial navigation system records the position information of the ship on water and on land. Collect and organize these data for subsequent data processing.

[0078] (3) Data processing and analysis: Use professional geographic information system (GIS) software or similar tools to convert the coordinate point data into a three-dimensional terrain model and analyze and visualize it. By analyzing the three-dimensional point cloud data, the terrain changes and potential risk factors at the bank collapse can be understood.

[0079] For the embodiment, the visualization results are as Figure 7 shown, and the terrain information such as the scope of the bank collapse can be clearly observed.

[0080] Step 5: Construct a Bayesian model for the risk level of river channel bank collapse, select the data collected in the above steps as input, and conduct risk level assessment.

[0081] The steps are introduced as follows:

[0082] (1) Collect data: Let the collected historical bank collapse dataset be D, which includes N sample data. Each sample data includes the degree of surface deformation D_i 1 , the change amplitude of the waterline D_i 2, on-site monitoring data \(D_i\) 3 and the corresponding bank collapse risk level \(R_i\).

[0083] (2) Construct a Bayesian model: We assume that the bank collapse risk level is \(R\) and the sample data is \(D\). A conditional probability model \(P(R|D)\) can be established, which represents the conditional probability of the bank collapse risk level given the sample data. Specifically, we can use a Bayesian network or a probabilistic graphical model to represent the dependence relationship between various factors.

[0084] (3) Establish prior probabilities: We assume that the prior probability of high risk is \(P(R = high)\), the prior probability of medium risk is \(P(R = medium)\), and the prior probability of low risk is \(P(R = low)\). The prior probabilities can be determined based on past experience, expert knowledge, or according to the frequencies of various potential risk levels in the dataset \(D\).

[0085] (4) Update probabilities based on observed data: Assume that the sample data observed this time is \(D_{obs}\). Then, according to Bayes' theorem, the observed data can be used to update the probabilities of potential risk levels. Specifically, we have the following formula:

[0086] \(P(R|D_{obs}) = P(R) * P(D_{obs}|R) / P(D_{obs})\)

[0087] where \(P(R)\) is the prior probability, \(P(D_{obs}|R)\) is the probability of observing the sample data \(D_{obs}\) given the risk level \(R\), and \(P(D_{obs})\) is the marginal probability of the sample data being \(D_{obs}\).

[0088] For each potential risk level \(R\), we can calculate \(P(D_{obs}|R)\) through the following formula:

[0089] \(P(D_{obs}|R)=\int P(D_{obs}|D,R)*P(D|R)dD\)

[0090] where \(P(D_{obs}|D,R)\) is the probability density function of observing the sample data \(D_{obs}\) given the sample data \(D\) and the risk level \(R\), and \(P(D|R)\) is the probability density function of the sample data being \(D\) given the risk level \(R\).

[0091] \(P(D_{obs})\) can be calculated through the following formula:

[0092] \(P(D_{obs})=\sum P(D_{obs}|R)*P(R)\)

[0093] (5) Calculate posterior probabilities: Based on the probabilities updated according to the observed data, the posterior probabilities of each risk level can be calculated. According to the normalization principle, specifically, we have the following formula:

[0094] P(R = high|D_obs) = P(R = high) * P(D_obs|R = high) / P(D_obs)

[0095] P(R = medium|D_obs) = P(R = medium) * P(D_obs|R = medium) / P(D_obs)

[0096] P(R = low|D_obs) = P(R = low) * P(D_obs|R = low) / P(D_obs)

[0097] (6) Risk level classification: According to the posterior probability, the level with the largest posterior probability value is used as the final risk level.

[0098] For the embodiment, the degree of ground surface deformation is 50 mm, the change range of the water edge line is about 60 m, the field monitoring data shows that the landslide length is about 100 m, the landslide width and depth are about 60 m, and the height of the hanging bank is 12 m. Input into the established Bayesian model, the probability value of P(R = high|D_obs) is the largest. The early warning signals and countermeasures with a high risk level in Table 2 below should be adopted.

[0099] Table 2 Bank Collapse Risk Classification Criteria and Countermeasures

[0100]

[0101] The above is only the embodiment of the present application and is not used to limit the protection scope of the present application. For those skilled in the art, the present application can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for evaluating the risk level of river bank collapse, characterized in that, it includes the following steps: Step 1, collect remote sensing images and DEM data of the study area, including remote sensing image data of SAR satellites and optical satellites covering the study area, and DEM data of the study area; Step 2, use the SBAS technology to process the SAR satellite remote sensing images to obtain the deformation rate map of the study area; Step 3, after identifying the possible surface deformation in the study area, further conduct optical satellite image observations synchronously to obtain more detailed potential bank collapse information; Step 4, after identifying the bank collapse through optical satellite images, conduct emergency monitoring on the river channel topography near the bank collapse, and use the hardware integration of a three-dimensional laser scanner, a multibeam sounding system, and a GNSS / INS tightly coupled inertial navigation system to conduct shipborne land-water three-dimensional measurement of the bank collapse; Step 5, construct a Bayesian model for the risk level of river bank collapse, select the data collected in the above steps as input, and conduct risk level assessment; The shipborne land-water three-dimensional measurement in Step 4 is specifically: Shipborne land-water three-dimensional measurement: Mount a three-dimensional laser scanner and a multibeam sounding system on the same ship to integrally measure the onshore and underwater topography to improve the measurement efficiency and accuracy; Data collection: The three-dimensional laser scanner and multibeam technology continuously obtain a large amount of coordinate point data. At the same time, the GNSS / INS tightly coupled inertial navigation system records the position information of the ship on water and on land, and collect and organize these data for subsequent data processing; Data processing and analysis: Use professional geographic information system software to convert the coordinate point data into a three-dimensional terrain model, and conduct analysis and visualization on it. By analyzing the three-dimensional point cloud data, understand the terrain changes and potential risk factors at the bank collapse; The implementation method of Step 5 is specifically: Collect data: Let the collected historical bank collapse data set be D, which includes N sample data. Each sample data includes the degree of surface deformation, the change range of the water edge line, on-site monitoring data, and the corresponding bank collapse risk level; Construct a Bayesian model: Assume that the bank collapse risk level is R and the sample data is D, and establish a conditional probability model P(R|D), which represents the conditional probability of the bank collapse risk level given the sample data; Establish a prior probability: Determine based on past experience, expert knowledge, or according to the frequency of each potential risk level in the data set D. The prior probability of high risk is P(R = high), the prior probability of medium risk is P(R = medium), and the prior probability of low risk is P(R = low); Update the probability according to the observed data: The sample data observed this time is D_obs, then according to Bayes' theorem, use the observed data to update the probability of the potential risk level; Calculate the posterior probability: According to the probability updated by the observed data, calculate the posterior probability of each risk level, and use the level with the largest posterior probability value as the final risk level.

2. The method for evaluating the risk level of river bank collapse according to claim 1, characterized in that, the implementation method of Step 2 is specifically: Preprocessing: Preprocess the SAR and DEM data collected in Step 1, including range correction, radiometric correction, covariance matrix filtering, and DEM registration with the main image; Interferometry and differential interferometry processing: Using the preprocessed SAR data, perform interferometry processing, that is, register and perform phase difference on SAR images at multiple time points to obtain surface deformation information; Phase unwrapping: After interferometry processing, perform phase unwrapping to obtain accurate deformation amounts; Temporal deformation analysis: Use the processed SAR dataset above for SBAS-InSAR analysis, that is, perform temporal interferometry analysis to obtain the deformation amount of each pixel point; Result analysis: Analyze the temporal deformation results, and use statistical methods or model fitting to obtain the deformation rate characteristics of the surface; 3. A method for evaluating the risk level of river bank collapse according to claim 1, characterized in that, the specific content of Step 3 is as follows: Image preprocessing: Preprocess the optical satellite image obtained in Step 1, including atmospheric correction, geometric correction, and radiometric correction; Image registration: Corresponding to the InSAR data, register the optical satellite image with the previously used SAR data to ensure that the image positions corresponding to the surface deformation area are consistent; Image interpretation: Use the registered optical satellite image to interpret and analyze the previously identified surface deformation area; Extract change information: During the image interpretation process, judge the water body and river bank parts in the image, extract the water edge coordinates, compare and calculate the difference in water edge coordinates at different times, and analyze the occurrence and expansion of the river bank collapse.

Images