A geological disaster early warning method, system, device and storage medium

Abstract

The application discloses a geological disaster early warning method, system, device and storage medium. The geological disaster early warning method comprises the following steps: determining a first time sequence optical change type sequence of a target area through a cloud detection algorithm; determining a first time sequence deformation sequence through a preset screening rule based on single-view complex product data and pixel distance; determining a first soil water content sequence of the target area based on optical remote sensing data and ground distance product data; inputting the first time sequence rainfall sequence, the second time sequence rainfall sequence, the first time sequence optical change type sequence, the first time sequence deformation sequence and the first soil water content sequence into a geological disaster prediction model to obtain a geological disaster occurrence probability output by the geological disaster prediction model, and performing early warning based on the geological disaster occurrence probability, thereby improving the accuracy of geological disaster early warning.

Description

Technical Field

[0001] This application relates to the technical field of geological disaster early warning, and in particular to a geological disaster early warning method, system, equipment and storage medium. Background Technology

[0002] Currently, geological disaster early warning methods mainly include optical remote sensing technology, time-series synthetic aperture radar interferometry (TISAR) technology, and hydrological monitoring data. While optical remote sensing can identify potential disaster hazards through changes in land cover, it is greatly affected by cloud and rain weather and cannot capture deep surface deformation and soil hydrological conditions. TSAR technology can monitor surface deformation at the millimeter level, but it struggles to directly correlate deformation with triggering factors such as rainfall and soil moisture content. Although hydrological monitoring data reflects changes in surface hydrological conditions, it lacks dynamic coupling analysis with geological deformation, making it difficult to form effective synergy among different data sources. Therefore, breaking down data barriers and integrating multi-source heterogeneous data has become a key breakthrough point for geological disaster early warning. Summary of the Invention

[0003] This application aims to at least address the technical problems existing in the prior art. To this end, this application proposes a geological disaster early warning method, system, device, and storage medium, which can integrate first time-series rainfall sequences, second time-series rainfall sequences, optical remote sensing data, single-view complex product data, and ground distance product data to improve the accuracy of geological disaster early warning.

[0004] The first aspect of this application provides a geological disaster early warning method, comprising the following steps: The system acquires a first time-series rainfall sequence, a second time-series rainfall sequence, optical remote sensing data, single-view complex product data, ground distance product data, and pixel distance for the target area. The first time-series rainfall sequence consists of the total rainfall for each first time period, which is obtained by dividing a first preset historical time period by a first preset time length. The second time-series rainfall sequence consists of the total rainfall for each second time period, which is obtained by dividing a second preset historical time period by a second preset time length. The single-view complex product data and the ground distance product data are both data obtained from a double-orbit time-series synthetic aperture radar (SAPRA) within the satellite revisit period to which the predicted time belongs. The pixel distance is the pixel distance of the synthetic aperture radar data obtained from the double-orbit time-series SAPRA. Based on the optical remote sensing data, the first temporal optical change type sequence of the target area is determined by a cloud detection algorithm; Based on the single-view complex product data and pixel distance, a first temporal deformation sequence is determined by a preset filtering rule; Based on the optical remote sensing data and the ground distance product data, a first soil moisture content sequence for the target area is determined; The first time-series rainfall sequence, the second time-series rainfall sequence, the first time-series optical change type sequence, the first time-series deformation sequence, and the first soil moisture content sequence are input into the geological disaster prediction model to obtain the probability of geological disaster occurrence output by the geological disaster prediction model, and early warning is issued based on the probability of geological disaster occurrence.

[0005] The geological disaster early warning method according to the embodiments of this application has at least the following beneficial effects: This method acquires the first and second time-series rainfall sequences, optical remote sensing data, single-view complex product data, ground distance product data, and pixel distance of the target area. Based on the optical remote sensing data, it determines the first time-series optical change type sequence of the target area using a cloud detection algorithm; based on the single-view complex product data and pixel distance, it determines the first time-series deformation sequence using a preset filtering rule; and based on the optical remote sensing data and ground distance product data, it determines the first soil moisture content sequence of the target area. The method then combines the first time-series rainfall sequence, the second time-series rainfall sequence, the first time-series optical change type sequence, and the second time-series complex product data... This application integrates a first-time-series deformation sequence and a first-time-series soil moisture content sequence into a geological disaster prediction model to obtain the probability of geological disaster occurrence output by the model. Based on this probability, early warning is issued. This application constructs a geological disaster prediction model by integrating a first-time-series rainfall sequence, a second-time-series rainfall sequence, optical remote sensing data, single-view complex product data, and ground distance product data, thereby achieving accurate early warning of geological disasters. This application breaks through the data silos and mechanism deficiencies of traditional technologies, provides a brand-new solution for disaster prevention and mitigation in complex geological environments, and improves the accuracy of geological disaster early warning.

[0006] According to some embodiments of this application, determining the first temporal optical change type sequence of the target area based on the first optical remote sensing data using a cloud detection algorithm includes: The first optical remote sensing data is evenly divided according to a first preset time interval to obtain the second optical remote sensing data. Traverse all the second optical remote sensing data, filter all the second optical remote sensing data with cloud obstruction using the cloud detection algorithm, use them as the third optical remote sensing data, and mark the third optical remote sensing data with the first preset classification label to obtain the first-labeled optical remote sensing data; Determine the fourth optical remote sensing data, wherein the fourth optical remote sensing data is all other optical remote sensing data in the second optical remote sensing data except for the third optical remote sensing data; The fourth optical remote sensing data is input into the temporal optical change classification network model so that the temporal optical change classification network model uses a second preset classification label symbol to label each of the fourth optical remote sensing data, thereby obtaining the second-labeled optical remote sensing data, wherein the second preset classification label symbol is different from the first preset classification label symbol; The first temporal optical change type sequence is determined based on the first labeled optical remote sensing data and the second labeled optical remote sensing data.

[0007] According to some embodiments of this application, determining the first temporal optical change type sequence based on the first labeled optical remote sensing data and the second labeled optical remote sensing data includes: Each first-labeled optical remote sensing data is uniformly divided according to a second preset time interval to obtain a first segmented optical remote sensing sequence for each first-labeled optical remote sensing data; each second-labeled optical remote sensing data is uniformly divided according to the second preset time interval to obtain a second segmented optical remote sensing sequence for each second-labeled optical remote sensing data, wherein the classification label symbol of each first-segmented optical remote sensing data in the first segmented optical remote sensing sequence is the same as the classification label symbol of the first-labeled optical remote sensing data with the closest time distance, and the classification label symbol of each second-segmented optical remote sensing data in the second segmented optical remote sensing sequence is the same as the classification label symbol of the second-labeled optical remote sensing data with the closest time distance; By combining all the first segmented optical remote sensing sequences and all the second segmented optical remote sensing sequences, the first temporal optical change type sequence is obtained.

[0008] According to some embodiments of this application, the single-view complex product data includes multi-period image data, and determining the first temporal deformation sequence of the target region based on the single-view complex product data and pixel distance includes: Based on preset time baseline thresholds and preset spatial baseline thresholds, image pairs are filtered on the single-view complex product data to obtain a set of image pairs and the date of each image pair; Construct an image pair index sequence for the image pair set based on the date of each image pair; Based on the image pair index sequence, image registration is performed on the image pair set to obtain the first registered image pair sequence; Based on the first registered image pair sequence, the phase-unwrapped image pair sequence is determined by the phase unwrapping method; Based on the phase-unwrapped image pair sequence, the second temporal deformation sequence is determined by the singular value decomposition method; Based on the first preset filtering value, the second time-series deformation sequence is filtered to obtain the third time-series deformation sequence; Calculate the average coherence of each period's image data of the single-view complex product data; The image data of each period is traversed, and the image data with the maximum value of the average coherence is selected as the master image data. The remaining image data are all image registered with the master image data to obtain the second registered image pair sequence. Differential interferometry is performed on the second registered image pair sequence to obtain differential interferometry data; The differential interferometric data are filtered using the amplitude deviation index and a preset screening threshold to obtain a permanent scatterer sequence; The deformation rate of the permanent scatterer sequence was calculated using the least squares method to obtain the fourth time-series deformation sequence; Based on the second preset filtering value, the fourth time-series deformation sequence is filtered to obtain the fifth time-series deformation sequence; The first temporal deformation sequence is determined based on the third temporal deformation sequence, the fifth temporal deformation sequence, and the pixel distance.

[0009] According to some embodiments of this application, determining the first temporal deformation sequence based on the third temporal deformation sequence, the fifth temporal deformation sequence, and the pixel distance includes: Calculate the distance between each point in the third temporal deformation sequence and each point in the fifth temporal deformation sequence in the preset coordinate system. The temporal deformation sequence includes several temporal deformation values ​​and the latitude and longitude corresponding to each temporal deformation value. The points in the temporal deformation sequence are obtained by projecting the latitude and longitude corresponding to each temporal deformation value onto the preset coordinate system. Traverse each point in the third temporal deformation sequence, and filter out points in the third temporal deformation sequence whose distance from any point in the fifth temporal deformation sequence is less than a preset distance threshold to obtain points with the same name in the third temporal deformation sequence, wherein the preset distance threshold is one-tenth of the pixel distance; Determine the sixth temporal deformation sequence, wherein the sixth temporal deformation sequence is a sequence composed of all points other than the corresponding points in the third temporal deformation sequence; The fifth temporal deformation sequence and the sixth temporal deformation sequence are combined to obtain the seventh temporal deformation sequence; The seventh time-series deformation sequence is uniformly divided according to the third preset time interval to obtain the eighth time-series deformation sequence; the seventh time-series deformation sequence is uniformly divided according to the fourth preset time interval to obtain the ninth time-series deformation sequence. The first time-series deformation sequence is obtained by linear fitting based on the eighth time-series deformation sequence and the ninth time-series deformation sequence.

[0010] According to some embodiments of this application, determining the first soil moisture content sequence of the target area based on the first optical remote sensing data and the ground distance product data includes: Extract the homopolarization and crosspolarization data of the ground distance product data; The same-polarization data is filtered to obtain filtered same-polarization data; the cross-polarization data is filtered to obtain filtered cross-polarization data. Geocoding is performed on the filtered homopolarized data to obtain encoded homopolarized data; geocoding is performed on the filtered cross-polarized data to obtain encoded cross-polarized data. Radiometric calibration is performed on the encoded homopolarized data to obtain homopolarized backscattering coefficients; radiometric calibration is performed on the encoded crosspolarized data to obtain crosspolarized backscattering coefficients. The same polarization backscattering coefficient is input into the water cloud model to obtain the first backscattering coefficient output by the water cloud model; the cross polarization backscattering coefficient is input into the water cloud model to obtain the second backscattering coefficient output by the water cloud model. Calculate the normalized vegetation index of the first optical remote sensing data; The first soil moisture content sequence of the target area is determined based on the first backscattering coefficient, the second backscattering coefficient, and the normalized vegetation index.

[0011] According to some embodiments of this application, determining the first soil moisture content sequence of the target area based on the first backscattering coefficient, the second backscattering coefficient, and the normalized vegetation index includes: The first backscattering coefficient, the second backscattering coefficient, and the normalized vegetation index are input into the soil moisture content prediction model to obtain the first soil moisture content prediction sequence of the target area output by the soil moisture content prediction model. The first soil moisture content prediction sequence is uniformly divided according to a fourth preset time interval to obtain a second soil moisture content prediction sequence; the first soil moisture content prediction sequence is uniformly divided according to a fifth preset time interval to obtain a third soil moisture content prediction sequence. The first soil moisture content sequence is obtained by linear fitting based on the second and third soil moisture content prediction sequences.

[0012] A second aspect of this application provides a geological disaster early warning system, the geological disaster early warning system comprising: The data acquisition module is used to acquire a first time-series rainfall sequence, a second time-series rainfall sequence, first optical remote sensing data, single-view complex product data, ground distance product data, and pixel distance for a target area. The first time-series rainfall sequence is a rainfall sequence composed of the total rainfall for each first time period, which is obtained by dividing a first preset historical time period by a first preset time length. The second time-series rainfall sequence is a rainfall sequence composed of the total rainfall for each second time period, which is obtained by dividing a second preset historical time period by a second preset time length. The single-view complex product data and the ground distance product data are both data from the satellite revisit period to which the predicted time belongs, obtained through a double-orbit time-series synthetic aperture radar. The pixel distance is the pixel distance of the synthetic aperture radar data obtained through a double-orbit time-series synthetic aperture radar. The first temporal optical change type sequence determination module is used to determine the first temporal optical change type sequence of the target area based on the first optical remote sensing data and through a cloud detection algorithm; The first temporal deformation sequence determination module is used to determine the first temporal deformation sequence of the target region based on the single-view complex product data and pixel distance; The first soil moisture content sequence determination module is used to determine the first soil moisture content sequence of the target area based on the first optical remote sensing data and the ground distance product data; The early warning module is used to input the first time-series rainfall sequence, the second time-series rainfall sequence, the first time-series optical change type sequence, the first time-series deformation sequence, and the first soil moisture content sequence into the geological disaster prediction model, obtain the probability of geological disaster occurrence output by the geological disaster prediction model, and issue an early warning based on the probability of geological disaster occurrence.

[0013] This system acquires the first and second time-series rainfall sequences, optical remote sensing data, single-view complex product data, ground distance product data, and pixel distance of the target area. Based on the optical remote sensing data, it determines the first time-series optical change type sequence of the target area using a cloud detection algorithm; based on the single-view complex product data and pixel distance, it determines the first time-series deformation sequence using preset filtering rules; and based on the optical remote sensing data and ground distance product data, it determines the first soil moisture content sequence of the target area; and then combines the first time-series rainfall sequence, the second time-series rainfall sequence, the first time-series optical change type sequence, and the second time-series complex product data... This application integrates a first-time-series deformation sequence and a first-time-series soil moisture content sequence into a geological disaster prediction model to obtain the probability of geological disaster occurrence output by the model. Based on this probability, early warning is issued. This application constructs a geological disaster prediction model by integrating a first-time-series rainfall sequence, a second-time-series rainfall sequence, optical remote sensing data, single-view complex product data, and ground distance product data, thereby achieving accurate early warning of geological disasters. This application breaks through the data silos and mechanism deficiencies of traditional technologies, provides a brand-new solution for disaster prevention and mitigation in complex geological environments, and improves the accuracy of geological disaster early warning.

[0014] A third aspect of this application provides a geological disaster early warning electronic device, including at least one control processor and a memory for communicatively connecting to the at least one control processor; the memory stores instructions executable by the at least one control processor, which are executed by the at least one control processor to enable the at least one control processor to perform the above-described geological disaster early warning method.

[0015] In a fourth aspect, this application provides a computer-readable storage medium storing computer-executable instructions for causing a computer to execute the aforementioned geological disaster early warning method.

[0016] It should be noted that the beneficial effects of the second to fourth aspects of this application with respect to the prior art are the same as the beneficial effects of the geological disaster early warning system described above with respect to the prior art, and will not be described in detail here.

[0017] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0018] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a flowchart illustrating an embodiment of the geological disaster early warning method provided in this application; Figure 2This is a schematic diagram of an embodiment of the geological disaster early warning system provided in this application; Figure 3 This is a schematic diagram of the structure of an embodiment of the electronic device provided in this application. Detailed Implementation

[0019] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0020] In the description of this application, the use of terms such as "first," "second," etc., is for the purpose of distinguishing technical features only and should not be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated or the order of the technical features indicated.

[0021] In the description of this application, it should be understood that the orientation descriptions, such as up, down, etc., are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0022] In the description of this application, it should be noted that, unless otherwise explicitly defined, terms such as "setup," "installation," and "connection" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this application in conjunction with the specific content of the technical solution.

[0023] Currently, geological disaster early warning methods mainly include optical remote sensing technology, time-series synthetic aperture radar interferometry (TISAR) technology, and hydrological monitoring data. While optical remote sensing can identify potential disaster hazards through changes in land cover, it is greatly affected by cloud and rain weather and cannot capture deep surface deformation and soil hydrological conditions. TSAR technology can monitor surface deformation at the millimeter level, but it struggles to directly correlate deformation with triggering factors such as rainfall and soil moisture content. Although hydrological monitoring data reflects changes in surface hydrological conditions, it lacks dynamic coupling analysis with geological deformation, making it difficult to form effective synergy among different data sources. Therefore, breaking down data barriers and integrating multi-source heterogeneous data has become a key breakthrough point for geological disaster early warning.

[0024] To address the aforementioned technical deficiencies, embodiments of this application provide a geological disaster early warning method, system, device, and storage medium.

[0025] Please see Figure 1This is a flowchart illustrating a geological disaster early warning method provided in an embodiment of this application. The method is applied to an electronic device, which may be a server, etc. Figure 1 As shown, this geological disaster early warning method includes: Step S101: Obtain the first time-series rainfall sequence, the second time-series rainfall sequence, the first optical remote sensing data, the single-view complex product data, the ground distance product data, and the pixel distance for the target area. The first time-series rainfall sequence is a rainfall sequence composed of the total rainfall of each first time period, which is obtained by dividing the first preset historical time period according to the first preset time length. The second time-series rainfall sequence is a rainfall sequence composed of the total rainfall of each second time period, which is obtained by dividing the second preset historical time period according to the second preset time length. The single-view complex product data and the ground distance product data are data of the satellite revisit period to which the predicted time belongs, obtained by using a double-orbit time-series synthetic aperture radar. The pixel distance is the pixel distance of the synthetic aperture radar data obtained by using a double-orbit time-series synthetic aperture radar. Step S102: Based on the first optical remote sensing data, determine the first temporal optical change type sequence of the target area using a cloud detection algorithm; Step S103: Based on single-view complex product data and pixel distance, determine the first temporal deformation sequence of the target area; Step S104: Determine the first soil moisture content sequence of the target area based on the first optical remote sensing data and ground distance product data; Step S105: Input the first time-series rainfall sequence, the second time-series rainfall sequence, the first time-series optical change type sequence, the first time-series deformation sequence, and the first soil moisture content sequence into the geological disaster prediction model to obtain the geological disaster occurrence probability output by the geological disaster prediction model, and issue an early warning based on the geological disaster occurrence probability.

[0026] The aforementioned first preset historical time period can be a time length value preset according to actual needs.

[0027] The aforementioned second preset historical time period can be a time length value preset according to actual needs.

[0028] The aforementioned first preset time length can be a time length value preset according to actual needs.

[0029] The aforementioned second preset time length can be a time length value preset according to actual needs.

[0030] The aforementioned single-view multiple product data may include multi-period image data.

[0031] The aforementioned first optical remote sensing data may include multiple periods of optical remote sensing data.

[0032] The above-mentioned first time-series rainfall sequence can be a rainfall sequence composed of the total rainfall of each first time period, arranged in chronological order.

[0033] The aforementioned second time-series rainfall sequence can be a rainfall sequence composed of the total rainfall in each second time period, arranged in chronological order.

[0034] The above-mentioned first temporal optical change type sequence can be optical change types arranged in chronological order. The optical change types can include, but are not limited to, -1, 1, 2 and 3. Among them, -1 can indicate that there is cloud cover in the target area, 1 can indicate that there are fallen trees in the target area, 2 can indicate that there are ground cracks in the target area, and 3 can indicate that there are slipped trees in the target area.

[0035] The aforementioned first temporal deformation sequence may include, but is not limited to, a first temporal deformation variable sequence and a first temporal deformation rate sequence. The aforementioned first temporal deformation sequence may be a temporal deformation sequence arranged in chronological order.

[0036] The first soil moisture content sequence mentioned above can be a soil moisture content sequence arranged in chronological order.

[0037] In step S105 above, early warning is given based on the probability of geological disasters. This can be done by sending an SMS notification to the user when the probability of a geological disaster is greater than a preset early warning threshold. The preset early warning threshold can be a constant value that is pre-set according to actual needs.

[0038] This method acquires the first and second time-series rainfall sequences, optical remote sensing data, single-view complex product data, ground distance product data, and pixel distance of the target area. Based on the optical remote sensing data, it determines the first time-series optical change type sequence of the target area using a cloud detection algorithm; based on the single-view complex product data and pixel distance, it determines the first time-series deformation sequence using a preset filtering rule; and based on the optical remote sensing data and ground distance product data, it determines the first soil moisture content sequence of the target area. The method then combines the first time-series rainfall sequence, the second time-series rainfall sequence, the first time-series optical change type sequence, and the second time-series complex product data... This application integrates a first-time-series deformation sequence and a first-time-series soil moisture content sequence into a geological disaster prediction model to obtain the probability of geological disaster occurrence output by the model. Based on this probability, early warning is issued. This application constructs a geological disaster prediction model by integrating a first-time-series rainfall sequence, a second-time-series rainfall sequence, optical remote sensing data, single-view complex product data, and ground distance product data, thereby achieving accurate early warning of geological disasters. This application breaks through the data silos and mechanism deficiencies of traditional technologies, provides a brand-new solution for disaster prevention and mitigation in complex geological environments, and improves the accuracy of geological disaster early warning. In some embodiments, step S102 may include, but is not limited to, steps S201 to S205: Step S201: Divide the first optical remote sensing data into equal parts according to the first preset time interval to obtain the second optical remote sensing data; Step S202: Traverse all second optical remote sensing data, filter all second optical remote sensing data with cloud obstruction using cloud detection algorithm, use them as third optical remote sensing data, and mark the third optical remote sensing data with the first preset classification mark symbol to obtain the first marked optical remote sensing data; Step S203: Determine the fourth optical remote sensing data, wherein the fourth optical remote sensing data is all the other optical remote sensing data in the second optical remote sensing data except for the third optical remote sensing data; Step S204: Input the fourth optical remote sensing data into the temporal optical change classification network model so that the temporal optical change classification network model uses the second preset classification label symbol to label each fourth optical remote sensing data, and obtain the second-labeled optical remote sensing data, wherein the second preset classification label symbol is different from the first preset classification label symbol; Step S205: Determine the first temporal optical change type sequence based on the first labeled optical remote sensing data and the second labeled optical remote sensing data.

[0039] The first preset classification marker symbol mentioned above can be -1.

[0040] The aforementioned second preset classification marker symbol can be 1, 2, or 3.

[0041] This application uses a time-series optical change classification network model to label each fourth optical remote sensing data point, thereby achieving the purpose of classifying each fourth optical remote sensing data point, providing data basis for subsequent geological disaster early warning, and improving the accuracy of geological disaster early warning.

[0042] In some embodiments, step S205 may include, but is not limited to, steps S301 to S302: Step S301: Divide each first-marked optical remote sensing data into equal segments according to a second preset time interval to obtain a first segmented optical remote sensing sequence for each first-marked optical remote sensing data; divide each second-marked optical remote sensing data into equal segments according to a second preset time interval to obtain a second segmented optical remote sensing sequence for each second-marked optical remote sensing data, wherein the classification label symbol of each first-marked optical remote sensing data in the first segmented optical remote sensing sequence is the same as the classification label symbol of the first-marked optical remote sensing data with the closest time interval, and the classification label symbol of each second-marked optical remote sensing data in the second segmented optical remote sensing sequence is the same as the classification label symbol of the second-marked optical remote sensing data with the closest time interval. Step S302: Combine all the first segmented optical remote sensing sequences and all the second segmented optical remote sensing sequences to obtain the first temporal optical change type sequence.

[0043] The aforementioned second preset time interval can be a time interval value preset according to actual needs.

[0044] Specifically, in step S301 above, each first-marked optical remote sensing data is uniformly divided according to a second preset time interval to obtain the first segmented optical remote sensing sequence of each first-marked optical remote sensing data. and The time interval between them is divided into equal parts according to the second preset time interval. Segment, of which, 0 to The values ​​within the segment are all equal to The value, arrive The value of each segment is equal to The value of is used to obtain the first segmented optical remote sensing sequence, where For the first The first marked optical remote sensing value, For the first The first marked optical remote sensing value, These are constant values ​​that are pre-set according to actual needs.

[0045] The process of dividing each second-marked optical remote sensing data into a second segmented optical remote sensing sequence by uniformly dividing each second-marked optical remote sensing data into a second preset time interval is similar to the process of dividing each first-marked optical remote sensing data into a first segmented optical remote sensing sequence by uniformly dividing each first-marked optical remote sensing data into a second preset time interval, and will not be described in detail here.

[0046] This application obtains a first temporal optical change type sequence by combining all first-segmented optical remote sensing sequences and all second-segmented optical remote sensing sequences, providing data support for subsequent geological disaster early warning and improving the accuracy of geological disaster early warning.

[0047] In some embodiments, step S103 may include, but is not limited to, steps S401 to S413: Step S401: Based on the preset time baseline threshold and the preset spatial baseline threshold, perform image pair filtering on the single-view multiple product data to obtain the image pair set and the date of each image pair; Step S402: Construct an image pair index sequence for the image pair set based on the date of each image pair; Step S403: Based on the image pair index sequence, perform image registration on the image pair set to obtain the first registered image pair sequence; Step S404: Based on the first registered image pair sequence, determine the phase-unwrapped image pair sequence using the phase unwrapping method; Step S405: Based on the phase-unwrapped image pair sequence, determine the second temporal deformation sequence using the singular value decomposition method; Step S406: Based on the first filter preset value, filter the second time-series deformation sequence to obtain the third time-series deformation sequence; Step S407: Calculate the average coherence of each period's image data for the single-view complex product data; Step S408: Traverse the image data of each period, select the image data with the maximum average coherence as the master image data, and perform image registration with the master image data for the remaining image data to obtain the second registered image pair sequence. Step S409: Perform differential interferometry on the second registered image pair sequence to obtain differential interferometry data; Step S410: Filter the differential interferometric data using the amplitude deviation index and a preset screening threshold to obtain a permanent scatterer sequence; Step S411: Calculate the deformation rate of the permanent scatterer sequence using the least squares method to obtain the fourth time-series deformation sequence; Step S412: Based on the second filter preset value, filter the fourth time-series deformation sequence to obtain the fifth time-series deformation sequence; Step S413: Determine the first temporal deformation sequence based on the third temporal deformation sequence, the fifth temporal deformation sequence, and the pixel distance.

[0048] The aforementioned preset time baseline threshold is a value set in advance according to actual needs, and can be 60 days.

[0049] The aforementioned preset spatial baseline threshold is a value pre-set according to actual needs, and can be 0.02.

[0050] The aforementioned single-view multiple-view product data includes data from multiple film and television episodes.

[0051] In step S401 above, based on the preset time baseline threshold and the preset spatial baseline threshold, image pairs are filtered for single-view multiple product data to obtain the image pair set and the date of each image pair. This can be done by pairing all period film and television data, calculating the time difference and spatial difference of the paired image combinations, filtering out image pairs with a time difference less than the preset time baseline threshold and a spatial difference less than the preset spatial baseline threshold, and combining all the filtered image pairs to obtain the image pair set and the date of each image pair.

[0052] In step S402 above, the image pair index sequence of the image pair set can be constructed based on the date of each image pair, which can be the date of each image pair as the corresponding image pair index.

[0053] In step S403 above, image registration is performed on the image pair set based on the image pair index sequence to obtain the first registered image pair sequence. This can be achieved by sequentially selecting image pairs in the image pair set according to the order of the image pair index sequence and performing image registration.

[0054] In step S404 above, the determination of the phase-unwrapped image pair sequence based on the first registered image pair sequence may include, but is not limited to, steps S4041 to S4045: Step S4041: Perform differential interferometry on the first registered image pair sequence to obtain a differential interferometric image pair sequence; Step S4042: De-flatten the differential interferometric image pair sequence to obtain the de-flattened image pair sequence; Step S4043: Perform terrain phase removal on the image pair sequence after flattening to obtain the image pair sequence after terrain phase removal; Step S4044: Perform differential interferogram filtering on the topographic phase-degraded image pair sequence to obtain the differential interferogram-filtered image pair sequence; Step S4045: Perform phase unwrapping on the image pair sequence after differential interferogram filtering to obtain the phase unwrapped image pair sequence.

[0055] In step S4041 above, differential interferometry is performed on the first registered image pair sequence to obtain a differential interferometric image pair sequence. This can be achieved by performing differential interferometry on the first registered image pair sequence using the two-track D-InSAR method.

[0056] In step S4042 above, the differential interferometric image pair sequence is de-flattened to obtain the de-flattened image pair sequence. This can be achieved by de-flattening the differential interferometric image pair sequence using the orbital parameter method.

[0057] In step S4043 above, the terrain phase is removed from the image pair sequence after the flattening is removed to obtain the terrain phase removed image pair sequence. This can be done by simulating the terrain phase using external DEM data.

[0058] In step S4044 above, differential interferogram filtering is performed on the topographic phase-removed image pair sequence to obtain a differential interferogram-filtered image pair sequence. This can be achieved by performing differential interferogram filtering on the topographic phase-removed image pair sequence using the Goldstein filtering method.

[0059] In step S4045 above, the phase unwrapping of the differential interferogram filtered image pair sequence is performed to obtain the phase unwrapped image pair sequence. This can be obtained by performing phase unwrapping of the differential interferogram filtered image pair sequence using the minimum cost flow method.

[0060] In step S403 above, image registration is performed on the image pair set based on the image pair index sequence to obtain the first registered image pair sequence. This can be achieved by sequentially selecting image pairs in the image pair set according to the order of the image pair index sequence and performing image registration.

[0061] In step S405 above, based on the phase-unwrapped image pair sequence, the second temporal deformation sequence is determined by the singular value decomposition method. This can be achieved by establishing the phase equation corresponding to the phase-unwrapped image pair sequence, and then solving the phase equation by the singular value decomposition method to obtain the second temporal deformation sequence.

[0062] The aforementioned first filter preset value includes a first high-pass filter preset value and a first low-pass filter preset value. The first high-pass filter preset value can be 365 days, and the first low-pass filter preset value can be 2000 meters.

[0063] In step S406 above, the second time-series deformation sequence is filtered based on the first preset filter value to obtain the third time-series deformation sequence. This can be achieved by performing a high-pass filter on the second time-series deformation sequence using the first preset high-pass filter value to obtain the second high-pass filtered time-series deformation sequence, and then performing a low-pass filter on the second high-pass filtered time-series deformation sequence using the first preset low-pass filter value to obtain the third time-series deformation sequence.

[0064] The above preset filtering thresholds are values ​​that are pre-set according to actual needs.

[0065] The aforementioned second filter preset value includes a second high-pass filter preset value and a second low-pass filter preset value. The second high-pass filter preset value can be 365 days, and the second low-pass filter preset value can be 1200 meters.

[0066] The aforementioned second filter preset value is a value that is preset according to actual needs.

[0067] In step S412 above, the fourth time-series deformation sequence is filtered based on the second filter preset value to obtain the fifth time-series deformation sequence. This can be achieved by performing a high-pass filter on the fourth time-series deformation sequence using the second high-pass filter preset value to obtain the fourth high-pass filtered time-series deformation sequence, and then performing a low-pass filter on the fourth high-pass filtered time-series deformation sequence using the second low-pass filter preset value to obtain the fifth time-series deformation sequence.

[0068] In some embodiments, step S413 may include, but is not limited to, steps S501 to S506: Step S501: Calculate the distance between each point in the third time-series deformation sequence and each point in the fifth time-series deformation sequence in the preset coordinate system. The time-series deformation sequence includes several time-series deformation values ​​and the latitude and longitude corresponding to each time-series deformation value. The points in the time-series deformation sequence are obtained by projecting the latitude and longitude corresponding to each time-series deformation value onto the preset coordinate system. Step S502: Traverse each point in the third temporal deformation sequence, and filter out points in the third temporal deformation sequence whose distance from any point in the fifth temporal deformation sequence is less than a preset distance threshold, to obtain the same-name points in the third temporal deformation sequence, wherein the preset distance threshold is one-tenth of the pixel distance; Step S503: Determine the sixth time series deformation sequence, wherein the sixth time series deformation sequence is a sequence composed of all points other than the corresponding points in the third time series deformation sequence; Step S504: Combine the fifth and sixth time-series deformation sequences to obtain the seventh time-series deformation sequence; Step S505: Divide the seventh time-series deformation sequence evenly according to the third preset time interval to obtain the eighth time-series deformation sequence; divide the seventh time-series deformation sequence evenly according to the fourth preset time interval to obtain the ninth time-series deformation sequence. Step S506: Perform linear fitting based on the eighth time-series deformation sequence and the ninth time-series deformation sequence to obtain the first time-series deformation sequence.

[0069] The aforementioned preset coordinate system can be a geodetic coordinate system.

[0070] The aforementioned third preset time interval is a value preset according to actual needs.

[0071] The aforementioned fourth preset time interval is a value preset according to actual needs.

[0072] Specifically, in step S506 above, a linear fit is performed based on the eighth and ninth time-series deformation sequences to obtain the first time-series deformation sequence, which can be used for... ,exist The date is less than The date is less than In the case of the date, establish The linear fitting equation, and through and( and The number of days between intervals, To solve the linear fitting equation. Then and The number of days between intervals is used as Inputting it into the solved linear fitting equation, the calculation yields... The values ​​are then applied sequentially to obtain the first temporal deformation sequence, where, The eighth time-series deformation sequence The eighth time-series deformation value. For the ninth time-series deformation sequence The ninth time-series deformation value. The ninth time-series deformation sequence The ninth time-series deformation value.

[0073] In some embodiments, step S104 may include, but is not limited to, steps S601 to S607: Step S601: Extract the homopolarization data and crosspolarization data of the ground distance product data; Step S602: Filter the same polarization data to obtain filtered same polarization data; filter the cross polarization data to obtain filtered cross polarization data; Step S603: Geocode the filtered homopolarized data to obtain encoded homopolarized data; geocode the filtered cross-polarized data to obtain encoded cross-polarized data. Step S604: Perform radiometric calibration on the encoded co-polarized data to obtain the co-polarized backscattering coefficients; perform radiometric calibration on the encoded cross-polarized data to obtain the cross-polarized backscattering coefficients. Step S605: Input the same polarization backscattering coefficient into the water cloud model to obtain the first backscattering coefficient output by the water cloud model; input the cross polarization backscattering coefficient into the water cloud model to obtain the second backscattering coefficient output by the water cloud model. Step S606: Calculate the normalized vegetation index of the first optical remote sensing data; Step S607: Determine the first soil moisture content sequence of the target area based on the first backscattering coefficient, the second backscattering coefficient, and the normalized vegetation index.

[0074] This application calculates the first backscattering coefficient, the second backscattering coefficient, and the normalized vegetation index, and then uses the first backscattering coefficient, the second backscattering coefficient, and the normalized vegetation index to determine the first soil moisture content sequence of the target area, providing data basis for subsequent geological disaster early warning and improving the accuracy of geological disaster early warning.

[0075] In some embodiments, step S607 may include, but is not limited to, steps S701 to S703: Step S701: Input the first backscattering coefficient, the second backscattering coefficient, and the normalized vegetation index into the soil moisture content prediction model to obtain the first soil moisture content prediction sequence of the target area output by the soil moisture content prediction model. Step S702: Divide the first soil moisture content prediction sequence evenly according to the fourth preset time interval to obtain the second soil moisture content prediction sequence; divide the first soil moisture content prediction sequence evenly according to the fifth preset time interval to obtain the third soil moisture content prediction sequence. Step S703: Perform linear fitting based on the second and third soil moisture content prediction sequences to obtain the first soil moisture content sequence.

[0076] In step S703, the linear fitting method for obtaining the first soil moisture content sequence based on the second and third soil moisture content prediction sequences is similar to the linear fitting method for obtaining the first time-series deformation sequence based on the eighth and ninth time-series deformation sequences in step S506 above, and will not be described again here.

[0077] This application uses a linear fitting method to calculate the first soil moisture content sequence, thereby improving the accuracy of the first soil moisture content sequence.

[0078] Additionally, refer to Figure 2 One embodiment of this application provides a geological disaster early warning system, including a data acquisition module 1100, a first time-series optical change type sequence determination module 1200, a first time-series deformation sequence determination module 1300, a first soil moisture content sequence determination module 1400, and an early warning module 1500, wherein: The data acquisition module 1100 is used to acquire the first time-series rainfall sequence, the second time-series rainfall sequence, the first optical remote sensing data, the single-view complex product data, the ground distance product data, and the pixel distance of the target area. The first time-series rainfall sequence is a rainfall sequence composed of the total rainfall of each first time period, which is obtained by dividing the first preset historical time period according to the first preset time length. The second time-series rainfall sequence is a rainfall sequence composed of the total rainfall of each second time period, which is obtained by dividing the second preset historical time period according to the second preset time length. The single-view complex product data and the ground distance product data are both data of the satellite revisit period to which the predicted time belongs, obtained by the double-orbit time-series synthetic aperture radar. The pixel distance is the pixel distance of the synthetic aperture radar data obtained by the double-orbit time-series synthetic aperture radar. The first temporal optical change type sequence determination module 1200 is used to determine the first temporal optical change type sequence of the target area based on the first optical remote sensing data and through a cloud detection algorithm. The first temporal deformation sequence determination module 1300 is used to determine the first temporal deformation sequence of the target area based on single-view complex product data and pixel distance; The first soil moisture content sequence determination module 1400 is used to determine the first soil moisture content sequence of the target area based on the first optical remote sensing data and the ground distance product data; The early warning module 1500 is used to input the first time-series rainfall sequence, the second time-series rainfall sequence, the first time-series optical change type sequence, the first time-series deformation sequence, and the first soil moisture content sequence into the geological disaster prediction model, obtain the probability of geological disaster occurrence output by the geological disaster prediction model, and issue an early warning based on the probability of geological disaster occurrence.

[0079] This system acquires the first and second time-series rainfall sequences, optical remote sensing data, single-view complex product data, ground distance product data, and pixel distance of the target area. Based on the optical remote sensing data, it determines the first time-series optical change type sequence of the target area using a cloud detection algorithm; based on the single-view complex product data and pixel distance, it determines the first time-series deformation sequence using preset filtering rules; and based on the optical remote sensing data and ground distance product data, it determines the first soil moisture content sequence of the target area; and then combines the first time-series rainfall sequence, the second time-series rainfall sequence, the first time-series optical change type sequence, and the second time-series complex product data... This application integrates a first-time-series deformation sequence and a first-time-series soil moisture content sequence into a geological disaster prediction model to obtain the probability of geological disaster occurrence output by the model. Based on this probability, early warning is issued. This application constructs a geological disaster prediction model by integrating a first-time-series rainfall sequence, a second-time-series rainfall sequence, optical remote sensing data, single-view complex product data, and ground distance product data, thereby achieving accurate early warning of geological disasters. This application breaks through the data silos and mechanism deficiencies of traditional technologies, provides a brand-new solution for disaster prevention and mitigation in complex geological environments, and improves the accuracy of geological disaster early warning.

[0080] It should be noted that the system embodiments described above are based on the same inventive concept as the method embodiments described above. Therefore, the relevant content of the method embodiments described above is also applicable to the system embodiments described above, and will not be repeated here.

[0081] Figure 3 A schematic diagram of the hardware structure for geological disaster early warning provided in an embodiment of this application is shown.

[0082] The geological disaster early warning equipment may include a processor 301 and a memory 302 storing computer program instructions.

[0083] Specifically, the processor 301 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0084] Memory 302 may include mass storage for data or instructions. For example, and not limitingly, memory 302 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 302 may include removable or non-removable (or fixed) media. Where appropriate, memory 302 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 302 is non-volatile solid-state memory.

[0085] In some embodiments, memory 302 may include read-only memory (ROM), random access memory (RAM), disk storage media device, optical storage media device, flash memory device, electrical, optical, or other physical / tangible memory storage device. Thus, generally, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to one aspect of this disclosure.

[0086] The processor 301 reads and executes computer program instructions stored in the memory 302 to implement any of the geological disaster early warning methods in the above embodiments.

[0087] In one example, the geological disaster early warning device may also include a communication interface 303 and a bus 310. For example, Figure 3 As shown, the processor 301, memory 302, and communication interface 303 are connected through bus 310 and complete communication with each other.

[0088] The communication interface 303 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0089] Bus 310 includes hardware, software, or both, that couples components of a geological disaster early warning device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 310 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, any suitable bus or interconnect is contemplated herein.

[0090] This geological disaster early warning device can execute the geological disaster early warning method in the embodiments of this application based on a three-dimensional design model, thereby achieving a combination of... Figure 1 and Figure 2 Describes geological disaster early warning methods and systems.

[0091] Furthermore, in conjunction with the geological disaster early warning methods in the above embodiments, this application embodiment can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the geological disaster early warning methods in the above embodiments.

[0092] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0093] The functional blocks shown in the above block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0094] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0095] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0096] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A geological disaster early warning method, characterized in that, The geological disaster early warning method includes: The system acquires a first time-series rainfall sequence, a second time-series rainfall sequence, first optical remote sensing data, single-view complex product data, ground distance product data, and pixel distance for the target area. The first time-series rainfall sequence consists of the total rainfall for each first time period, which is obtained by dividing a first preset historical time period by a first preset time length. The second time-series rainfall sequence consists of the total rainfall for each second time period, which is obtained by dividing a second preset historical time period by a second preset time length. The single-view complex product data and the ground distance product data are both data from the satellite revisit period to which the predicted time belongs, obtained through a double-orbit time-series synthetic aperture radar. The pixel distance is the pixel distance of the synthetic aperture radar data obtained through the double-orbit time-series synthetic aperture radar. Based on the first optical remote sensing data, a first temporal optical change type sequence of the target area is determined using a cloud detection algorithm, specifically as follows: The first optical remote sensing data is evenly divided according to a first preset time interval to obtain the second optical remote sensing data. Traverse all the second optical remote sensing data, filter all the second optical remote sensing data with cloud obstruction using the cloud detection algorithm, use them as the third optical remote sensing data, and mark the third optical remote sensing data with the first preset classification label to obtain the first-labeled optical remote sensing data; Determine the fourth optical remote sensing data, wherein the fourth optical remote sensing data is all other optical remote sensing data in the second optical remote sensing data except for the third optical remote sensing data; The fourth optical remote sensing data is input into the temporal optical change classification network model so that the temporal optical change classification network model uses a second preset classification label symbol to label each of the fourth optical remote sensing data, thereby obtaining the second-labeled optical remote sensing data, wherein the second preset classification label symbol is different from the first preset classification label symbol; The first temporal optical change type sequence is determined based on the first labeled optical remote sensing data and the second labeled optical remote sensing data, specifically as follows: Each first-labeled optical remote sensing data is uniformly divided according to a second preset time interval to obtain a first segmented optical remote sensing sequence for each first-labeled optical remote sensing data; each second-labeled optical remote sensing data is uniformly divided according to the second preset time interval to obtain a second segmented optical remote sensing sequence for each second-labeled optical remote sensing data, wherein the classification label symbol of each first-segmented optical remote sensing data in the first segmented optical remote sensing sequence is the same as the classification label symbol of the first-labeled optical remote sensing data with the closest time distance, and the classification label symbol of each second-segmented optical remote sensing data in the second segmented optical remote sensing sequence is the same as the classification label symbol of the second-labeled optical remote sensing data with the closest time distance; By combining all the first segmented optical remote sensing sequences and all the second segmented optical remote sensing sequences, the first temporal optical change type sequence is obtained; Based on the single-view complex product data and pixel distance, a first temporal deformation sequence of the target region is determined; Based on the first optical remote sensing data and the ground distance product data, a first soil moisture content sequence for the target area is determined; The first time-series rainfall sequence, the second time-series rainfall sequence, the first time-series optical change type sequence, the first time-series deformation sequence, and the first soil moisture content sequence are input into the geological disaster prediction model to obtain the probability of geological disaster occurrence output by the geological disaster prediction model, and early warning is issued based on the probability of geological disaster occurrence.

2. The geological disaster early warning method according to claim 1, characterized in that, The single-view complex product data includes multi-period image data. Determining the first temporal deformation sequence of the target region based on the single-view complex product data and pixel distance includes: Based on preset time baseline thresholds and preset spatial baseline thresholds, image pairs are filtered on the single-view complex product data to obtain a set of image pairs and the date of each image pair; Construct an image pair index sequence for the image pair set based on the date of each image pair; Based on the image pair index sequence, image registration is performed on the image pair set to obtain the first registered image pair sequence; Based on the first registered image pair sequence, the phase-unwrapped image pair sequence is determined by the phase unwrapping method; Based on the phase-unwrapped image pair sequence, the second temporal deformation sequence is determined by the singular value decomposition method; Based on the first preset filtering value, the second time-series deformation sequence is filtered to obtain the third time-series deformation sequence; Calculate the average coherence of each period's image data of the single-view complex product data; The image data of each period is traversed, and the image data with the maximum value of the average coherence is selected as the master image data. The remaining image data are all image registered with the master image data to obtain the second registered image pair sequence. Differential interferometry is performed on the second registered image pair sequence to obtain differential interferometry data; The differential interferometric data are filtered using the amplitude deviation index and a preset screening threshold to obtain a permanent scatterer sequence; The deformation rate of the permanent scatterer sequence was calculated using the least squares method to obtain the fourth time-series deformation sequence; Based on the second preset filtering value, the fourth time-series deformation sequence is filtered to obtain the fifth time-series deformation sequence; The first temporal deformation sequence is determined based on the third temporal deformation sequence, the fifth temporal deformation sequence, and the pixel distance.

3. The geological disaster early warning method according to claim 2, characterized in that, Determining the first temporal deformation sequence based on the third temporal deformation sequence, the fifth temporal deformation sequence, and the pixel distance includes: Calculate the distance between each point in the third temporal deformation sequence and each point in the fifth temporal deformation sequence in the preset coordinate system. The temporal deformation sequence includes several temporal deformation values ​​and the latitude and longitude corresponding to each temporal deformation value. The points in the temporal deformation sequence are obtained by projecting the latitude and longitude corresponding to each temporal deformation value onto the preset coordinate system. Traverse each point in the third temporal deformation sequence, and filter out points in the third temporal deformation sequence whose distance from any point in the fifth temporal deformation sequence is less than a preset distance threshold to obtain points with the same name in the third temporal deformation sequence, wherein the preset distance threshold is one-tenth of the pixel distance; A sixth temporal deformation sequence is determined, wherein the sixth temporal deformation sequence is a sequence composed of all points in the third temporal deformation sequence except for the points with the same name in the third temporal deformation sequence; the fifth temporal deformation sequence and the sixth temporal deformation sequence are combined to obtain a seventh temporal deformation sequence; The seventh time-series deformation sequence is uniformly divided according to the third preset time interval to obtain the eighth time-series deformation sequence; the seventh time-series deformation sequence is uniformly divided according to the fourth preset time interval to obtain the ninth time-series deformation sequence. The first time-series deformation sequence is obtained by linear fitting based on the eighth time-series deformation sequence and the ninth time-series deformation sequence.

4. The geological disaster early warning method according to claim 1, characterized in that, Determining the first soil moisture content sequence of the target area based on the first optical remote sensing data and the ground distance product data includes: Extract the homopolarization and crosspolarization data of the ground distance product data; The same-polarization data is filtered to obtain filtered same-polarization data; the cross-polarization data is filtered to obtain filtered cross-polarization data. Geocoding is performed on the filtered homopolarized data to obtain encoded homopolarized data; geocoding is performed on the filtered cross-polarized data to obtain encoded cross-polarized data. Radiometric calibration is performed on the encoded homopolarized data to obtain homopolarized backscattering coefficients; radiometric calibration is performed on the encoded crosspolarized data to obtain crosspolarized backscattering coefficients. The same polarization backscattering coefficient is input into the water cloud model to obtain the first backscattering coefficient output by the water cloud model; the cross polarization backscattering coefficient is input into the water cloud model to obtain the second backscattering coefficient output by the water cloud model. Calculate the normalized vegetation index of the first optical remote sensing data; The first soil moisture content sequence of the target area is determined based on the first backscattering coefficient, the second backscattering coefficient, and the normalized vegetation index.

5. A geological disaster early warning method according to claim 4, characterized in that, The determination of the first soil moisture content sequence of the target area based on the first backscattering coefficient, the second backscattering coefficient, and the normalized vegetation index includes: The first backscattering coefficient, the second backscattering coefficient, and the normalized vegetation index are input into the soil moisture content prediction model to obtain the first soil moisture content prediction sequence of the target area output by the soil moisture content prediction model. The first soil moisture content prediction sequence is uniformly divided according to a fourth preset time interval to obtain a second soil moisture content prediction sequence; the first soil moisture content prediction sequence is uniformly divided according to a fifth preset time interval to obtain a third soil moisture content prediction sequence. The first soil moisture content sequence is obtained by linear fitting based on the second and third soil moisture content prediction sequences.

6. A geological disaster early warning system, characterized in that, The geological disaster early warning system includes: The data acquisition module is used to acquire a first time-series rainfall sequence, a second time-series rainfall sequence, first optical remote sensing data, single-view complex product data, ground distance product data, and pixel distance for a target area. The first time-series rainfall sequence is a rainfall sequence composed of the total rainfall for each first time period, which is obtained by dividing a first preset historical time period by a first preset time length. The second time-series rainfall sequence is a rainfall sequence composed of the total rainfall for each second time period, which is obtained by dividing a second preset historical time period by a second preset time length. The single-view complex product data and the ground distance product data are both data from the satellite revisit period to which the predicted time belongs, obtained through a double-orbit time-series synthetic aperture radar. The pixel distance is the pixel distance of the synthetic aperture radar data obtained through a double-orbit time-series synthetic aperture radar. The first temporal optical change type sequence determination module is used to determine the first temporal optical change type sequence of the target area based on the first optical remote sensing data and through a cloud detection algorithm, specifically: The first optical remote sensing data is evenly divided according to a first preset time interval to obtain the second optical remote sensing data. Traverse all the second optical remote sensing data, filter all the second optical remote sensing data with cloud obstruction using the cloud detection algorithm, use them as the third optical remote sensing data, and mark the third optical remote sensing data with the first preset classification label to obtain the first-labeled optical remote sensing data; Determine the fourth optical remote sensing data, wherein the fourth optical remote sensing data is all other optical remote sensing data in the second optical remote sensing data except for the third optical remote sensing data; The fourth optical remote sensing data is input into the temporal optical change classification network model so that the temporal optical change classification network model uses a second preset classification label symbol to label each of the fourth optical remote sensing data, thereby obtaining the second-labeled optical remote sensing data, wherein the second preset classification label symbol is different from the first preset classification label symbol; The first temporal optical change type sequence is determined based on the first labeled optical remote sensing data and the second labeled optical remote sensing data, specifically as follows: Each first-labeled optical remote sensing data is uniformly divided according to a second preset time interval to obtain a first segmented optical remote sensing sequence for each first-labeled optical remote sensing data; each second-labeled optical remote sensing data is uniformly divided according to the second preset time interval to obtain a second segmented optical remote sensing sequence for each second-labeled optical remote sensing data, wherein the classification label symbol of each first-segmented optical remote sensing data in the first segmented optical remote sensing sequence is the same as the classification label symbol of the first-labeled optical remote sensing data with the closest time distance, and the classification label symbol of each second-segmented optical remote sensing data in the second segmented optical remote sensing sequence is the same as the classification label symbol of the second-labeled optical remote sensing data with the closest time distance; By combining all the first segmented optical remote sensing sequences and all the second segmented optical remote sensing sequences, the first temporal optical change type sequence is obtained; The first temporal deformation sequence determination module is used to determine the first temporal deformation sequence of the target region based on the single-view complex product data and pixel distance; The first soil moisture content sequence determination module is used to determine the first soil moisture content sequence of the target area based on the first optical remote sensing data and the ground distance product data; The early warning module is used to input the first time-series rainfall sequence, the second time-series rainfall sequence, the first time-series optical change type sequence, the first time-series deformation sequence, and the first soil moisture content sequence into the geological disaster prediction model, obtain the probability of geological disaster occurrence output by the geological disaster prediction model, and issue an early warning based on the probability of geological disaster occurrence.

7. A geological disaster early warning device, characterized in that, It includes at least one control processor and a memory for communicatively connecting to the at least one control processor; the memory stores instructions executable by the at least one control processor, which, when executed by the at least one control processor, enable the at least one control processor to perform a geological disaster early warning method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions for causing a computer to perform a geological disaster early warning method as described in any one of claims 1 to 5.

Images