A flood monitoring method, system, device and medium

Abstract

The present invention discloses a flood monitoring method, system, equipment and medium, the method comprising: using satellite remote sensing images and SAR image information for fusion, adopting different water body extraction methods for different terrain environments such as plains and towns, reducing the influence of environmental factors such as ground objects on water body target extraction, and improving the accuracy of water body identification. For the flood disaster range and water depth prediction part, real-time meteorological information is selected as the input data of the time series, combined with the geographical environment, and an LSTM artificial neural network is used as a prediction model to analyze and predict the changes in the time, rainfall, evaporation, and disaster degree changes at the flood-affected points. This method greatly reduces manual intervention, and with the support of high-time satellite remote sensing and high-precision data, it can play a greater role in flood monitoring, evaluation and early warning.

Description

Technical Field

[0001] The present invention relates to the technical field of flood disaster prevention and control, and in particular to a flood monitoring method, system, equipment and medium. Background Art

[0002] With the rapid development of information technology, it has become possible to monitor and evaluate floods using new remote sensing technologies. Ground remote sensing has advantages in providing small-scale and local point observations, while satellite remote sensing has more significant advantages in monitoring large-scale floods. Through the effective combination of different remote sensing methods, the actual needs of overall large-scale flood monitoring and small-scale local embankment breach rescue can be better met, and integrated monitoring and emergency assessment of flood disasters can be truly realized. By combining image information with data resources such as geographic data, flood disasters can be monitored and evaluated automatically and quickly, reducing human intervention, and will certainly further improve the scientific and technological benefits of flood prevention and disaster reduction.

[0003] However, most of the existing flood monitoring methods of multi-source remote sensing satellite images require water extraction first. However, due to the strong subjectivity of threshold selection, the extracted information is relatively simple, and most methods are complex to operate and have poor robustness, resulting in poor recognition accuracy. Further use of other data information, edge methods and other means is required. In addition, in flood monitoring, analysis and early warning, satellite remote sensing images are limited by weather, environment and other factors. In practical applications, they need to be combined with high-precision terrain data to calculate and predict flood depth, affected range and other data. However, there are usually relatively many variables that need to be referenced for prediction. The calculation accuracy of flood inundation range and water depth is not high enough. The accuracy of available terrain data is still far from the actual needs. Therefore, more reliable prediction methods are needed to improve accuracy and reduce human intervention. Summary of the invention

[0004] The purpose of the present invention is to provide a flood monitoring method, system, equipment and medium, which can monitor, evaluate and warn flood disasters with the support of high-time satellite remote sensing and high-precision data.

[0005] In order to achieve the above object, the present invention provides a flood monitoring method, comprising:

[0006] Acquire real-time image data within the target range, and extract the real-time image data to obtain the location range of the flood within the target range, wherein the real-time image data includes: satellite remote sensing real-time image data and SAR real-time image data;

[0007] Using geospatial tools, the location range of the flood within the target range is measured to obtain the real-time flooded area and water depth within the target range;

[0008] Real-time meteorological data within the target range is obtained, and the real-time meteorological data and the real-time flooded area and water depth are input into a preset neural network model to predict the flooded area and water depth of the target area, wherein the preset neural network model is specifically constructed as follows:

[0009] Historical flood disaster data within the target range is obtained, and the historical flood disaster data is input into a first neural network model for training to obtain a preset neural network model, wherein the flood disaster data includes: meteorological data, flooded area and water depth.

[0010] Furthermore, the real-time image data within the target range is obtained, and the real-time image data is extracted to obtain the location range of the flood within the target range, including:

[0011] The real-time image data within the target range is divided into real-time image data of plain areas and real-time image data of urban areas;

[0012] Perform image extraction on the real-time image data of the plain area to obtain the location range of floods in the plain area within the target range;

[0013] Image extraction is performed on the real-time image data of the urban area to obtain the location range of the flood in the urban area within the target range.

[0014] Furthermore, the real-time image data of the plain area is subjected to image extraction to obtain the location range of floods in the plain area within the target range, including:

[0015] The flood edge is extracted from the real-time image data of the plain area within the target range using the big law method;

[0016] If the edge of the detected water body is discontinuous, a continuous flood target boundary is obtained through surface fitting, thereby obtaining the location range of floods in the plain area within the target range.

[0017] Furthermore, the real-time image data of the urban area is extracted to obtain the location range of the flood in the urban area within the target range, including:

[0018] The real-time image data of the urban area is segmented by using the threshold segmentation method to obtain the segmentation result of the real-time image data of the urban area;

[0019] Adopting object-oriented multi-level method, constructing object-oriented multi-level decision tree through threshold;

[0020] The result of the real-time image data segmentation of the urban area is input into the object-oriented multi-level decision tree to obtain the location range of the urban area flood within the target range.

[0021] The present invention also provides a flood monitoring system, comprising: a data acquisition module, a data processing module, and a prediction module, wherein:

[0022] The data acquisition module is used to acquire real-time image data within the target range, and extract the real-time image data to obtain the location range of the flood within the target range, wherein the real-time image data includes: satellite remote sensing real-time image data and SAR real-time image data;

[0023] The data processing module is used to measure the location range of the flood within the target range using geospatial tools to obtain the real-time flooded area and water depth within the target range;

[0024] The prediction module is used to obtain real-time meteorological data within the target range, and input the real-time meteorological data and the real-time flooded area and water depth into a preset neural network model to predict the flooded area and water depth of the target area, wherein the preset neural network model is specifically constructed as follows:

[0025] Historical flood disaster data within the target range is obtained, and the historical flood disaster data is input into a first neural network model for training to obtain a preset neural network model, wherein the flood disaster data includes: meteorological data, flooded area and water depth.

[0026] Furthermore, the data acquisition module is specifically used to:

[0027] The real-time image data within the target range is divided into real-time image data of plain areas and real-time image data of urban areas;

[0028] Perform image extraction on the real-time image data of the plain area to obtain the location range of floods in the plain area within the target range;

[0029] Image extraction is performed on the real-time image data of the urban area to obtain the location range of the flood in the urban area within the target range.

[0030] Furthermore, the real-time image data of the plain area is subjected to image extraction to obtain the location range of floods in the plain area within the target range, including:

[0031] The flood edge is extracted from the real-time image data of the plain area within the target range using the big law method;

[0032] If the edge of the detected water body is discontinuous, a continuous flood target boundary is obtained through surface fitting, thereby obtaining the location range of floods in the plain area within the target range.

[0033] Furthermore, the real-time image data of the urban area is extracted to obtain the location range of the flood in the urban area within the target range, including:

[0034] The real-time image data of the urban area is segmented by using the threshold segmentation method to obtain the segmentation result of the real-time image data of the urban area;

[0035] Adopting object-oriented multi-level method, constructing object-oriented multi-level decision tree through threshold;

[0036] The result of the real-time image data segmentation of the urban area is input into the object-oriented multi-level decision tree to obtain the location range of the urban area flood within the target range.

[0037] The present invention also provides a computer terminal device, comprising: one or more processors; a memory coupled to the processor and used to store one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the flood monitoring method as described in any one of the above items.

[0038] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any of the flood monitoring methods described above.

[0039] Compared with the prior art, the flood monitoring method, system, device and medium provided by the present invention have the following beneficial effects:

[0040] The present invention uses satellite remote sensing images and SAR image information for fusion, adopts different water body extraction methods for different terrain environments such as plains and towns, reduces the impact of environmental factors such as ground objects on water body target extraction, and improves the accuracy of water body identification. For the flood disaster range and water depth prediction part, real-time meteorological information is selected as the input data of the time series, combined with the geographical environment, and the LSTM artificial neural network is used as the prediction model to analyze and predict the changes in time, rainfall, evaporation, and the disaster degree of the flood-affected points. This method greatly reduces human intervention and, with the support of high-time satellite remote sensing and high-precision data, can play a greater role in flood monitoring, assessment and early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a flow chart of a flood monitoring method provided by the present invention;

[0042] Figure 2 It is a schematic diagram of the process of the flood information extraction technology provided by the present invention;

[0043] Figure 3 It is a schematic diagram of the multi-level decision-making process provided by the present invention;

[0044] Figure 4 It is a schematic diagram of a process for estimating the flooded area and water depth provided by the present invention;

[0045] Figure 5 It is a schematic diagram of the process of predicting flood water volume using the LSTM model provided by the present invention;

[0046] Figure 6 It is a structural schematic diagram of a flood monitoring system provided by the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] It should be understood that the step numbers used in this document are only for convenience of description and are not intended to limit the order in which the steps are executed.

[0049] It should be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0050] The terms “include” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0051] The term "and / or" means and includes any and all possible combinations of one or more of the associated listed items.

[0052] like Figure 1 As shown, a flood monitoring method provided by the present invention at least includes steps S1-S3, and the specific steps are as follows:

[0053] S1, acquiring real-time image data within a target range, and extracting the real-time image data to obtain the location range of the flood within the target range, wherein the real-time image data includes: satellite remote sensing real-time image data and SAR real-time image data;

[0054] Specifically, the present invention uses multi-source fusion information images to extract water body boundaries, fuses satellite remote sensing images with SAR image information, selects different extraction methods for water bodies in different locations according to their characteristics, and calculates the shape and position of the water body in combination with factors such as image resolution.

[0055] It should be noted that if Figure 2 As shown, when extracting the flood location range within the target range, it is first necessary to divide the real-time image data within the target range into plain area images and urban area images, and then extract the flood range in the plain area image and the flood range in the urban area image in turn;

[0056] Specifically, the flood range extraction in the plain area image is as follows:

[0057] The terrain in the plain area is relatively flat. Although the images are slightly different at different times and seasons, water bodies and bare land can be distinguished well due to the small number of surrounding objects. The image is converted to grayscale and detected using the OTSU method.

[0058] The pixels are divided into two categories according to the pixel value m, and the w0 value is set as the most appropriate threshold. If the number of water pixels is x0, the proportion of the total number of pixels in the image is y0, and the average pixel value of the water is m0; the number of bare ground pixels is x1, the proportion of the total number of pixels in the image is y1, and the average pixel value is m1; the total number of pixels in the image is x, and the average pixel value is m. The two have the following relationship:

[0059]

[0060] y0+y1=1

[0061] The formula for calculating the between-class variance is:

[0062] g=y0*(m0-m) 2 +y1*(m1-m) 2

[0063] g=y0*y1*(m0-m1) 2

[0064] The inter-class variance of the two classes is maximized to extract the edge of the water body. After the edge of the water body contour is detected, the detection edge may be discontinuous. Through surface fitting, a continuous water body target boundary can be finally obtained.

[0065] Specifically, the scheme for extracting flood range in urban area images is as follows:

[0066] There may be a large number of buildings and trees in the town that affect the identification of water bodies, so the water bodies need to be finely extracted. First, in the image segmentation part, the KI binary threshold segmentation method is used for image segmentation. The discrimination criterion of this method under Gaussian distribution is:

[0067]

[0068] Among them, S is the mean grayscale of the target and background, P is the prior probability of segmenting the target, and when T is the optimal segmentation threshold, K(T) can reach the minimum value.

[0069] Based on the water body segmentation, the present invention adopts an object-oriented multi-level method, constructs an object-oriented multi-level decision tree through thresholds, thereby extracting the water body and classifying different ground objects in a hierarchical manner through specific rules. The first level mainly divides the flood coverage area based on the remote sensing image after the flood; the second level mainly focuses on the ground objects in the image of the flood coverage area, including bare land, buildings, trees, etc. Further division of ground objects can facilitate subsequent damage assessment of buildings and other equipment in the flooded area. As shown in Table 1:

[0070] Table 1 Hierarchical order diagram

[0071]

[0072] For the segmentation of objects covered by flood in remote sensing images, it is necessary to use the image information before the flood to perform the second-level segmentation to achieve the purpose of distinguishing bare land, buildings and trees. By dividing different objects twice through segmentation parameters, a multi-layer classifier with target category information can be obtained. For example, the decision rule construction for buildings is as follows:

[0073] DR: if(ρ 建筑 <T 建筑 )and(area>T a

[0074] and(width>T w )then

[0075] class label = object 建筑

[0076] Among them, ρ is the reflectivity of the object in the image, T is the threshold, and T a With T wThey are area threshold and width threshold respectively. The pixels that meet the threshold conditions are combined into ground objects through cluster analysis, and segmented by thresholds to suppress noise interference and maintain the regional continuity of ground objects, and finally segment the required building vector layer. The second-level division is mainly aimed at the following segmentation parameters: color, density and shape, etc. Different weights are set for each segmentation parameter. The color weight mainly measures the band changes in the image, the density weight mainly targets the correlation and integrity of objects in the image, and the shape weight mainly measures the shape changes of the covered area before and after the flood disaster. In the brightness decision rule, the water body brightness is generally selected from [60, 80], the bare land brightness is selected from [90, 110], the rectangle fit coefficient is greater than 0.9, and the normalized vegetation index NDVI selects a value less than 0 to distinguish it from terrain disasters. Specifically, Figure 3 shown.

[0077] S2. Using geospatial tools, measure the location range of the flood within the target range to obtain the real-time flooded area and water depth within the target range;

[0078] Specifically, Figure 4 As shown, the spatial information of remote sensing images is used, and with the help of the measurement function of geospatial tools, the length and width of the flood coverage area are measured. By combining terrain data, the flood coverage area, water depth and other data are estimated.

[0079] S3, obtaining real-time meteorological data within the target range, and inputting the real-time meteorological data and the real-time flooded area and water depth into a preset neural network model to predict the flooded area and water depth of the target area, wherein the preset neural network model is specifically constructed as follows:

[0080] Historical flood disaster data within the target range is obtained, and the historical flood disaster data is input into a first neural network model for training to obtain a preset neural network model, wherein the flood disaster data includes: meteorological data, flooded area and water depth.

[0081] Specifically, for the input variables of LSTM, the Pearson correlation coefficient is first used to calculate the correlation of temperature, rainfall, evaporation, runoff and other factors at different times. Sequences with high correlation and meeting the 95% confidence test are selected as pre-selected input items. On this basis, the random forest method is used to determine the final input items.

[0082] By calling the input item historical data and combining it with other information to form a time series data set, the input layer, loop layer and output layer are constructed. The input layer nodes are used to receive data. The Tanh function is used as the threshold activation function to bring the input layer data into the second layer of the neural network and define the parameters. This layer passes the neurons to another LSTM layer, and then uses the linear activation function to pass a fully connected normal layer for the prediction of the next time step.

[0083] In the model error backtracking parameter adjustment stage, the gradient descent method is used to update each weight and select appropriate value ranges for different parameters. The training sample set is used to complete the training of the LSTM neural network model, the test sample set is used to test the accuracy of the obtained model, and the mean absolute error (MAE) indicator is used to evaluate the model.

[0084] It should be noted that if Figure 5 As shown in the figure, real-time meteorological information is selected as the input data of the time series, and each monitoring terminal retrieves data. The rainfall, evaporation and terrain data of 0.5h, 1h, 2h, 3h and 6h are selected as the initial data. The LSTM artificial neural network is used as the prediction model to analyze and predict the changes in rainfall and evaporation over time, and predict the flood depth and inundation range in a few hours. According to the predicted flood development trend, relevant personnel are warned in time to carry out maintenance to reduce the impact and scope of flood disasters.

[0085] In one embodiment of the present invention, the acquiring of real-time image data within the target range and extracting the real-time image data to obtain the location range of the flood within the target range includes:

[0086] The real-time image data within the target range is divided into real-time image data of plain areas and real-time image data of urban areas;

[0087] Perform image extraction on the real-time image data of the plain area to obtain the location range of floods in the plain area within the target range;

[0088] Image extraction is performed on the real-time image data of the urban area to obtain the location range of the flood in the urban area within the target range.

[0089] In one embodiment of the present invention, the image extraction of the real-time image data of the plain area to obtain the location range of the flood in the plain area within the target range includes:

[0090] The flood edge is extracted from the real-time image data of the plain area within the target range using the big law method;

[0091] If the edge of the detected water body is discontinuous, a continuous flood target boundary is obtained through surface fitting, thereby obtaining the location range of floods in the plain area within the target range.

[0092] In one embodiment of the present invention, the step of performing image extraction on the real-time image data of the urban area to obtain the location range of the flood in the urban area within the target range includes:

[0093] The real-time image data of the urban area is segmented by using the threshold segmentation method to obtain the segmentation result of the real-time image data of the urban area;

[0094] Adopting object-oriented multi-level method, constructing object-oriented multi-level decision tree through threshold;

[0095] The result of the real-time image data segmentation of the urban area is input into the object-oriented multi-level decision tree to obtain the location range of the urban area flood within the target range.

[0096] Compared with the prior art, the flood monitoring method provided by the present invention has the following beneficial effects:

[0097] The present invention uses satellite remote sensing images and SAR image information for fusion, adopts different water body extraction methods for different terrain environments such as plains and towns, reduces the impact of environmental factors such as ground objects on water body target extraction, and improves the accuracy of water body identification. For the flood disaster range and water depth prediction part, real-time meteorological information is selected as the input data of the time series, combined with the geographical environment, and the LSTM artificial neural network is used as the prediction model to analyze and predict the changes in time, rainfall, evaporation, and the disaster degree of the flood-affected points. This method greatly reduces human intervention and, with the support of high-time satellite remote sensing and high-precision data, can play a greater role in flood monitoring, assessment and early warning.

[0098] like Figure 6 As shown, the present invention also provides a flood monitoring system 200, comprising: a data acquisition module 201, a data processing module 202, and a prediction module 203, wherein:

[0099] The data acquisition module 201 is used to acquire real-time image data within the target range, and extract the real-time image data to obtain the location range of the flood within the target range, wherein the real-time image data includes: satellite remote sensing real-time image data and SAR real-time image data;

[0100] The data processing module 202 is used to measure the location range of the flood within the target range using geospatial tools to obtain the real-time flooded area and water depth within the target range;

[0101] The prediction module 203 is used to obtain real-time meteorological data within the target range, and input the real-time meteorological data and the real-time flooded area and water depth into a preset neural network model to predict the flooded area and water depth of the target area. The specific construction process of the preset neural network model is as follows:

[0102] Historical flood disaster data within the target range is obtained, and the historical flood disaster data is input into a first neural network model for training to obtain a preset neural network model, wherein the flood disaster data includes: meteorological data, flooded area and water depth.

[0103] In one embodiment of the present invention, the data acquisition module is specifically used to:

[0104] The real-time image data within the target range is divided into real-time image data of plain areas and real-time image data of urban areas;

[0105] Perform image extraction on the real-time image data of the plain area to obtain the location range of floods in the plain area within the target range;

[0106] Image extraction is performed on the real-time image data of the urban area to obtain the location range of the flood in the urban area within the target range.

[0107] In one embodiment of the present invention, the image extraction of the real-time image data of the plain area to obtain the location range of the flood in the plain area within the target range includes:

[0108] The flood edge is extracted from the real-time image data of the plain area within the target range using the big law method;

[0109] If the edge of the detected water body is discontinuous, a continuous flood target boundary is obtained through surface fitting, thereby obtaining the location range of floods in the plain area within the target range.

[0110] In one embodiment of the present invention, the step of performing image extraction on the real-time image data of the urban area to obtain the location range of the flood in the urban area within the target range includes:

[0111] The real-time image data of the urban area is segmented by using the threshold segmentation method to obtain the segmentation result of the real-time image data of the urban area;

[0112] Adopting object-oriented multi-level method, constructing object-oriented multi-level decision tree through threshold;

[0113] The result of the real-time image data segmentation of the urban area is input into the object-oriented multi-level decision tree to obtain the location range of the urban area flood within the target range.

[0114] The present invention also provides a computer terminal device, comprising: one or more processors; a memory coupled to the processor and used to store one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the flood monitoring method as described in any one of the above items.

[0115] It should be noted that the processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or the processor can be any conventional processor. The processor is the control center of the terminal device, and various parts of the terminal device are connected using various interfaces and lines.

[0116] The memory mainly includes a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function, etc., and the data storage area can store related data, etc. In addition, the memory can be a high-speed random access memory, or a non-volatile memory, such as a plug-in hard disk, a smart memory card (SmartMediaCard, SMC), a secure digital (SecureDigital, SD) card, and a flash card (FlashCard), etc., or the memory can also be other volatile solid-state storage devices.

[0117] It should be noted that the above-mentioned terminal device may include, but is not limited to, a processor and a memory. Those skilled in the art will understand that the above-mentioned terminal device is merely an example and does not constitute a limitation on the terminal device. It may include more or fewer components, or a combination of certain components, or different components.

[0118] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any of the flood monitoring methods described above.

[0119] It should be noted that the computer program can be divided into one or more modules / units (such as computer program, computer program), and the one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0120] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A flood monitoring method, characterized in that: include: Acquire real-time image data within the target range, and extract the real-time image data to obtain the location range of the flood within the target range, wherein the real-time image data includes: satellite remote sensing real-time image data and SAR real-time image data; Using geospatial tools, the location range of the flood within the target range is measured to obtain the real-time flooded area and water depth within the target range; Real-time meteorological data within the target range is obtained, and the real-time meteorological data and the real-time flooded area and water depth are input into a preset neural network model to predict the flooded area and water depth of the target area, wherein the preset neural network model is specifically constructed as follows: Acquire historical flood disaster data within the target range, and input the historical flood disaster data into a first neural network model for training to obtain a preset neural network model, wherein the flood disaster data includes: meteorological data, flooded area and water depth; The step of extracting the real-time image data to obtain the location range of the flood within the target range includes: The flood edge is extracted from the real-time image data of the plain area within the target range using the big law method; if there is a discontinuity in the edge of the detected water body, the continuous flood target boundary is obtained through surface fitting, thereby obtaining the location range of the flood in the plain area within the target range; The real-time image data of urban areas within the target range are segmented by a threshold segmentation method to obtain the results of the real-time image data segmentation of the urban areas; an object-oriented multi-level method is adopted to construct an object-oriented multi-level decision tree through thresholds; the results of the real-time image data segmentation of the urban areas are input into the object-oriented multi-level decision tree to obtain the location range of floods in the urban areas within the target range.

2. The flood monitoring method according to claim 1, characterized in that: The step of acquiring real-time image data within the target range and extracting the real-time image data to obtain the location range of the flood within the target range includes: The real-time image data within the target range is divided into real-time image data of plain areas and real-time image data of urban areas; Perform image extraction on the real-time image data of the plain area to obtain the location range of floods in the plain area within the target range; Image extraction is performed on the real-time image data of the urban area to obtain the location range of the flood in the urban area within the target range.

3. A flood monitoring system, characterized in that: include: Data acquisition module, data processing module, prediction module, among which, The data acquisition module is used to acquire real-time image data within the target range, and extract the real-time image data to obtain the location range of the flood within the target range, wherein the real-time image data includes: satellite remote sensing real-time image data and SAR real-time image data; The data processing module is used to measure the location range of the flood within the target range using geospatial tools to obtain the real-time flooded area and water depth within the target range; The prediction module is used to obtain real-time meteorological data within the target range, and input the real-time meteorological data and the real-time flooded area and water depth into a preset neural network model to predict the flooded area and water depth of the target area, wherein the preset neural network model is specifically constructed as follows: Acquire historical flood disaster data within the target range, and input the historical flood disaster data into a first neural network model for training to obtain a preset neural network model, wherein the flood disaster data includes: meteorological data, flooded area and water depth; The step of extracting the real-time image data to obtain the location range of the flood within the target range includes: The flood edge is extracted from the real-time image data of the plain area within the target range using the big law method; if there is a discontinuity in the edge of the detected water body, the continuous flood target boundary is obtained through surface fitting, thereby obtaining the location range of the flood in the plain area within the target range; The real-time image data of urban areas within the target range are segmented by a threshold segmentation method to obtain the results of the real-time image data segmentation of the urban areas; an object-oriented multi-level method is adopted to construct an object-oriented multi-level decision tree through thresholds; the results of the real-time image data segmentation of the urban areas are input into the object-oriented multi-level decision tree to obtain the location range of floods in the urban areas within the target range.

4. The flood monitoring system according to claim 3, characterized in that: The data acquisition module is specifically used for: The real-time image data within the target range is divided into real-time image data of plain areas and real-time image data of urban areas; Perform image extraction on the real-time image data of the plain area to obtain the location range of floods in the plain area within the target range; Image extraction is performed on the real-time image data of the urban area to obtain the location range of the flood in the urban area within the target range.

5. A computer terminal device, characterized in that: include: one or more processors; A memory, coupled to the processor, for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the flood monitoring method according to any one of claims 1 to 2.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the flood monitoring method according to any one of claims 1 to 2 is implemented.

Images