A method, system, device, and storage medium for identifying floodwater bodies

Abstract

The application provides a floodwater body identification method, system, device and storage medium, and belongs to the technical field of floodwater body identification. The method comprises the following steps: acquiring GRD image data of a target region in a set time period; acquiring a time sequence of a grid point for each pixel of the GRD image data, inputting the time sequence into a harmonic regression model, and obtaining a predicted value of the pixel; performing abnormal image identification according to a difference between an observed value and the predicted value, and obtaining a binary flood abnormal image set corresponding to all the GRD images; statistically analyzing the precipitation in a set time period before and after each flood abnormal image in the binary flood abnormal image set, comparing the precipitation with average cumulative precipitation in the same period, retaining a region in the flood abnormal image that is greater than the average cumulative precipitation, and obtaining a floodwater body identification result of the target region. The method realizes stable identification of floodwater bodies in different regional environments and has high identification accuracy.

Description

Technical Field

[0001] This invention belongs to the field of flood identification technology, specifically relating to a method, system, device, and storage medium for identifying flood water bodies. Background Technology

[0002] Flooding, a natural phenomenon triggered by heavy rainfall, persistent downpours, or prolonged concentrated rainfall, is one of the most frequent and destructive natural disasters globally. Therefore, timely and accurate monitoring and assessment of flood inundation areas are crucial for disaster prevention and mitigation. With the comprehensive coverage and rapid development of global satellites, all-weather disaster identification is now possible, overcoming the limitations of traditional flood monitoring methods, improving monitoring efficiency, and saving time and manpower costs.

[0003] The Sentinel-1 satellite, equipped with a synthetic aperture radar (SAR) system, acquires raw SAR imagery data. Existing methods for flood identification based on SAR imagery data mainly include visual interpretation, threshold segmentation, machine learning, and deep learning. Visual interpretation is not only time-consuming and labor-intensive but also susceptible to the experience and knowledge level of the interpreters, making it difficult to guarantee accuracy and consistency. Threshold segmentation is simple and easy to implement, but in complex terrain and weather conditions, factors such as surface roughness and vegetation cover make threshold selection difficult, affecting identification accuracy. Machine learning requires a large number of training samples and computational resources, and the model's generalization ability is limited. Deep learning model training is complex, consumes significant computational resources, and its accuracy is easily affected by data quality and application scenarios. Therefore, existing methods all have certain limitations in the accuracy of flood identification. Summary of the Invention

[0004] To improve the accuracy of flood water body identification, this invention provides a flood water body identification method, system, device, and storage medium.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for identifying flood-prone water bodies includes the following steps:

[0007] Acquire GRD image data of the target area within a set time period; obtain the observed values ​​of the backscattering intensity of pixels within the set time period based on the GRD image data;

[0008] The time series of grid points is obtained pixel by pixel from the GRD image data; the time series is input into the harmonic regression model to obtain the predicted value of the backscattering intensity of the pixel;

[0009] Anomaly identification is performed on the observed and predicted values ​​according to the anomaly identification criteria to obtain a set of binary flood anomaly images corresponding to all GRD images within a set time period.

[0010] The precipitation amount of each image in the binarized flood anomaly image within a set time period is statistically analyzed and compared with the average cumulative precipitation within the same time period. Binarized flood anomaly images with precipitation exceeding the average cumulative precipitation are retained. The flood water body identification result of the target area is obtained based on the retained binarized flood anomaly images.

[0011] Preferably, the harmonic regression model is as follows:

[0012]

[0013] Where x represents the Julian day, T represents the number of days in a year, a represents the coefficient of the overall value, b represents the coefficient of trend change, and c and d represent the coefficients of seasonal change. This represents the predicted value at time x on the Julian Day.

[0014] Preferably, the outlier identification criterion is as follows: the difference between the observed value and the model prediction value of a pixel is less than an empirical threshold and lower than the minimum threshold calculated using the Pauta criterion method.

[0015] Preferably, before obtaining the time series of raster points pixel by pixel from the GRD image data, the method further includes preprocessing the GRD image data, specifically including the following steps:

[0016] Remove regions in GRD image data where the signal-to-noise ratio (SNR) dB value is less than a set value;

[0017] Remove mountain shadows, bare ground, water bodies, and areas covered by ice and snow from the image;

[0018] The images acquired from the ascending and descending orbits within a set time are combined by minimizing the total number of images, and the orbit with the most images is selected as the GRD image dataset.

[0019] Preferably, it also includes outputting the flood water body identification results of the target area through the GEE cloud platform, wherein the GEE cloud platform can select annual scale, seasonal scale and event scale for the flood water body identification results.

[0020] Preferably, the method further includes estimating the coefficients of the harmonic regression model using ordinary least squares (OLS).

[0021] The present invention also provides a flood water body identification system, specifically including:

[0022] The image acquisition module is used to acquire GRD image data of the target area within a set time period; and to obtain the observed values ​​of the backscattering intensity of the pixels within the set time period based on the GRD image data.

[0023] The model processing module is used to obtain the time series of grid points pixel by pixel from the GRD image data; and input the time series into the harmonic regression model to obtain the predicted value of the backscattering intensity of the pixel.

[0024] Anomaly identification module is used to identify anomalies in the observed and predicted values ​​according to anomaly identification criteria, and obtain a set of binarized flood anomaly images corresponding to all GRD images within a set time period;

[0025] The flood identification module is used to count the precipitation of each image in the binarized flood anomaly image within a set time period, compare it with the average cumulative precipitation within the same time period, retain the binarized flood anomaly images with precipitation exceeding the average cumulative precipitation, and obtain the flood water body identification result of the target area based on the retained binarized flood anomaly images.

[0026] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps described in the flood water body identification method.

[0027] The present invention also provides a computer-readable storage medium storing a computer program that, when loaded by a processor, can execute the steps described in the flood water body identification method.

[0028] The flood water body identification method provided by this invention has the following beneficial effects:

[0029] This invention acquires GRD image data of a target area within a set time period, obtaining observed values ​​of backscattering intensity of pixels within that time period and a time series of raster points obtained pixel-by-pixel from the GRD image data. Based on the time series, the predicted values ​​of backscattering intensity of pixels are obtained using the coefficients of a harmonic regression model. The harmonic regression model can flexibly fit periodic fluctuations in the image, improving the accuracy of the predictions. Anomaly image identification is performed based on the difference between observed and predicted values ​​according to anomaly identification criteria, effectively distinguishing between flooded and non-flooded water bodies. By statistically analyzing the precipitation in each image of the flood anomaly image within a set time period and comparing it with the average cumulative precipitation within the same time period, and considering precipitation deviation factors directly related to flooding, the reliability of flood water body identification in the target area is enhanced. Simultaneously, it can achieve stable identification of flood water bodies in different regional environments, demonstrating high universality. Attached Figure Description

[0030] To more clearly illustrate the embodiments and design schemes of the present invention, the accompanying drawings required for this embodiment will be briefly described below. The drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] Figure 1 This is a flowchart of a flood water body identification method according to an embodiment of the present invention.

[0032] Figure 2 This is a comparison diagram of a flood event occurring in region A according to an embodiment of the present invention, wherein, Figure 2 (a) is the SAR image of region A on October 31, 2023; Figure 2 (b) is a SAR image of region A on November 10, 2023.

[0033] Figure 3 This is the flood water body identification result for region A in this embodiment of the invention, wherein, Figure 3 (a) is a graph showing the monitoring results of emergency management services; Figure 3 (b) is a flood inundation extent map drawn using a standardized method; Figure 3 (c) is a flood map generated by the method used in this invention.

[0034] Figure 4 This is a diagram showing the flood verification results in an embodiment of the present invention. Detailed Implementation

[0035] To enable those skilled in the art to better understand and implement the technical solutions of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be construed as limiting the scope of protection of the present invention.

[0036] Example

[0037] Sentinel-1A's Level-1 products include SLC (SingleLook Complex) and GRD (Ground Range Detected). GRD data is obtained by performing multi-look processing and georegistration on the raw Sentinel-1 SAR data.

[0038] This invention provides a method for identifying flood-prone water bodies, such as... Figure 1 As shown, the GEE cloud platform specifically includes the following steps:

[0039] S1. Obtain ground distance GRD (VH band) images from 2015 to the present using the Sent inel-1 method. Perform preprocessing on individual GRD images, specifically including the following steps:

[0040] (1) Generate a new edge file (the actual range of the image is cropped by an 8km buffer), and remove areas on the image with a signal-to-noise ratio of less than -40 dB.

[0041] (2) Remove mountain shadows; the current data comes from the publicly available datasets from GEE: Slope (WWF / HydroSHEDS / 03VFDEM) and HAND (users / gena / GlobalHAND / 30m / hand-1000). The slope threshold is set to 5° and the HAND threshold is set to 5m to mask the shadowed areas of flat areas where the vertical distance from the nearest water system is greater than 5m.

[0042] (3) Remove bare land, water bodies, ice and snow areas; global land use data (ESA / WorldCover / v200).

[0043] (4) Image selection. Images acquired on both ascending and descending orbits within a 6-day period are mosaicked (minimum value synthesis). The orbit with the most images is selected as the image set to ensure the most comprehensive and representative coverage of the flood event. This helps to retain the features most relevant to the flood while reducing redundant data.

[0044] S2. Anomaly detection is performed based on the preprocessed GRD image data.

[0045] S21. Obtain the time series of grid points pixel by pixel from the preprocessed GRD image, input the time series into the harmonic regression model, and use the ordinary least squares (OLS) method to estimate the model coefficients to capture the dynamic changes of the land surface. The model includes seasonal, trend and constant components.

[0046]

[0047] Where x represents Julian Day, T represents the number of days in a year (T=365), a represents the coefficient of the overall value, b represents the coefficient of trend change, c and d represent the coefficients of seasonal change, and y(x) represents the predicted value at time x on Julian Day.

[0048] S22. The difference between the actual observed value of each pixel in the preprocessed SAR image and the predicted value output by the harmonic regression model is used as the basis for anomaly identification. The observed value of each pixel is the backscattering intensity time series of that point obtained by retrieval within a certain time range. After a flood, the scattering intensity of the pixel will suddenly decrease, i.e., the observed value decreases, and the difference between the observed value and the predicted value will be large. This phenomenon is used for flood identification. When the difference between the observed value and the predicted value is regarded as an anomaly, the following two specific criteria must be met: a) less than an empirically set predefined threshold of -5; b) lower than the minimum threshold calculated using the Pauta criterion method. Each GRD image in the image set has a corresponding binary flood anomaly image (0: normal point, 1: anomaly point).

[0049] S23. For each image where anomalies were identified, the ERA5P cumulative precipitation for the 10 days before and after that image was recalculated, and these values ​​were compared with the multi-year average cumulative precipitation for the same period as a reference background value. Regions in the flood anomaly images where the cumulative precipitation exceeded the corresponding reference background precipitation value were retained, while other regions were ignored. This refinement process ensured that only anomalies directly related to significant precipitation deviations were considered in the final flood identification results.

[0050] S24. Use connectivity functions to obtain the target objects, count them (using 8-neighborhood statistical analysis), and remove discrete points with fewer than 6 points. Removing small, disconnected flood areas helps improve the overall accuracy and reliability of the flood detection process.

[0051] S25. The GRD image data from the orbit raising and lowering detections are processed by S21 to S24 respectively and then merged.

[0052] S3. Image export yields the flood water body identification results for the target area. The GEE cloud platform allows users to select three scales: annual, seasonal, and event-based, catering to diverse user needs for long-term trend analysis, seasonal change studies, and specific event monitoring.

[0053] Taking the flood event that occurred in region A in November 2023 as an example:

[0054] Figure 2 Sentinel SAR imagery acquired over the timeframe of a flood event visually reflects the extent of inundation at the time of the flood. Flooded water bodies have very low BCSG, which helps to quickly and accurately locate flood-inundated areas.

[0055] Figure 3 For the results of flood water body identification, Figure 3 (a) is the Emergency Management Service Monitoring Result Map (EMSR). Figure 3(b) is a flood inundation extent map (Z-score) drawn using a standardized method. Figure 3 (c) shows the flood mapping (HRM) generated by the method used in this invention. It can be seen that the flood body identification results generated by this product are quite consistent with the spatial distribution of EMSR.

[0056] Figure 4 To validate the results at random points, 1000 validation points were randomly selected, and the Z-score and HRM were tested using the EMSR (Emergency Response System) as the validation benchmark. If a point fell within the flood zone defined by the EMSR, it was considered a flood point and marked as 1; conversely, if the point was outside the EMSR zone, it was considered a non-flood point and marked as 0. To measure the accuracy of the classification, the overall precision was used as a key parameter. The overall precision calculated using the Z-score method was 0.74, meaning that 74% of the selected 1000 random points correctly classified the flood state (whether flood or non-flood). However, when the model proposed in this patent was used for flood water body classification, the overall precision improved to 0.78, meaning that 78% of the measurement points correctly classified the flood state. Compared to the Z-score method, the overall precision of this patent method improved by 5 percentage points, a statistically significant improvement, indicating that the patent model has superior performance in flood water body identification compared to traditional methods.

[0057] The present invention also provides a flood water body identification system, specifically including:

[0058] The image acquisition module is used to acquire GRD image data of the target area within a set time period; and to obtain the observed values ​​of the backscattering intensity of pixels within the set time period based on the GRD image data.

[0059] The model processing module is used to obtain the time series of grid points on a pixel-by-pixel basis from GRD image data; the time series is input into the harmonic regression model to obtain the predicted value of the backscattering intensity of the pixel.

[0060] The anomaly identification module is used to identify anomalies in observed and predicted values ​​according to anomaly identification criteria, and obtain a set of binary flood anomaly images corresponding to all GRD images within a set time period.

[0061] The flood identification module is used to statistically analyze the precipitation of each image in the binarized flood anomaly image within a set time period, compare it with the average cumulative precipitation within the same time period, retain the binarized flood anomaly images with precipitation exceeding the average cumulative precipitation, and obtain the flood water body identification result of the target area based on the retained binarized flood anomaly images.

[0062] The modules in the aforementioned flood water body identification system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0063] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory. The processor executes the computer program to implement the steps in an embodiment of a flood water body identification method. Specific implementation methods can be found in the method embodiments, and will not be repeated here.

[0064] Furthermore, the present invention also provides a non-transitory computer-readable storage medium containing instructions on which a computer program is stored. For example, a memory containing instructions that can be executed by a processor of a computer device to perform the aforementioned method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc. When the computer program is executed by the processor, it can implement the steps in an embodiment of a flood water body identification method. Specific implementation methods can be found in the method embodiments, which will not be repeated here.

[0065] Those skilled in the art will understand that embodiments of the present invention can provide methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0066] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0067] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0068] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0069] It should be noted that the specific embodiments described above enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any way. Therefore, although the present invention has been described in detail in this specification and embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the present invention; and all technical solutions and improvements that do not depart from the spirit and scope of the present invention are covered within the protection scope of the present invention patent. No reference numerals in the claims should be construed as limiting the scope of the claims. Any simple variations or equivalent substitutions of technical solutions that can be readily obtained by those skilled in the art within the scope of the technology disclosed in the present invention are within the protection scope of the present invention.

Claims

1. A method for identifying a flood water body, characterized in that, The method comprises the following steps: Obtaining GRD image data of a target area within a set time period; obtaining observation values of pixel point backscattering intensity within the set time period according to the GRD image data; Obtaining a time sequence of a grid point for each pixel of the GRD image data; inputting the time sequence into a harmonic regression model to obtain prediction values of the pixel point backscattering intensity; the harmonic regression model is specifically: ； wherein x represents the Julian day, T represents the number of days per year, a represents a coefficient of the overall value, b represents a coefficient of the trend change, c, d represent coefficients of the seasonal change, represents the predicted value at the Julian day x; Performing abnormal identification on the observation values and the prediction values according to an abnormal point identification criterion to obtain a set of binary flood abnormal images corresponding to all GRD images within the set time period; Statistically counting accumulated precipitation within a 10-day period before and after each binary flood abnormal image, comparing the counted accumulated precipitation with average accumulated precipitation within the same period, retaining the binary flood abnormal image whose precipitation exceeds the average accumulated precipitation, and obtaining a floodwater body identification result of the target area according to the retained binary flood abnormal image.

2. The method of claim 1, wherein, The abnormal point identification criterion is specifically that a difference between the observation values and the prediction values of a pixel is less than an empirical threshold value and is lower than a minimum threshold value calculated by using a Pauta criterion method.

3. The method of claim 1, wherein, Before the step of obtaining the time sequence of the grid point for each pixel of the GRD image data, the method further comprises preprocessing the GRD image data, specifically The method comprises the following steps: Eliminating a region in the GRD image data with a signal-to-noise ratio dB value less than a set value; Removing mountain shadow, bare land, water body, ice and snow-covered regions in the image; Synthesizing minimum values of images obtained by ascending tracks and descending tracks within a set time to select a track with the largest number of images as a GRD image data set.

4. The method of claim 1, wherein, The method further comprises outputting the floodwater body identification result of the target area through a GEE cloud platform, and the GEE cloud platform can select an annual scale, a seasonal scale and an event scale for the floodwater body identification result.

5. The method of claim 2, wherein, The method further comprises estimating coefficients of the harmonic regression model by using an ordinary least squares method OLS.

6. A floodwater body identification system characterized by, The method comprises: An image acquisition module is configured to obtain GRD image data of a target area within a set time period; Obtaining observation values of pixel point backscattering intensity within the set time period according to the GRD image data; A model processing module is configured to obtain a time sequence of a grid point for each pixel of the GRD image data; inputting the time sequence into a harmonic regression model to obtain prediction values of the pixel point backscattering intensity; the harmonic regression model is specifically: ； wherein x represents the Julian day, T represents the number of days per year, a represents a coefficient of the overall value, b represents a coefficient of the trend change, c, d represent coefficients of the seasonal change, represents the predicted value at the Julian day x; An abnormal point identification module is configured to perform abnormal identification on the observation values and the prediction values according to an abnormal point identification criterion to obtain a set of binary flood abnormal images corresponding to all GRD images within the set time period; A flood identification module is configured to statistically count precipitation of each image in the binary flood abnormal image within the set time period, compare the counted precipitation with average accumulated precipitation within the same period, retain the binary flood abnormal image whose precipitation exceeds the average accumulated precipitation, and obtain a floodwater body identification result of the target area according to the retained binary flood abnormal image.

7. A computer device comprising a memory, a processor, and a computer program stored on the memory, wherein the computer program comprises instructions that, when executed by the processor, cause the processor to perform the method of any one of claims 1-6. The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program can implement the steps of the method according to any one of claims 1 to 5 when loaded by the processor.

Images