Water depth inversion method and device, nonvolatile storage medium and electronic device

Abstract

The application discloses a water depth inversion method and device, a nonvolatile storage medium and an electronic device. The method comprises the following steps: acquiring historical remote sensing image data and historical ground water level gauge data in a target area; inputting a second historical radiation brightness value into a physical model to obtain predicted water depth information output by the physical model; correcting the predicted water depth information by using second historical real water depth information, and updating the physical model according to a correction result to obtain a target physical model; and inputting a current radiation brightness value into the target physical model to obtain target predicted water depth information output by the target physical model. The application solves the technical problem that the accuracy of an inversion result is low when a satellite remote sensing technology is used to perform large-range water depth inversion, due to the influence of factors such as clouds, atmosphere and surface types.

Description

Technical Field

[0001] The present application relates to the technical field of water depth quantitative inversion, and in particular to a water depth inversion method and device, a non-volatile storage medium, and an electronic device. Background Art

[0002] In the field of power grid security monitoring, the impact of floods on transmission lines has long been a focus of research and attention. Because transmission lines cover a wide area and are often located in complex terrain, traditional ground-based monitoring methods struggle to achieve real-time monitoring of large-scale flooding. Therefore, remote sensing technology, particularly satellite remote sensing, has gained widespread application in power grid flood monitoring. Satellite remote sensing technology offers advantages such as wide coverage, periodic observations, and the absence of on-site operations, enabling rapid acquisition of surface water information over a wide area.

[0003] Existing technologies for monitoring flood depth in power grids primarily rely on water depth inversion using satellite remote sensing imagery. These methods typically include visible light reflectance models based on optical remote sensing imagery and synthetic aperture radar interferometry (SARI) based on radar remote sensing imagery. While these technologies can provide a certain degree of water depth information for flood disasters, their accuracy for complex terrain and urban areas still needs to be improved. Specifically, existing satellite remote sensing technology is limited in its accuracy when performing large-scale water depth inversion due to the influence of factors such as cloud cover, the atmosphere, and surface type. This results in significant errors, particularly in complex terrain and urban areas.

[0004] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0005] The embodiments of the present application provide a water depth inversion method and device, a non-volatile storage medium, and an electronic device to at least solve the technical problem that when existing satellite remote sensing technology performs large-scale water depth inversion, the inversion results have low accuracy due to the influence of factors such as clouds, atmosphere, and surface type.

[0006] According to one aspect of an embodiment of the present application, a water depth inversion method is provided, including: acquiring historical remote sensing image data and historical ground water level gauge data in a target area, wherein the historical remote sensing image data includes: a first number of first historical radiation brightness values ​​and a second number of second historical radiation brightness values, and the historical ground water level gauge data includes: a third number of first historical real water depth information and a fourth number of second historical real water depth information, wherein the second number is smaller than the first number, and the fourth number is smaller than the third number; determining a physical model for representing the correlation relationship between the first historical radiation brightness value and the first historical real water depth information; inputting the second historical radiation brightness value into the physical model to obtain predicted water depth information output by the physical model; correcting the predicted water depth information using the second historical real water depth information, and updating the physical model according to the correction result to obtain a target physical model; inputting the current radiation brightness value into the target physical model to obtain target predicted water depth information output by the target physical model.

[0007] Optionally, determining a physical model for representing the correlation between the first historical radiation brightness value and the first historical real water depth information includes: obtaining the daily average change value of the solar radiation intensity of the target area within a preset historical time period; if the daily average change value of the solar radiation intensity is within a first preset interval, using a single-band model to determine the correlation relationship, wherein the single-band model is used to determine the linear or nonlinear relationship between the water depth and the reflectivity of the preset band and the radiation brightness value; if the daily average change value of the solar radiation intensity is within a second preset interval, using a dual-band model to determine the correlation relationship, wherein the dual-band model is used to determine the linear or nonlinear relationship between the water depth and the ratio or logarithm of the difference between the reflectivity of two different bands and the radiation brightness value. The linear relationship is determined by the logarithmic ratio model, and the minimum value of the second preset interval is greater than the maximum value of the first preset interval; if the daily average change value of the solar radiation intensity is within the third preset interval, the logarithmic ratio model is used to determine the linear or nonlinear relationship between the natural logarithm of the water depth and the reflectivity ratio and the radiation brightness value, and the minimum value of the third preset interval is greater than the maximum value of the second preset interval; if the daily average change value of the solar radiation intensity is within the fourth preset interval, the multi-band model is used to determine the linear or nonlinear relationship between the water depth and the reflectivity and radiation brightness values ​​of multiple preset bands, and the minimum value of the fourth preset interval is greater than the maximum value of the third preset interval.

[0008] Optionally, the first preset interval is 45W / m 2 Up to 55W / m 2 ; The second preset interval is 100W / m 2 Up to 180W / m 2 ; The third preset interval is 200W / m 2 Up to 220W / m 2 ; The fourth preset interval is 300W / m2 Up to 350W / m 2 .

[0009] Optionally, determining the correlation relationship using a dual-band model includes: determining the correlation relationship using the following formula: Z = (1 / fk i )(lnr Bi -X i ) Where Z is the relative water depth, f is the length of the water path, k i is the effective attenuation coefficient of water, r Bi is the bottom reflectivity at water depth Z, L i is the radiation brightness value at water depth Z, L si is the radiation brightness value at the preset water depth, R i It is a function that represents the relationship between solar radiation value, transmission flux at the interface between atmosphere and water body, and refractive index of water surface.

[0010] Optionally, the predicted water depth information is corrected using the second historical real water depth information, including: dividing the second historical real water depth information into training set data and verification set data; using the training set data to train a polynomial regression model, and using the verification set data to verify the mean square error between the output value of the polynomial regression model and the true value, and when the mean square error is less than a preset threshold, a trained polynomial regression model is obtained; and the predicted water depth information is input into the trained polynomial regression model to obtain a correction result for the predicted water depth information.

[0011] Optionally, the target predicted water depth information includes: multiple discrete water depth measurement points and water depth values ​​corresponding to the water depth measurement points, wherein each water depth measurement point has a geographic coordinate identifier; after obtaining the target predicted water depth information output by the target physical model, the method also includes: converting the geographic coordinate identifier of the water depth measurement point into plane coordinates suitable for visual display; constructing a visual graphic based on the converted plane coordinates and the water depth values ​​corresponding to the water depth measurement points; setting different color mapping rules according to different ranges of water depth values ​​to perform color rendering on the constructed visual graphic; and displaying the rendered visual graphic on a display device.

[0012] Optionally, historical remote sensing image data and historical ground water level gauge data in the target area are obtained, including: determining an area whose distance from a transmission line of a preset length meets a preset distance as the target area; downloading initial remote sensing image data of the target area within a preset time window from a satellite image database, selecting remote sensing image data with a resolution greater than a first preset threshold and a cloud coverage percentage less than a second preset threshold from the initial remote sensing image data, performing radiation correction, geometric correction and atmospheric correction on the selected remote sensing image data to obtain historical remote sensing image data; determining the target acquisition time corresponding to the initial remote sensing image data, and obtaining historical ground water level gauge data corresponding to the target acquisition time.

[0013] According to another aspect of the embodiment of the present application, a water depth inversion device is also provided, including: an acquisition module for acquiring historical remote sensing image data and historical ground water level gauge data in a target area, wherein the historical remote sensing image data includes: a first number of first historical radiation brightness values ​​and a second number of second historical radiation brightness values, and the historical ground water level gauge data includes: a third number of first historical real water depth information and a fourth number of second historical real water depth information, wherein the second number is smaller than the first number, and the fourth number is smaller than the third number; a determination module for determining a physical model for representing the correlation relationship between the first historical radiation brightness value and the first historical real water depth information; a first processing module for inputting the second historical radiation brightness value into the physical model to obtain predicted water depth information output by the physical model; a correction module for correcting the predicted water depth information using the second historical real water depth information, and updating the physical model according to the correction result to obtain a target physical model; and a second processing module for inputting the current radiation brightness value into the target physical model to obtain target predicted water depth information output by the target physical model.

[0014] According to another aspect of the embodiments of the present application, a non-volatile storage medium is further provided, the storage medium including a stored program, wherein when the program is executed, the device where the storage medium is located is controlled to execute the above water depth inversion method.

[0015] According to another aspect of the embodiments of the present application, an electronic device is provided, including: a memory and a processor, wherein the processor is configured to run a program stored in the memory, wherein the above water depth inversion method is executed when the program is run.

[0016] According to yet another aspect of the embodiments of the present application, a computer program is further provided, wherein when the computer program is executed by a processor, the above water depth inversion method is implemented.

[0017] According to another aspect of an embodiment of the present application, a computer program product is further provided, which includes a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above water depth inversion method is implemented.

[0018] In an embodiment of the present application, historical remote sensing image data and historical ground water level gauge data in the target area are obtained, wherein the historical remote sensing image data includes: a first number of first historical radiation brightness values ​​and a second number of second historical radiation brightness values, and the historical ground water level gauge data includes: a third number of first historical real water depth information and a fourth number of second historical real water depth information, wherein the second number is smaller than the first number, and the fourth number is smaller than the third number; a physical model for representing the correlation between the first historical radiation brightness value and the first historical real water depth information is determined; the second historical radiation brightness value is input into the physical model, and a The method obtains the predicted water depth information output by the physical model; uses the second historical real water depth information to correct the predicted water depth information, and updates the physical model according to the correction result to obtain the target physical model; inputs the current radiation brightness value into the target physical model to obtain the target predicted water depth information output by the target physical model, thereby achieving the purpose of improving the accuracy of the water depth inversion result, thereby realizing the technical effect of providing more accurate data support for flood disaster warning and emergency response, and further solving the technical problem that the existing satellite remote sensing technology has low accuracy of the inversion result when performing large-scale water depth inversion due to the influence of factors such as cloud layer, atmosphere, and surface type. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0020] Figure 1 is a flow chart of a water depth inversion method according to an embodiment of the present application;

[0021] Figure 2 is a flow chart of another water depth inversion method according to an embodiment of the present application;

[0022] Figure 3 is a structural diagram of a water depth inversion device according to an embodiment of the present application;

[0023] Figure 4 This is a hardware structure block diagram of a computer terminal for a water depth inversion method according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] According to an embodiment of the present application, a method embodiment of a water depth inversion method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0027] Figure 1 is a flow chart of a water depth inversion method according to an embodiment of the present application, such as Figure 1 As shown, the method includes the following steps:

[0028] Step S101, obtain historical remote sensing image data and historical ground water level gauge data in the target area, wherein the historical remote sensing image data includes: a first number of first historical radiation brightness values ​​and a second number of second historical radiation brightness values, and the historical ground water level gauge data includes: a third number of first historical real water depth information and a fourth number of second historical real water depth information, wherein the second number is smaller than the first number, and the fourth number is smaller than the third number.

[0029] By logging into public satellite image databases (such as NASA Earthdata and ESA Copernicus OpenAccess Hub), historical remote sensing image data of the target area are downloaded. At the same time, historical ground water level gauge data in the buffer zone along the transmission lines in the target area are collected.

[0030] The historical remote sensing image data includes a first number of first historical radiance values. To improve model training efficiency and avoid data redundancy, a second number of second historical radiance values ​​is selected from this data for model training, where the second number is smaller than the first number. The historical surface water level gauge data includes a third number of first historical real water depth information, which is used for model calibration and validation. Similarly, to optimize model training, a fourth number of second historical real water depth information is selected from this data, where the fourth number is smaller than the third number.

[0031] Step S102: determining a physical model for representing the correlation relationship between the first historical radiance value and the first historical true water depth information.

[0032] Based on the target area's water quality, bottom type, and lighting conditions, determine a physical model to represent the relationship between the first historical radiance value and the first historical true water depth information. For example, if the target area's water quality is clear, a two-band model or a log-ratio model might be preferred; if the water quality is turbid, a multi-band model might be used.

[0033] Step S103: input the second historical radiance value into the physical model to obtain predicted water depth information output by the physical model.

[0034] The second historical radiance value is input into the selected physical model, which is then used to calculate and predict water depth information. This process is part of the model training process, which aims to preliminarily evaluate the model's inversion capabilities using limited, high-quality historical data.

[0035] Step S104: Correct the predicted water depth information using the second historical real water depth information, and update the physical model according to the correction result to obtain a target physical model.

[0036] The fourth amount of second historical actual water depth information is compared with the predicted water depth information obtained in step S103, and error metrics (such as RMSE and MAE) are calculated to evaluate the model's prediction accuracy. The preliminary inversion results are corrected based on the second historical actual water depth information. A regression model is constructed, using the predicted water depth information as the independent variable and the actual water depth information as the dependent variable, and the model parameters are fitted to reduce the prediction error. The result of the model update is a more accurate target physical model.

[0037] Step S105 : inputting the current radiation brightness value into the target physical model to obtain target predicted water depth information output by the target physical model.

[0038] Acquire current remote sensing image data of the target area and perform preprocessing such as radiometric calibration, geometric correction, and atmospheric correction to obtain the current radiometric brightness value. This radiometric brightness value is then input into the target physical model, which is then used to perform real-time water depth inversion to obtain the target predicted water depth information. This result more accurately reflects the current water depth conditions in the target area.

[0039] Through these steps, not only can a physical model based on historical data be established, but model parameters can also be continuously optimized through a calibration process, ultimately achieving accurate predictions of real-time water depth conditions in the target area. This method combines the wide-area coverage of satellite remote sensing with the high-precision measurements of ground-based water level gauges, effectively improving the accuracy and real-time nature of water depth inversion. This is of great significance for power grid security monitoring, disaster warning, and emergency response.

[0040] The following Figure 1 The steps shown are exemplified and explained.

[0041] According to some optional embodiments of the present application, determining a physical model for representing the correlation between a first historical radiation brightness value and a first historical real water depth information can be achieved by the following method: obtaining the daily average change value of the solar radiation intensity of the target area within a preset historical time period; if the daily average change value of the solar radiation intensity is within a first preset interval, using a single-band model to determine the correlation relationship, wherein the single-band model is used to determine the linear or nonlinear relationship between the water depth and the reflectivity and radiation brightness value of the preset band; if the daily average change value of the solar radiation intensity is within a second preset interval, using a dual-band model to determine the correlation relationship, wherein the dual-band model is used to determine the logarithm of the ratio or difference between the water depth and the reflectivity of two different bands, and the radiation brightness The linear or nonlinear relationship between the water depth and the reflectivity values ​​is determined, and the minimum value of the second preset interval is greater than the maximum value of the first preset interval; if the daily average change value of the solar radiation intensity is within the third preset interval, the logarithmic ratio model is used to determine the association relationship, wherein the logarithmic ratio model is used to determine the linear or nonlinear relationship between the natural logarithm of the water depth and the reflectivity ratio and the radiation brightness value, and the minimum value of the third preset interval is greater than the maximum value of the second preset interval; if the daily average change value of the solar radiation intensity is within the fourth preset interval, the multi-band model is used to determine the association relationship, wherein the multi-band model is used to determine the linear or nonlinear relationship between the water depth and the reflectivity and radiation brightness values ​​of multiple preset bands, and the minimum value of the fourth preset interval is greater than the maximum value of the third preset interval.

[0042] Preferably, the first preset interval is 45W / m 2 Up to 55W / m 2 ; The second preset interval is 100W / m 2 Up to 180W / m 2 ; The third preset interval is 200W / m 2Up to 220W / m 2 ; The fourth preset interval is 300W / m 2 Up to 350W / m 2 .

[0043] Specifically, before analyzing the remote sensing image data of the target area, it is first necessary to obtain the daily average variation of solar radiation intensity in the target area over a preset historical period. This can be estimated by collecting metadata from historical remote sensing data, particularly solar altitude and atmospheric conditions, and combining it with a solar radiation transfer model (such as the MODTRAN model). The selection of the historical period should take into account factors such as seasonal changes and weather conditions to ensure that the collected data reflects the typical lighting conditions of the target area.

[0044] According to the daily average change in solar radiation intensity, determine which physical model to use for water depth inversion. The change in solar radiation intensity will directly affect the reflectivity of the remote sensing image, and thus affect the accuracy of water depth inversion. Therefore, it is necessary to use different physical models for different lighting conditions. When the daily average change in solar radiation intensity is within the first preset range (45W / m 2 Up to 55W / m 2 ), a single-band model is used. Single-band models are usually based on the linear or nonlinear relationship between the reflectivity of a specific band and the water depth, and are suitable for situations where the lighting conditions are relatively stable. When the daily average change in solar radiation intensity is within the second preset interval (100W / m 2 Up to 180W / m 2 ), the dual-band model is used. The dual-band model uses the relationship between the ratio or logarithm of the difference of the reflectance of the two bands and the water depth to better eliminate the impact of light changes on the inversion results. It is suitable for environments with moderate light changes. When the daily average change value of solar radiation intensity is within the third preset interval (200W / m 2 Up to 220W / m 2 ), select the logarithmic ratio model. The logarithmic ratio model further improves the ability to handle light changes by calculating the relationship between the natural logarithm of the reflectivity ratio and the water depth. It is suitable for situations where light conditions vary greatly. When the daily average change in solar radiation intensity is within the fourth preset interval (300W / m 2 Up to 350W / m 2 ), a multi-band model is used. The multi-band model integrates information from multiple bands to establish a more complex water depth inversion model. It can effectively cope with environments with extreme changes in lighting conditions and improve the robustness of the inversion.

[0045] After deciding which physical model to use, appropriate parameters, such as reflectivity and water optical properties, must be set based on the model's characteristics. A preliminary water depth inversion calculation is then performed. Radiometric brightness values ​​from historical remote sensing data are input into the model to obtain a preliminary water depth estimate. Next, the initial inversion results are corrected using actual water depth data from historical surface water level gauges. By constructing a regression model to fit the relationship between predicted and actual water depths, the physical model parameters are optimized and adjusted to ensure the accuracy and reliability of the inversion results.

[0046] According to some other optional embodiments of the present application, the dual-band model is used to determine the correlation relationship, which can be achieved by the following method: The correlation relationship is determined using the following formula: Z = (1 / fk i )(lnr Bi -X i ) Where Z is the relative water depth, f is the length of the water path, k i is the effective attenuation coefficient of water, r Bi is the bottom reflectivity at water depth Z, L i is the radiation brightness value at water depth Z, L si is the radiation brightness value at the preset water depth, R i It is a function that represents the relationship between solar radiation value, transmission flux at the interface between atmosphere and water body, and refractive index of water surface.

[0047] In some optional embodiments of the present application, the use of the second historical real water depth information to correct the predicted water depth information can be achieved by the following method: dividing the second historical real water depth information into training set data and verification set data; using the training set data to train a polynomial regression model, and using the verification set data to verify the mean square error between the output value of the polynomial regression model and the true value, and when the mean square error is less than a preset threshold, obtaining a trained polynomial regression model; inputting the predicted water depth information into the trained polynomial regression model to obtain a correction result for the predicted water depth information.

[0048] The second historical real water depth information is randomly divided into a training set and a validation set. Typically, the training set contains the majority of the data (e.g., 70%-80%) for model training, while the validation set retains a smaller portion (e.g., 20%-30%) for evaluating model performance. A polynomial regression model is constructed using the training set data. The model can be linear or a higher-order polynomial. The goal of training is to find the optimal parameters for the polynomial model so that the difference between the model's predicted values ​​and the true water depth values ​​in the training set data is minimized. This can be accomplished using the polynomial regression function in statistical software, by setting the model order and adjusting the model parameters until model training is complete. The performance of the polynomial regression model is evaluated using the validation set data. The predicted water depth information from the validation set data is input into the model and compared with the true water depth information from the same dataset to calculate the mean squared error (MSE). Once the polynomial regression model passes validation, i.e., the mean squared error (MSE) is less than a preset threshold, the predicted water depth information can be fed into the model for correction. The correction output by the model yields a predicted value closer to the true water depth, improving the accuracy of water depth inversion.

[0049] The polynomial regression model trained on historical data not only improves the accuracy of water depth inversion but also ensures its stability and generalization, enabling its effective application under varying lighting and environmental conditions. This model calibration method provides an important data processing and precision improvement tool for quantitative inversion of flood depth in satellite-ground coordinated power grids.

[0050] As some optional embodiments of the present application, the target predicted water depth information includes: a plurality of discrete water depth measurement points and water depth values ​​corresponding to the water depth measurement points, wherein each water depth measurement point has a geographic coordinate identifier.

[0051] Furthermore, after obtaining the target predicted water depth information output by the target physical model, the following steps can also be performed: converting the geographic coordinate identification of the water depth measurement point into plane coordinates suitable for visual display; constructing a visual graphic based on the converted plane coordinates and the water depth values ​​corresponding to the water depth measurement point; setting different color mapping rules according to different ranges of water depth values ​​to perform color rendering on the constructed visual graphic; and displaying the rendered visual graphic on a display device.

[0052] Specifically, the geographic coordinates (usually longitude and latitude) of the bathymetric measurement points are first converted into planar coordinates suitable for two-dimensional map display. This can be achieved by using map projection technology, such as the UTM (Universal Transverse Mercator) projection or the Web Mercator projection, to convert the three-dimensional coordinates on the Earth's surface into planar coordinates on a two-dimensional map for subsequent processing in GIS software.

[0053] Next, based on the converted plane coordinates and the water depth values ​​corresponding to the bathymetry points, a visualization is constructed using GIS software or a programming language (such as Python's Matplotlib or Seaborn library). This typically involves creating a scatter plot or contour map, where the location of each bathymetry point is represented by its plane coordinates, while the water depth values ​​are displayed by point size, color, or contour line values.

[0054] Third, to make visualizations easier to interpret, color mapping rules are set based on different ranges of water depth values. For example, shallow water areas might be represented by light blue, while deep water areas might be represented by dark blue or black. The color mapping rule should ensure that the depth changes are continuous and contrasting, allowing viewers to quickly identify patterns in water depth changes. This can be achieved by defining a color gradient table (Color Map), which maps the range of values ​​to a series of colors.

[0055] Finally, the constructed visualization is rendered, with the color of each bathymetric measurement point or grid point set according to color mapping rules, to generate a colorful bathymetric distribution map. The rendered image can then be displayed on a computer screen, tablet, or any other device that supports graphic display, using the GIS software interface or a developed visualization application. Additional map elements, such as scale bars, north arrows, and legends, can be added to enhance the map's readability and information integrity.

[0056] Through the above steps, not only can the water depth distribution in the target area be visually displayed, but the depth values ​​can also be color-coded according to their numerical range, allowing viewers to quickly understand the changing trends of water depth. This has important practical value for power grid flooding depth monitoring and disaster warning, helping decision-makers and emergency response teams to more effectively assess flood risks and develop response strategies. Furthermore, this visualization can further identify potential deviations or abnormal areas in model predictions, providing a basis for model optimization and iteration.

[0057] In some optional embodiments of the present application, obtaining historical remote sensing image data and historical ground water level gauge data in the target area can be achieved by the following method: determining an area that is at a preset distance from a transmission line of a preset length as the target area; downloading initial remote sensing image data of the target area within a preset time window from a satellite image database, selecting remote sensing image data with a resolution greater than a first preset threshold and a cloud coverage percentage less than a second preset threshold from the initial remote sensing image data, performing radiation correction, geometric correction and atmospheric correction on the selected remote sensing image data, and obtaining historical remote sensing image data; determining the target acquisition time corresponding to the initial remote sensing image data, and obtaining historical ground water level gauge data corresponding to the target acquisition time.

[0058] Specifically, first, a buffer zone with a predetermined length from the transmission line is identified based on the transmission line's location coordinates, serving as the target area. The predetermined length should take into account the width of the transmission line potentially affected by flooding, ensuring that the target area encompasses all potentially affected sections of the transmission line. For example, if the predetermined length is 500 meters, the target area would be within 500 meters on either side of the transmission line.

[0059] Next, log in to satellite imagery databases, such as NASA Earthdata and ESA Copernicus Open Access Hub, and download initial remote sensing imagery data based on the latitude and longitude of the target area and a pre-set time window. The pre-set time window should include the period before and after the flood event to ensure that the complete process of water body changes is captured.

[0060] Again, from the downloaded initial remote sensing image data, filter out data that meets the following conditions: The resolution is greater than the first preset threshold: The choice of resolution directly affects the accuracy of water depth inversion. The setting of the preset threshold should ensure that the data can distinguish the details of the transmission line and the surrounding area. For example, the first preset threshold can be set to 30 meters to ensure that the image can clearly display the surface features. The cloud cover percentage is less than the second preset threshold: Clouds will obscure the surface and affect the quality of remote sensing data. The setting of the preset threshold should exclude images that are severely interfered with by clouds. For example, the second preset threshold can be set to 10% to ensure the availability and accuracy of the data. And pre-process the filtered remote sensing image data, including: Radiometric correction: Convert digital numbers (DN) into surface reflectivity to eliminate the influence of sensor and atmospheric effects. Geometric correction: Adjust the geometric deformation of the image to ensure the consistency of image pixels and ground coordinates. Atmospheric correction: Reduce the impact of the atmosphere on the image and improve the accuracy of the data.

[0061] Next, determine the capture time of the historical remote sensing image data, i.e., the target acquisition time. Match the target acquisition time with the recording time of the ground water level gauge, and select water level gauge data with the same time or a time difference within the allowable range. This ensures data synchronization and effectively combines remote sensing data with ground data.

[0062] Finally, historical remote sensing imagery and surface water level gauge data were integrated to provide the foundational data for building the physical model. Following the previously discussed technical briefing, the team then carried out the physical model construction, model training and validation, and result calibration and inversion steps, ultimately achieving an accurate inversion of the water depth in the target area.

[0063] Through the above steps, it can be ensured that the remote sensing data and ground water level gauge data used for water depth inversion meet the high-precision requirements in terms of time, space and quality, providing a solid data foundation for the quantitative inversion of flood depth in satellite-ground coordinated power grid.

[0064] Figure 2 is a flow chart of another water depth inversion method according to an embodiment of the present application, such as Figure 2 As shown, the method includes the following steps:

[0065] Step 201, data collection and preprocessing, involves acquiring and organizing remote sensing imagery and surface water level gauge data. The downloaded remote sensing data undergoes a series of preprocessing steps, including radiometric correction, geometric correction, and atmospheric correction, to eliminate the influence of external factors like the atmosphere and illumination, ensuring data quality and accuracy. Surface water level gauge data, including its location coordinates and real-time water level information, is also collected. This data will be used for subsequent model calibration and verification. Data collection and preprocessing are fundamental steps in remote sensing water depth inversion projects, centered on ensuring data quality so that subsequent analysis and modeling can be based on accurate and reliable raw data.

[0066] The data collection and preprocessing process in this embodiment mainly consists of the following four sub-steps:

[0067] Step 211: Acquire remote sensing data. A buffer zone along the transmission line at a certain distance is designated as the study area. The time range for the required data is determined, and remote sensing imagery data for the target area is downloaded from a satellite imagery database. Remote sensing data acquisition involves collecting electromagnetic radiation signals from the Earth's surface using satellite platforms. These signals contain physical and chemical information about the surface. Satellite-borne sensors can receive and record radiation across different wavelengths, generating remote sensing images that are then used to analyze and monitor surface features. Based on the location of the transmission line, a buffer zone of a certain width along both sides of the line is designated as the study area to ensure coverage of all sections of the transmission line potentially affected by flooding. A time window is selected for data collection, typically around predicted or actual flood events, to capture changes in the water body. Access public satellite imagery databases such as NASA Earthdata and the ESA Capronicus Open Access Hub, select appropriate satellites (e.g., Landsat or Sentinel series) and bands, and download the corresponding remote sensing imagery data based on the study area and time range. Furthermore, select a file containing all necessary bands and metadata when downloading.

[0068] Step 2012: Data screening, through metadata inspection and image quality assessment, select cloudless or low cloud cover, high-definition images, and exclude data affected by weather conditions. Data screening is based on image quality and time series consistency to ensure that the data used can meet the accuracy requirements of water depth inversion. The judgment of image quality is mainly based on factors such as cloud cover, lighting conditions and sensor performance. Specific practices include: checking the metadata of each image, confirming information such as shooting time, cloud cover percentage, sensor type, etc. Use image processing software (such as ENVI, GDAL, QGIS) to open the image, check the image clarity, and ensure that there are no clouds or the cloud cover is less than 10%. Eliminate images with severe cloud obstruction, poor lighting conditions or obvious sensor stripes to ensure data quality.

[0069] Step 2013 involves image preprocessing. Radiometric correction is performed to convert pixel values ​​into true surface reflectances; geometric correction is applied to ensure the accuracy of the image's spatial position; and atmospheric correction is performed to reduce image distortion caused by atmospheric effects. Radiometric calibration is a crucial step in remote sensing data processing. It involves converting the digital quantization (DN) values ​​of remote sensing images into physical quantities, such as radiance, reflectance, or surface temperature. Radiometric calibration eliminates errors inherent in the sensor itself, improving the accuracy and comparability of remote sensing data. Geometric correction of remote sensing images eliminates spatial distortion caused by factors such as sensor attitude, platform motion, earth curvature, and atmospheric refraction, ensuring that the imaged features match their actual positions on the ground. Geometrically corrected remote sensing images can be used to create accurate geographic coordinate maps and overlay analysis with other geospatial data. Geometric correction works by converting image pixel coordinates into ground coordinates by establishing a transformation relationship between the image coordinate system and the ground coordinate system. Atmospheric correction eliminates atmospheric effects on remote sensing images, ensuring that the reflectance or radiance of features in the images approximates their true values. Atmospheric correction can improve the accuracy and comparability of remote sensing data, making it more suitable for quantitative analysis and application.

[0070] In step 2014, water level gauge data is collected in the target area, ensuring data time synchronization. This ensures that the water level data matches the capture time of the remote sensing image. A preliminary check is performed on the water level data to remove outliers and ensure data accuracy and reliability. The water level gauge data collection time is ensured to match the remote sensing image capture time, typically with a time difference of no more than a few hours. Statistical methods are used to remove outliers from the water level gauge data to ensure data continuity and consistency.

[0071] Step 202 constructs a physical model for preliminary water depth inversion, estimating water depth using the principles of physical optics and remote sensing data. This involves establishing a relationship between the radiance values ​​received by the sensor and the reflectivity of the bottom, water depth, and the optical properties of the water. Using the radiative transfer equation, the reflectivity of the remote sensing image is converted into water depth information, providing a foundation for subsequent corrections.

[0072] The physical model for water depth inversion is primarily based on the correlation between the optical properties of water and remote sensing data. Radiation transfer theory is used to explain the relationship between the radiance values ​​received by the sensor and water depth, bottom reflectivity, and the optical properties of the water. The radiation transfer equation describes the propagation, absorption, scattering, and reflection of light in water. The optical properties of the water (such as absorption and scattering coefficients) and bottom reflectivity have a direct impact on the radiance values ​​received by the remote sensor.

[0073] Step 202 specifically includes the following steps:

[0074] Step 221, water depth inversion model selection, select or customize the water depth inversion physical model according to regional characteristics, such as single-band model, dual-band (ratio logarithm) model, log-ratio model and multi-band model. When selecting a water depth inversion model, factors such as water quality, bottom type, and light conditions of the study area need to be considered. Different models are based on different physical principles and assumptions, for example:

[0075] Single-band model: Simply use the reflectivity of a specific band to establish a linear or nonlinear relationship with water depth.

[0076]

[0077] Where, Z is the relative water depth, f is the length of the water path, k i is the effective attenuation coefficient of water, r Bi is the bottom reflectivity at water depth Z, L i is the radiation brightness value at water depth Z, L si is the radiation brightness value of deep water, R i It is a function of solar radiation, transmission at the air-water interface, and refraction at the water surface.

[0078] Dual-band model: The water depth is inverted by the ratio or logarithm of the difference of the reflectivity of two different bands. It is suitable for clearer waters.

[0079]

[0080] Where, (i=1, 2, representing the 1st and 2nd bands respectively), Z is the relative water depth, f is the length of the water path, k i is the effective attenuation coefficient of water in band i, r Bi is the bottom reflectivity at the water depth Z in the i-th band, L i is the radiation brightness value at the water depth Z in the i-th band, L si is the radiation brightness value of deep water, R i It is a function of solar radiation, transmission at the air-water interface, and refraction at the water surface.

[0081] Logarithmic ratio model: Using the relationship between the natural logarithm of the reflectivity ratio and water depth can reduce water surface fluctuations and atmospheric influences.

[0082]

[0083] Where m1 is the constant that adjusts the ratio to water depth, n is a fixed value for the entire area, and λ i is the blue band, λ j is the green band, R w(λ) is the water reflectivity in the λ band at depth Z, and m0 is the offset relative to the water depth of 0 meters.

[0084] Multi-band model: Integrates information from multiple bands to improve inversion accuracy and is suitable for complex water environments.

[0085]

[0086] Where, (i=1, 2, representing the 1st and 2nd bands respectively), Z is the relative water depth, f is the length of the water path, k i is the effective attenuation coefficient of water in band i, r Bi is the bottom reflectivity at the water depth Z in the i-th band, L i is the radiation brightness value at the water depth Z in the i-th band, L si is the radiation brightness value of deep water, R i It is a function of solar radiation, transmission at the air-water interface, and refraction at the water surface.

[0087] Analyze the characteristics of the target area, evaluate the applicability and accuracy of each model, test multiple models and compare the results, and select the most appropriate model for subsequent analysis.

[0088] Step 2022, parameter setting, sets the parameters required for the model. Determine the input parameters required for the inversion model, such as spectral reflectance and topographic data. Collect and process the required parameter data to ensure its accuracy and reliability. Set the initial model parameter values, perform parameter sensitivity analysis, and determine the optimal parameter combination. Model parameters include spectral reflectance, water optical properties (such as absorption and scattering coefficients), and bottom reflectance. The accuracy and reliability of these parameters are crucial to model performance.

[0089] Specifically, remote sensing imagery, topographic data, and possibly field observation data are collected and processed. Initial parameter values ​​are determined using existing databases or literature. Parameter sensitivity analysis is performed to identify which parameters have the greatest impact on model output. These parameters are adjusted to optimize model performance. Model parameters are estimated using statistical methods or empirical formulas, such as calibrating bottom reflectance and water optical properties using field measurements.

[0090] Step 2023: Execute a preliminary inversion, using the remote sensing data and the physical model to perform preliminary water depth inversion calculations. The preprocessed remote sensing image data is input into the physical model, which is then run to calculate the water depth inversion results for the transmission line buffer zone. The results are then output as image files for subsequent analysis. Using the preprocessed remote sensing data as input, combined with the selected physical model and parameter settings, water depth inversion calculations are performed.

[0091] Specifically, the preprocessed image data is imported into the water depth inversion model, and the model calculation is performed to obtain preliminary water depth inversion results. The inversion results are output as geospatial data formats, such as GeoTIFF, so that they can be read by GIS software.

[0092] Step 2024: Visualize the results. Present the preliminary inversion results in a map format to facilitate subsequent analysis and correction. Use GIS software (such as ArcGIS, QGIS, etc.) to load the inversion results, visualize the inversion results, and generate a water depth distribution map.

[0093] Step 203, regression model construction and correction, uses the actual measurement data of the surface water level gauge to correct the preliminary inversion results to improve their accuracy. By constructing a regression model, the water depth estimate of the preliminary inversion is used as the independent variable, and the true value of the water level gauge is used as the dependent variable, and the model parameters are fitted by minimizing the difference between the predicted value and the actual value. The regression model can capture the relationship between the preliminary inversion result and the actual water depth, thereby systematically correcting the preliminary inversion result on a global scale and improving the overall accuracy of the water depth inversion. Regression is used to analyze the relationship between one or more independent variables (explanatory variables) and dependent variables (response variables). In this context, we use a polynomial regression model to correct the results of the water depth inversion to make it closer to the actual measurement data of the surface water level gauge.

[0094] The step 203 specifically includes the following steps:

[0095] Step 2031, data pairing, pairing the preliminary inversion results with the true data of the water level gauge. Determine the corresponding positions of the preliminary inversion results and the water level gauge data, match the two, ensure consistency in time and space, check the paired data, and eliminate data points that do not meet the requirements. Data pairing is to ensure the consistency in time and space between the preliminary inversion results and the measured data of the ground water level gauge, which is the prerequisite for model construction and correction. The specific operation is: use GIS software or programming language to match the position coordinates of the water level gauge with the pixels on the remote sensing image. Ensure that the measurement time of the water level gauge matches the shooting time of the remote sensing image, usually allowing a small time window (such as ±1 hour). Eliminate data points containing outliers or missing values ​​to ensure the integrity and accuracy of the data.

[0096] Step 2032: Model training. Use the paired data to build a regression model and estimate model parameters. Divide the paired data into a training set and a validation set. Use the training set data to build a regression model, estimate the intercept and slope parameters, and evaluate the model's fit, such as R. 2 Value, residual analysis, etc. Regression model training aims to estimate model parameters by minimizing the difference between predicted values ​​and actual values, thereby establishing a relationship between the preliminary inversion results and the measured water depth.

[0097] Specifically, paired data are randomly divided into a training set (approximately 70%-80%) and a validation set (approximately 20%-30%). Using the training set data, a linear regression or polynomial regression model is applied to estimate model parameters (e.g., intercept, slope, polynomial coefficients). Statistics such as the coefficient of determination (R²) and the standard deviation of the residuals are calculated to evaluate the model's fit to the training set.

[0098] The polynomial regression model is: y = β0 + β1x + β2x 2 +β3x 3 +...+β n x n +∈

[0099] Among them, y is the dependent variable (ground measured water depth), x is the independent variable (preliminary inversion water depth), β0, β1,…, β n are model parameters, and ∈ is the error term.

[0100] Step 2033, model validation, verifies the model's predictive ability and generalization ability. Use the validation set data to evaluate the model's prediction effect, calculate the model's prediction error, such as the mean square error, mean absolute error, etc., and optimize the model to ensure its generalization ability and robustness. Model validation tests the model's generalization ability and robustness by evaluating the model's performance on unseen validation set data. The specific operation is: using the validation set data, calculate the model's prediction indicators such as the mean square error and mean absolute error. Check the residual distribution between the predicted value and the actual value to ensure that it conforms to the normal distribution and has no obvious deviation or heteroscedasticity. Based on the verification results, adjust the model parameters or try different model structures to improve the prediction accuracy.

[0101] Error calculation is mainly used to quantify the difference between the model prediction value and the actual observation value. Common error indicators include mean square error, mean absolute error, etc.

[0102] The mean squared error measures the average of the squares of the differences between the predicted values ​​and the actual values, and its formula is:

[0103] Among them, y i is the actual value of the i-th sample, is the model's predicted value for the sample, and N is the total number of samples.

[0104] The mean absolute error measures the average of the absolute differences between the predicted values ​​and the actual values, and its formula is:

[0105]

[0106] Residual analysis is a further diagnosis of the model prediction results to check whether the model assumptions are valid and whether there are systematic deviations. The residual is the difference between the actual observation value and the model prediction value, that is:

[0107] To check whether the residuals follow a normal distribution, common methods include plotting residual histograms, QQ plots (QuantileQuantile Plots), and applying normality tests (such as the Shapiro-Wilk test). Heteroskedasticity means that the variance of the residuals changes with changes in the predicted value or an explanatory variable. Common tests include the Breusch-Pagan test and the White test. To check whether the residuals have systematic bias, that is, whether the mean of the residuals deviates significantly from zero. This can be observed by calculating the mean of the residuals or plotting a scatter plot of the residuals against the predicted values.

[0108] Step 2034, result correction, applies the trained regression model to correct the initial inversion results. The initial inversion results are input into the regression model for error correction, and the corrected depth inversion results are output. The corrected results are then preliminarily checked to ensure their rationality and accuracy. The trained regression model uses the initial inversion results as input and outputs the corrected depth estimate to improve the accuracy of the inversion results.

[0109] Step 204: Result verification and output, check the inversion results after correction to ensure their quality and reliability. This includes comparing with the measured value of the water level meter and calculating statistical indicators (RMSE, coefficient of determination R 2 The inversion results are visualized and output as high-resolution maps to intuitively display the water depth distribution.

[0110] Comparing the inverted data with ground-truth data (e.g., measurements from a water level gauge) is a direct way to verify the accuracy of the inversion results. The ground-truth data provides a real-world reference point for evaluating the accuracy of the model predictions. To quantitatively evaluate the model performance, a series of statistical indicators can be calculated, such as the root mean square error, the coefficient of determination (R 2 )wait.

[0111] The root mean square error measures the root mean square of the difference between the predicted value and the actual value, and its formula is:

[0112] Among them, y i is the actual value of the i-th sample, is the model's predicted value for the sample, and N is the total number of samples.

[0113] Coefficient of Determination (R 2 ):R 2 It reflects the proportion of the variation explained by the model to the total variation, and its value ranges from 0 to 1. The closer the value is to 1, the better the model fitting effect. 2 The formula is:

[0114]

[0115] in, is the average of all actual observations.

[0116] Displaying the inversion results in the form of high-resolution maps allows for an intuitive understanding of the distribution of water depths. This involves the application of Geographic Information System (GIS) software to convert numerical data into spatial distribution maps. For water depth data, color gradients can be used to represent different depths of water, allowing viewers to quickly identify changes in water depth.

[0117] The inversion process is essentially solving an inverse problem, that is, inferring unknown physical parameters or states from indirect observational data. In hydrology, this may mean inferring water depth distribution from satellite imagery or sonar data. The inversion algorithm attempts to find an optimal solution that minimizes the difference between the model prediction and the observed data. This difference is usually measured by a loss function (such as MSE), and the optimization process may use various mathematical methods such as least squares, maximum likelihood estimation, or Bayesian inference. Ultimately, through the verification and output steps, it can be ensured that the inversion results are not only mathematically sound, but also accurate and reliable in real-world application scenarios.

[0118] The step 204 specifically includes the following steps:

[0119] Step 2041, result comparison, re-compares the corrected inversion result with the water level gauge's true value. Compare the corrected depth inversion result with the water level gauge data, calculate the difference between the two, evaluate the correction effect, and compare and analyze the results to identify possible sources of error and areas for improvement. True value acquisition: First, ensure that the water level gauge's measured value, serving as the "true value," is accurate. Water level gauges are typically deployed in fixed locations, continuously recording water level changes and providing accurate depth data over a time series. This data can serve as a benchmark for verifying the inversion results.

[0120] Data registration: Since the water level gauge data is a point measurement, and the inversion results are usually spatially distributed, it is necessary to match the position of the water level gauge with the inversion grid and extract the inverted water depth value at the corresponding position. Error calculation: Calculate the difference between the inversion result and the water level gauge data. Common error indicators include the root mean square error, which can reflect the square root of the average deviation between the predicted value and the observed value, and gives the absolute size of the error; in addition, there is the mean absolute error, which reflects the average absolute deviation between the predicted value and the observed value, regardless of the positive or negative sign. Correction effect evaluation: By comparing the RMSE or MAE before and after correction, the effectiveness of the correction method can be evaluated. If the error is significantly reduced after correction, it means that the correction method has effectively improved the accuracy of the inversion result.

[0121] Step 2042, error analysis, calculates the statistical error of the inversion result after correction. Calculates statistical error indicators such as root mean square error and mean absolute error. Analyze the error distribution, identify areas with large errors and their causes, and propose suggestions for further improving the model and method. Statistical error indicator calculation:

[0122] Producing an error distribution map, which visualizes the error values ​​at each grid point, helps identify spatial patterns of error. Analyzing the error distribution map identifies areas of high error and explores possible sources of error, such as inaccurate model assumptions, poor input data quality, and changes in environmental conditions. Based on the results of the error analysis, propose targeted improvement measures, such as optimizing model parameters, adding calibration sites, and improving data preprocessing.

[0123] Step 2043: Visualization output. The final corrected water depth inversion results are converted into a high-resolution map. GIS software is used to generate a high-resolution water depth distribution map, adding necessary map elements and annotations to ensure map clarity and readability.

[0124] Through the above steps, it can be seen that the advantages of this application are:

[0125] 1. Satellite-ground collaboration significantly improves the accuracy and reliability of water depth inversion. Traditional water depth inversion methods rely on single remote sensing data and are easily affected by atmospheric conditions, illumination changes, and sensor errors, resulting in uncertainty and deviation in the inversion results. Through satellite-ground collaboration, this application combines the wide-area coverage advantage of satellite remote sensing and the high-precision measurement of ground water level meters to form a complementary observation system. Satellite remote sensing data provides a macroscopic perspective of water distribution and preliminary water depth estimation, while ground water level meters provide key control points and correction standards. The combination of the two significantly improves the accuracy and reliability of water depth inversion, especially in complex water conditions.

[0126] 2. Achieve large-scale, highly timely water depth monitoring. This application leverages the real-time monitoring capabilities of satellite remote sensing to cover large areas in a short period of time, enabling rapid response and continuous monitoring. This is particularly important for flood warning, disaster assessment, and emergency management. Compared to traditional manual measurements or local monitoring methods, this application can provide more comprehensive and timely water depth information, facilitating the rapid implementation of response measures and reducing losses.

[0127] 3. Reduced uncertainty in remote sensing data. Remote sensing data is subject to a variety of factors, including cloud cover, sensor noise, atmospheric effects, etc., which will affect the accuracy of water depth inversion. By integrating the true value data of the ground water level gauge, this application can effectively correct and verify the preliminary inversion results, significantly reducing the inherent uncertainty of remote sensing data and improving the credibility of the inversion results. This correction mechanism is particularly important in complex environmental conditions, ensuring that the inversion results maintain a high degree of accuracy under all circumstances.

[0128] 4. Refined processing of transmission lines improves the accuracy of water depth assessment in specific areas. Transmission lines and their surrounding environment are the key areas of focus of this application because they play a vital role in power supply and safe operation. By inverting the water depth specifically for the transmission line buffer zone, this application can provide high-resolution water depth information, which is of great significance for assessing the impact of floods on power grid facilities, formulating emergency response plans, and optimizing power grid design. Refined processing methods ensure that reliable and detailed water depth data can be obtained even in narrow or complex terrain.

[0129] In summary, this application not only greatly improves the accuracy and practicality of water depth inversion through satellite-ground coordination, high-timeliness monitoring, uncertainty reduction, and refined processing of key areas, but also opens up new possibilities for research and application in related fields. It is expected to become an important tool for future hydrological monitoring and disaster management.

[0130] Figure 3 is a structural diagram of a water depth inversion device according to an embodiment of the present application, such as Figure 3 As shown, the device includes:

[0131] The acquisition module 31 is used to acquire historical remote sensing image data and historical ground water level gauge data in the target area, wherein the historical remote sensing image data includes: a first number of first historical radiation brightness values ​​and a second number of second historical radiation brightness values; the historical ground water level gauge data includes: a third number of first historical real water depth information and a fourth number of second historical real water depth information, wherein the second number is smaller than the first number, and the fourth number is smaller than the third number.

[0132] The determination module 32 is configured to determine a physical model for representing the correlation between the first historical radiance value and the first historical true water depth information.

[0133] The first processing module 33 is configured to input the second historical radiance value into the physical model to obtain predicted water depth information output by the physical model.

[0134] The correction module 34 is used to correct the predicted water depth information using the second historical real water depth information, and update the physical model according to the correction result to obtain the target physical model.

[0135] The second processing module 35 is used to input the current radiation brightness value into the target physical model to obtain the target predicted water depth information output by the target physical model.

[0136] Optionally, the determination module 32 is further used to perform the following steps: obtaining the daily average change value of the solar radiation intensity of the target area within a preset historical time period; if the daily average change value of the solar radiation intensity is within a first preset interval, using a single-band model to determine the correlation relationship, wherein the single-band model is used to determine the linear or nonlinear relationship between the water depth and the reflectivity of the preset band and the radiation brightness value; if the daily average change value of the solar radiation intensity is within a second preset interval, using a dual-band model to determine the correlation relationship, wherein the dual-band model is used to determine the linear or nonlinear relationship between the water depth and the ratio or logarithm of the difference of the reflectivity of two different bands and the radiation brightness value, and the maximum value of the second preset interval is 1 / 4. The minimum value is greater than the maximum value of the first preset interval; if the daily average change value of the solar radiation intensity is within the third preset interval, the logarithmic ratio model is used to determine the correlation relationship, wherein the logarithmic ratio model is used to determine the linear or nonlinear relationship between the natural logarithm of the water depth and reflectivity ratio and the radiation brightness value, and the minimum value of the third preset interval is greater than the maximum value of the second preset interval; if the daily average change value of the solar radiation intensity is within the fourth preset interval, the multi-band model is used to determine the correlation relationship, wherein the multi-band model is used to determine the linear or nonlinear relationship between the water depth and the reflectivity and radiation brightness values ​​of multiple preset bands, and the minimum value of the fourth preset interval is greater than the maximum value of the third preset interval.

[0137] Optionally, the first preset interval is 45W / m 2 Up to 55W / m 2 ; The second preset interval is 100W / m 2 Up to 180W / m 2 ; The third preset interval is 200W / m 2 Up to 220W / m 2 ; The fourth preset interval is 300W / m 2 Up to 350W / m 2 .

[0138] Optionally, determining the correlation relationship using a dual-band model includes: determining the correlation relationship using the following formula: Z = (1 / fk i )(lnr Bi -X i ) Where Z is the relative water depth, f is the length of the water path, k i is the effective attenuation coefficient of water, r Bi is the bottom reflectivity at water depth Z, L i is the radiation brightness value at water depth Z, L si is the radiation brightness value at the preset water depth, R i It is a function that represents the relationship between solar radiation value, transmission flux at the interface between atmosphere and water body, and refractive index of water surface.

[0139] Optionally, the correction module 34 is also used to perform the following steps: dividing the second historical real water depth information into training set data and verification set data; using the training set data to train the polynomial regression model, and using the verification set data to verify the mean square error between the output value of the polynomial regression model and the true value, and when the mean square error is less than a preset threshold, obtaining a trained polynomial regression model; inputting the predicted water depth information into the trained polynomial regression model to obtain a correction result for the predicted water depth information.

[0140] Optionally, the target predicted water depth information includes: a plurality of discrete water depth measurement points and water depth values ​​corresponding to the water depth measurement points, wherein each water depth measurement point has a geographic coordinate identifier. The water depth inversion device is further configured to, after obtaining the target predicted water depth information output by the target physical model, perform the following steps: converting the geographic coordinate identifiers of the water depth measurement points into plane coordinates suitable for visual display; constructing a visual graphic based on the converted plane coordinates and the water depth values ​​corresponding to the water depth measurement points; setting different color mapping rules based on different ranges of water depth values ​​to color render the constructed visual graphic; and displaying the rendered visual graphic on a display device.

[0141] Optionally, the acquisition module 31 is also used to perform the following steps: determining an area whose distance from a transmission line of a preset length meets a preset distance as a target area; downloading initial remote sensing image data of the target area within a preset time window from a satellite image database, selecting remote sensing image data with a resolution greater than a first preset threshold and a cloud coverage percentage less than a second preset threshold from the initial remote sensing image data, performing radiation correction, geometric correction and atmospheric correction on the selected remote sensing image data to obtain historical remote sensing image data; determining the target acquisition time corresponding to the initial remote sensing image data, and obtaining historical ground water level gauge data corresponding to the target acquisition time.

[0142] It should be noted that the above Figure 3The modules in the embodiment can be program modules (for example, a set of program instructions that implement a specific function) or hardware modules. For the latter, they can be expressed in the following forms, but are not limited to these: the expression form of each of the above modules is a processor, or the functions of each of the above modules are implemented by a processor.

[0143] It should be noted that Figure 3 The preferred implementation of the embodiment shown can be found in Figure 1 The relevant description of the illustrated embodiment will not be repeated here.

[0144] Figure 4 FIG1 shows a hardware structure block diagram of a computer terminal for implementing a water depth inversion method. Figure 4 As shown, the computer terminal 40 may include one or more (402a, 402b, ..., 402n are shown in the figure) processors 402 (the processor 402 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 404 for storing data, and a transmission module 406 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It will be understood by those skilled in the art that Figure 4 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 4 More or fewer components than shown, or with Figure 4 Different configurations shown.

[0145] It should be noted that the one or more processors 402 and / or other data processing circuits described above may generally be referred to herein as "data processing circuitry." The data processing circuitry may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuitry may be a single, independent processing module, or may be incorporated in whole or in part into any of the other components of the computer terminal 40. As described in the embodiments of the present application, the data processing circuitry serves as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0146] The memory 404 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the water depth inversion method in the embodiment of the present application. The processor 402 executes various functional applications and data processing by running the software programs and modules stored in the memory 404, thereby implementing the above-mentioned water depth inversion method. The memory 404 may include a high-speed random access memory and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 404 may further include a memory remotely located relative to the processor 402, and these remote memories may be connected to the computer terminal 40 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0147] The transmission module 406 is configured to receive or transmit data via a network. A specific example of the aforementioned network may include a wireless network provided by the communications provider of the computer terminal 40. In one embodiment, the transmission module 406 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In another embodiment, the transmission module 406 may be a radio frequency (RF) module, which is configured to communicate with the Internet wirelessly.

[0148] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 40 .

[0149] It should be noted that, in some optional embodiments, the above Figure 4 The computer terminal shown may include hardware elements (including circuits), software elements (including computer code stored on a computer-readable medium), or a combination of hardware elements and software elements. Figure 4 This is merely one example of a particular embodiment and is intended to illustrate the types of components that may be present in the computer terminal described above.

[0150] It should be noted that Figure 4 The computer terminal shown is used to execute Figure 1 The water depth inversion method shown, therefore the relevant explanations in the execution method of the above command are also applicable to the electronic device and will not be repeated here.

[0151] An embodiment of the present application further provides a non-volatile storage medium, which includes a stored program, wherein when the program is executed, the device where the storage medium is located is controlled to execute the above water depth inversion method.

[0152] A program for a non-volatile storage medium to perform the following functions: obtaining historical remote sensing image data and historical ground water level gauge data in a target area, wherein the historical remote sensing image data includes: a first number of first historical radiation brightness values ​​and a second number of second historical radiation brightness values, and the historical ground water level gauge data includes: a third number of first historical real water depth information and a fourth number of second historical real water depth information, wherein the second number is smaller than the first number, and the fourth number is smaller than the third number; determining a physical model for representing the correlation between the first historical radiation brightness value and the first historical real water depth information; inputting the second historical radiation brightness value into the physical model to obtain predicted water depth information output by the physical model; correcting the predicted water depth information using the second historical real water depth information, and updating the physical model according to the correction result to obtain a target physical model; inputting the current radiation brightness value into the target physical model to obtain target predicted water depth information output by the target physical model.

[0153] An embodiment of the present application further provides an electronic device, comprising: a memory and a processor, wherein the processor is configured to run a program stored in the memory, wherein the above water depth inversion method is executed when the program is run.

[0154] The processor is used to run a program that performs the following functions: obtaining historical remote sensing image data and historical ground water level gauge data in a target area, wherein the historical remote sensing image data includes: a first number of first historical radiation brightness values ​​and a second number of second historical radiation brightness values, and the historical ground water level gauge data includes: a third number of first historical real water depth information and a fourth number of second historical real water depth information, wherein the second number is smaller than the first number, and the fourth number is smaller than the third number; determining a physical model for representing the correlation between the first historical radiation brightness value and the first historical real water depth information; inputting the second historical radiation brightness value into the physical model to obtain predicted water depth information output by the physical model; correcting the predicted water depth information using the second historical real water depth information, and updating the physical model according to the correction result to obtain a target physical model; inputting the current radiation brightness value into the target physical model to obtain target predicted water depth information output by the target physical model.

[0155] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0156] In the above embodiments of the present application, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0157] In the above-mentioned embodiments of the present application, the collected information is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with relevant laws, regulations and standards, take necessary protection measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0158] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of the units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0159] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0160] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0161] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the relevant technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0162] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A water depth inversion method, characterized in that: include: Acquire historical remote sensing image data and historical surface water level gauge data in a target area, wherein the historical remote sensing image data includes: a first quantity of first historical radiance values ​​and a second quantity of second historical radiance values, and the historical surface water level gauge data includes: a third quantity of first historical true water depth information and a fourth quantity of second historical true water depth information, wherein the second quantity is smaller than the first quantity, and the fourth quantity is smaller than the third quantity; Determining a physical model for representing an association relationship between the first historical radiance value and the first historical true water depth information; Inputting the second historical radiance value into the physical model to obtain predicted water depth information output by the physical model; Correcting the predicted water depth information using the second historical real water depth information, and updating the physical model according to the correction result to obtain a target physical model; Inputting the current radiation brightness value into the target physical model to obtain target predicted water depth information output by the target physical model; Determining a physical model for representing the correlation between the first historical radiance value and the first historical true water depth information includes: obtaining a daily average change value of solar radiation intensity of the target area within a preset historical time period; if the daily average change value of solar radiation intensity is within a first preset interval, determining the correlation relationship using a single-band model, wherein the single-band model is used to determine a linear or nonlinear relationship between water depth and the reflectivity and radiance value of a preset band; if the daily average change value of solar radiation intensity is within a second preset interval, determining the correlation relationship using a dual-band model, wherein the dual-band model is used to determine a linear or nonlinear relationship between water depth and the logarithm of the ratio or difference of the reflectivity of two different bands and the radiance value, and the minimum value of the second preset interval is greater than the maximum value of the first preset interval; if the daily average change value of solar radiation intensity is within a third preset interval, determining the correlation relationship using a logarithmic ratio model, wherein the logarithmic ratio model is used to determine a linear or nonlinear relationship between the natural logarithm of the ratio of water depth to reflectivity and the radiance value, and the minimum value of the third preset interval is greater than the maximum value of the second preset interval; If the daily average change in solar radiation intensity is within a fourth preset interval, determining the correlation relationship using a multi-band model, wherein the multi-band model is used to determine a linear or nonlinear relationship between water depth and reflectivity and radiance values ​​in multiple preset bands, and the minimum value of the fourth preset interval is greater than the maximum value of the third preset interval; The first preset interval is 45 W / m² to 55 W / m²; the second preset interval is 100 W / m² to 180 W / m²; the third preset interval is 200 W / m² to 220 W / m²; and the fourth preset interval is 300 W / m² to 350 W / m².

2. The method according to claim 1, characterized in that Correcting the predicted water depth information using the second historical real water depth information includes: Dividing the second historical real water depth information into training set data and validation set data; The polynomial regression model is trained using the training set data, and the mean square error between the output value of the polynomial regression model and the true value is verified using the validation set data, and when the mean square error is less than a preset threshold, a polynomial regression model that has completed training is obtained; The predicted water depth information is input into the trained polynomial regression model to obtain a correction result of the predicted water depth information.

3. The method according to claim 1, characterized in that The target predicted water depth information includes: a plurality of discrete water depth measurement points and water depth values ​​corresponding to the water depth measurement points, wherein each of the water depth measurement points has a geographic coordinate identifier; After obtaining the target predicted water depth information output by the target physical model, the method further includes: Converting the geographic coordinate identifier of the water depth measurement point into a plane coordinate suitable for visual display; Constructing a visualization graph based on the converted plane coordinates and the water depth values ​​corresponding to the water depth measurement points; Different color mapping rules are set according to different ranges of water depth values ​​to render the constructed visualization graphics; The rendered visualization graphics are displayed on the display device.

4. The method according to claim 1, wherein Acquire historical remote sensing imagery data and historical surface water level gauge data in the target area, including: Determine an area whose distance from a transmission line of a preset length meets a preset distance as the target area; downloading initial remote sensing image data of the target area within a preset time window from a satellite image database, selecting remote sensing image data having a resolution greater than a first preset threshold and a cloud cover percentage less than a second preset threshold from the initial remote sensing image data, and performing radiation correction, geometric correction, and atmospheric correction on the selected remote sensing image data to obtain historical remote sensing image data; The target acquisition time corresponding to the initial remote sensing image data is determined, and historical surface water level gauge data corresponding to the target acquisition time is obtained.

5. A water depth inversion device, characterized in that: include: an acquisition module, configured to acquire historical remote sensing image data and historical surface water level gauge data in a target area, wherein the historical remote sensing image data includes: a first number of first historical radiance values ​​and a second number of second historical radiance values; and the historical surface water level gauge data includes: a third number of first historical true water depth information and a fourth number of second historical true water depth information, wherein the second number is smaller than the first number, and the fourth number is smaller than the third number; a determination module, configured to determine a physical model for representing an association relationship between the first historical radiance value and the first historical true water depth information; a first processing module, configured to input the second historical radiance value into the physical model to obtain predicted water depth information output by the physical model; a correction module, configured to correct the predicted water depth information using the second historical real water depth information, and update the physical model according to the correction result to obtain a target physical model; A second processing module is used to input the current radiation brightness value into the target physical model to obtain the target predicted water depth information output by the target physical model; The determination module is further configured to perform the following steps: obtaining the daily average change value of the solar radiation intensity of the target area within a preset historical time period; if the daily average change value of the solar radiation intensity is within a first preset interval, determining the association relationship using a single-band model, wherein the single-band model is used to determine the linear or nonlinear relationship between the water depth and the reflectivity and radiance value of the preset band; if the daily average change value of the solar radiation intensity is within a second preset interval, determining the association relationship using a dual-band model, wherein the dual-band model is used to determine the linear or nonlinear relationship between the water depth and the logarithm of the ratio or difference of the reflectivity of two different bands and the radiance value, and the minimum value of the second preset interval is greater than the maximum value of the first preset interval; if the daily average change value of the solar radiation intensity is within a third preset interval, determining the association relationship using a logarithmic ratio model, wherein the logarithmic ratio model is used to determine the linear or nonlinear relationship between the natural logarithm of the ratio of water depth to reflectivity and the radiance value, and the minimum value of the third preset interval is greater than the maximum value of the second preset interval; If the daily average change in solar radiation intensity is within a fourth preset interval, determining the correlation relationship using a multi-band model, wherein the multi-band model is used to determine a linear or nonlinear relationship between water depth and reflectivity and radiance values ​​in multiple preset bands, and the minimum value of the fourth preset interval is greater than the maximum value of the third preset interval; The first preset interval is 45 W / m² to 55 W / m²; the second preset interval is 100 W / m² to 180 W / m²; the third preset interval is 200 W / m² to 220 W / m²; and the fourth preset interval is 300 W / m² to 350 W / m².

6. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored program, wherein when the program is executed, the device where the non-volatile storage medium is located is controlled to execute the water depth inversion method according to any one of claims 1 to 4.

7. An electronic device, characterized in that: include: A memory and a processor, wherein the processor is used to run a program stored in the memory, wherein the water depth inversion method according to any one of claims 1 to 4 is executed when the program is run.

8. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the water depth inversion method according to any one of claims 1 to 4 is implemented.

Images