Rainfall analysis method, device and equipment based on image fusion and medium

Abstract

The embodiment of the invention discloses a rainfall analysis method and device based on image fusion, equipment and a medium. The method comprises the following steps: accessing multiple types of image data based on a preset database interface so as to carry out superposition processing on the multiple types of image data to obtain to-be-analyzed image data; wherein the multiple types of images comprise a meteorological bureau radar map, a satellite cloud map and rainfall forecast result data; performing administrative region division on the to-be-analyzed image data to extract required region data corresponding to each administrative region in the to-be-analyzed image data; generating chart data corresponding to the required region data, and obtaining rainfall characteristic data of the chart data based on a project target corresponding to the administrative region; and inputting the rainfall characteristic data into a machine learning model corresponding to the project target to obtain a rainfall analysis result.

Description

Technical Field

[0001] This specification relates to the field of rainfall analysis technology, and in particular to a rainfall analysis method, device, equipment, and medium based on image fusion. Background Art

[0002] With the development of society, accurate analysis and prediction of rainfall conditions are becoming increasingly important. For example, in agricultural production, farmers need to understand rainfall conditions to plan irrigation and planting; urban managers rely on rainfall forecasts to maintain drainage systems and prepare for flood control. Therefore, numerous fields, including agricultural production, urban flood control and drainage, water resource management, and transportation, all require accurate and timely rainfall information to make relevant decisions. Therefore, rainfall analysis is crucial for current practical applications.

[0003] Currently, radar images, satellite cloud images, and weather forecast images are all static images, and the areas shown are relatively wide, making it difficult to perform regional segmentation. This makes it difficult for personnel to conduct data analysis on the areas covered by the images and to perform simple interpretation of specific rainfall areas. Furthermore, radar images, satellite cloud images, and weather forecast results images currently require manual interpretation. First, they are unable to analyze the weather conditions corresponding to fine-grained regions. Second, interpretation is time-consuming and labor-intensive. Third, data correlation analysis is not high, and it is impossible to combine the radar echo data corresponding to the analysis area, the cloud top temperature data corresponding to the area, and the rainfall forecast results data corresponding to the area. This is not conducive to the future application analysis of forecasts and actual rainfall results. Summary of the Invention

[0004] In order to solve the above technical problems, one or more embodiments of this specification provide a rainfall analysis method, apparatus, device and medium based on image fusion.

[0005] One or more embodiments of this specification adopt the following technical solutions: One or more embodiments of this specification provide a rainfall analysis method based on image fusion, the method comprising: Access multiple types of image data based on a preset database interface to perform superposition processing on the multiple types of image data to obtain image data to be analyzed; wherein the multiple types of images include: meteorological bureau radar images, satellite cloud images and rainfall forecast results data; Dividing the image data to be analyzed into administrative regions to extract required regional data corresponding to each administrative region in the image data to be analyzed; Generate chart data corresponding to the required area data, and obtain rainfall characteristic data of the chart data based on the project target corresponding to the administrative area; The rainfall characteristic data is input into a machine learning model corresponding to the project objective to obtain a rainfall analysis result.

[0006] Optionally, in one or more embodiments of the present specification, performing superposition processing on the multiple types of image data to obtain the image data to be analyzed specifically includes: Based on the timestamps and time resolutions corresponding to the various types of image data, the various types of image data are time-aligned to obtain multiple types of initial image data to be analyzed; Determining the number of registration points of the initial image data to be analyzed based on a preset accuracy, so as to obtain point coordinates of the number of registration points on each type of initial image data to be analyzed; Based on the point coordinates and the corresponding map coordinates, spatially registering the various types of initial image data to be analyzed to achieve spatial alignment of the various types of initial image data to be analyzed, and obtaining aligned initial image data to be analyzed; The aligned various initial image data to be analyzed are superimposed to obtain the image data to be analyzed.

[0007] Optionally, in one or more embodiments of this specification, dividing the image data to be analyzed into administrative regions specifically includes: Determining regional boundary data of the image data to be analyzed based on the scale corresponding to the project goal and the public data corresponding to the current administrative region; Converting the region boundary data based on the data format and coordinate system corresponding to the image data to be analyzed to obtain processed region boundary data; The processed region boundary data is used as a mask to segment the image data to be analyzed based on the mask to obtain sub-image data to be analyzed of multiple administrative regions.

[0008] Optionally, in one or more embodiments of the present specification, extracting required area data corresponding to each administrative area in the image data to be analyzed specifically includes: Performing statistics on the sub-image data to be analyzed corresponding to each grid in the administrative region to obtain regional data corresponding to each administrative region; wherein the regional data includes: spatial data and numerical data; Based on the data information associated with the current project target, data matching is performed on each of the sub-image data to be analyzed to extract the required regional data corresponding to each of the administrative regions.

[0009] Optionally, in one or more embodiments of the present specification, generating chart data corresponding to the required area data, and obtaining rainfall characteristic data of the chart data based on the project target corresponding to the administrative area, specifically includes: Based on the historical chart generation process corresponding to each required area data and the preset corresponding table, each required area data is processed to generate chart data corresponding to each required area data; Obtaining a data range of interest corresponding to the project target corresponding to the administrative area, and determining a target combination factor corresponding to the project target based on rainfall factors corresponding to each data within the data range of interest; Data corresponding to the target combination factor in the chart data is obtained as rainfall characteristic data.

[0010] Optionally, in one or more embodiments of this specification, before inputting the rainfall characteristic data into a machine learning model corresponding to the project objective to obtain a rainfall analysis result, the method further includes: Based on the data volume corresponding to each rainfall characteristic data and the accuracy data corresponding to the project goal, a corresponding initial machine learning model is matched; wherein the initial machine learning model is embedded with a preset threshold rule; Collecting rainfall characteristics and actual rainfall data corresponding to the target combination factor as data samples, so as to train the initial machine learning model based on the data samples; Based on the comparison between the trained initial machine learning model and the actual rainfall data, the initial machine learning model is iteratively adjusted to obtain a machine learning model that meets the requirements.

[0011] Optionally, in one or more embodiments of this specification, after inputting the rainfall characteristic data into a machine learning model corresponding to the project objective to obtain a rainfall analysis result, the method further includes: Determining whether the rainfall analysis result exceeds a preset result threshold; If so, determining a threshold exceeding value based on the rainfall analysis result and the preset result threshold; Generate reminder information based on the threshold exceeding value, and send the reminder information to the corresponding associated device.

[0012] One or more embodiments of this specification provide a rainfall analysis device based on image fusion, the device comprising: An overlay unit is configured to access multiple types of image data based on a preset database interface to perform overlay processing on the multiple types of image data to obtain image data to be analyzed; wherein the multiple types of images include: radar images from the meteorological bureau, satellite cloud images, and rainfall forecast results data; an extraction unit, configured to divide the image data to be analyzed into administrative regions, so as to extract required region data corresponding to each administrative region in the image data to be analyzed; an acquisition unit, configured to generate chart data corresponding to the required area data, and acquire rainfall characteristic data of the chart data based on a project target corresponding to the administrative area; An analysis unit is used to input the rainfall characteristic data into a machine learning model corresponding to the project goal to obtain a rainfall analysis result.

[0013] One or more embodiments of this specification provide a rainfall analysis device based on image fusion, the device comprising: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: perform any of the above methods.

[0014] One or more embodiments of this specification provide a non-volatile computer storage medium storing computer-executable instructions, wherein the computer-executable instructions are configured to execute any of the above-described methods.

[0015] At least one of the above technical solutions adopted in the embodiments of this specification can achieve the following beneficial effects: By overlaying multiple types of image data, including meteorological bureau radar images, satellite cloud images, and rainfall forecast data, a more comprehensive and complete rainfall analysis dataset can be formed, providing a more robust information foundation for in-depth analysis of rainfall conditions. Different types of image data have their own advantages and characteristics, and overlaying them enables data complementarity. By dividing the image data for analysis by administrative regions, the required regional data corresponding to each administrative region can be accurately extracted. Converting the required regional data into graphical data allows for a more intuitive understanding of rainfall characteristics and distribution. By determining target combination factors based on project objectives and extracting corresponding rainfall characteristic data, the analysis results are more focused on key features. This not only improves analysis efficiency but also makes the results more interpretable. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are only some of the embodiments described in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without inventive work. In the drawings: Figure 1 A schematic diagram of a method flow of a rainfall analysis method based on image fusion provided in an embodiment of this specification; Figure 2 A schematic diagram of a technical route for rainfall analysis based on image fusion provided in an embodiment of this specification; Figure 3 A schematic structural diagram of a rainfall analysis device based on image fusion provided in an embodiment of this specification; Figure 4 A schematic diagram of the structure of a rainfall analysis device based on image fusion provided in an embodiment of this specification; Figure 5 A schematic diagram of the structure of a non-volatile storage medium provided in an embodiment of this specification. DETAILED DESCRIPTION

[0017] The embodiments of this specification provide a rainfall analysis method, apparatus, device, and medium based on image fusion.

[0018] To help those skilled in the art better understand the technical solutions in this specification, the following will provide a clear and complete description of the technical solutions in the embodiments of this specification, in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this specification without creative work should fall within the scope of protection of this specification.

[0019] like Figure 1 As shown in FIG, the embodiment of this specification provides a flowchart of a rainfall analysis method based on image fusion. Figure 1 It can be seen that in one or more embodiments of this specification, a rainfall analysis method based on image fusion includes: S101: Access multiple types of image data based on a preset database interface to perform overlay processing on the multiple types of image data to obtain image data to be analyzed; wherein the multiple types of images include: meteorological bureau radar images, satellite cloud images and rainfall forecast results data.

[0020] In the current traditional rainfall analysis method, the data correlation analysis is not high, and the correlation analysis between actual rainfall and three types of forecast results is not carried out. It is difficult to analyze the relationship between actual rainfall, radar map, satellite cloud map and weather forecast result map. Therefore, in order to solve this problem, Figure 2 In the embodiment of this specification, multiple types of image data are accessed according to a preset database interface to perform superposition processing on the multiple types of image data to obtain image data to be analyzed.

[0021] Specifically, in one or more embodiments of this specification, performing superposition processing on multiple types of image data to obtain image data to be analyzed specifically includes: Based on the timestamps and temporal resolutions corresponding to each type of image data, the various types of image data are temporally aligned to obtain multiple types of initial image data to be analyzed. The number of registration points for the initial image data to be analyzed is then determined based on a preset accuracy to obtain point coordinates corresponding to the number of registration points on each type of initial image data to be analyzed. It is understood that a greater number of registration points indicates higher accuracy. Spatial registration is then performed on the various types of initial image data to be analyzed based on the point coordinates and the corresponding map coordinates to achieve spatial alignment of the various types of initial image data to be analyzed, obtaining the aligned initial image data to be analyzed. The aligned initial image data to be analyzed are then overlaid to obtain the image data to be analyzed.

[0022] In one application scenario, the overlay process can involve first selecting multiple points on various image types, determining their X and Y coordinates, and then matching these points to the map coordinates. The more points selected, the higher the matching accuracy. Georeferencing is then performed. One configuration method involves directly entering X and Y coordinates or longitude and latitude information on the image to be registered. The second method involves linking control points. Within a single project, clicking a point on the reference image and then clicking the corresponding point on the image to be registered achieves the registration goal. The data is then overlaid with the image to form a map that can be zoomed in and out.

[0023] This process overlays various image data types, including radar images from the Meteorological Bureau, satellite cloud images, and rainfall forecast data. This integrates the rich information provided by these diverse data sources, addressing the fragmentation of rainfall data in traditional planning. This provides a more comprehensive and complete rainfall analysis dataset, providing a more robust information foundation for in-depth analysis of rainfall conditions. Furthermore, different types of image data have their own strengths and characteristics, and the overlay process enables complementary data.

[0024] S102: Divide the image data to be analyzed into administrative regions to extract required region data corresponding to each administrative region in the image data to be analyzed.

[0025] After obtaining the image data to be analyzed, in order to refine the region, the image data to be analyzed will be divided into administrative regions, thereby extracting the required regional data corresponding to each administrative region in the image data to be analyzed. Specifically, in one or more embodiments of this specification, the administrative region division of the image data to be analyzed specifically includes: Based on the scale corresponding to the project goal and the public data corresponding to the current administrative region, the regional boundary data of the image data to be analyzed is determined. Then, according to the data format and coordinate system corresponding to the image data to be analyzed, the regional boundary data is converted to obtain processed regional boundary data. The processed regional boundary data is used as a mask to cut the image data to be analyzed based on the mask to obtain sub-image data to be analyzed for multiple administrative regions. This process accurately focuses the analysis scope on each administrative region through administrative region division. In this way, rainfall analysis can be carried out according to the specific characteristics and needs of each administrative region, providing more targeted rainfall information and decision support for different administrative regions. In addition, through regional division, the analysis scope is narrowed, and the complexity and interference factors of the data are reduced. Within a smaller range, the distribution patterns and characteristics of rainfall can be analyzed in more detail, improving the accuracy of subsequent analysis.

[0026] Specifically, in one or more embodiments of the present specification, extracting required regional data corresponding to each administrative region in the image data to be analyzed specifically includes: Statistics are collected on the sub-image data to be analyzed corresponding to each grid in the administrative region to obtain the regional data corresponding to each administrative region; the regional data includes spatial data and numerical data. Then, based on the data information associated with the current project goal, data matching is performed on each sub-image data to be analyzed to extract the required regional data corresponding to each administrative region. It is understandable that different project goals will result in different key data. For example, if it is an urban waterlogging prevention and control project, the data information associated with the project goal may include areas prone to waterlogging, the distribution of urban drainage pipe networks, etc. By matching this associated data information with the sub-image data to be analyzed, the required regional data closely related to the current project goal can be accurately extracted from the regional data of each administrative region, which will serve as the key basis for subsequent rainfall analysis and provide support for subsequent in-depth rainfall analysis and decision-making.

[0027] This process, through data matching, accurately locates data within each administrative region that is closely related to the project objectives, avoiding interference from irrelevant data. This allows the extracted data from the required regions to more precisely focus on project needs, improving the pertinence and accuracy of rainfall analysis. Statistics on raster data within administrative regions meticulously reflect the spatial distribution and numerical characteristics of rainfall within each region, providing a richer and more accurate data foundation for subsequent analysis.

[0028] S103: Generate chart data corresponding to the required area data, and obtain rainfall characteristic data of the chart data based on the project target corresponding to the administrative area.

[0029] After obtaining the required regional data, corresponding chart data is generated, thereby obtaining rainfall characteristic data based on the project objectives corresponding to the administrative region. For example, in a certain scenario, based on the demarcated grid and regional computing technology, the required regional data such as echo intensity, temperature color difference, and rainfall intensity can be directly generated into chart data based on radar images, weather cloud maps, and weather forecast results maps, realizing the combined display of GIS maps and tabular data.

[0030] Specifically, in one or more embodiments of the present specification, generating chart data corresponding to required regional data, and obtaining rainfall characteristic data of the chart data based on the project target corresponding to the administrative region, specifically includes: Based on the historical chart generation process and preset corresponding tables corresponding to the data for each required area, the data for each required area is processed to generate chart data corresponding to the data for each required area. Chart data typically includes spatial distribution maps, time series graphs, bar charts, line charts, etc., which can intuitively display rainfall characteristic data. The data range corresponding to the project objective corresponding to the administrative area is then obtained, and based on the rainfall factors corresponding to each data within the data range, the target combination factor corresponding to the project objective is determined. In other words, different project objectives have different focus points on rainfall data. By determining the target combination factor, it is possible to focus on the rainfall characteristics most relevant to the project objective. The data corresponding to the target combination factor in the chart data is then obtained as the rainfall characteristic data.

[0031] This process generates charts and data, presenting complex rainfall data in an intuitive manner. This allows for a more intuitive understanding of rainfall characteristics and distribution, improving the scientific and rational nature of decision-making. Furthermore, by determining target combination factors based on different project objectives, rainfall analysis becomes more customized, providing targeted analysis results for each project to meet its specific needs.

[0032] S104: Inputting the rainfall characteristic data into a machine learning model corresponding to the project goal to obtain rainfall analysis results.

[0033] After obtaining rainfall characteristic data based on the above steps, it is input into a machine learning model corresponding to the project objectives to obtain rainfall analysis results. In certain application scenarios, massive data from historical experience databases can be extracted. Based on the selected rainfall characteristics, the characteristic data values ​​and the actual rainfall data set are combined. Using machine learning algorithms, statistical modeling is performed. The model trained with historical experience data calculates rainfall forecasts for a specific area in real time. The model's output is then compared with actual rainfall to determine if it is correct. This allows for further analysis of the rationality of the selected rainfall characteristics, and the model's rainfall characteristics can be adjusted accordingly, continuously optimizing and improving the model.

[0034] Furthermore, in one or more embodiments of the present specification, before inputting the rainfall characteristic data into a machine learning model corresponding to the project objective to obtain a rainfall analysis result, the method further includes: Based on the data volume corresponding to each rainfall feature data and the accuracy data corresponding to the project goal, a corresponding initial machine learning model is matched. The initial machine learning model is embedded with preset threshold rules. These rules can be set based on the radar echo data and rainfall condition correspondence table shown in Table 1, the meteorological cloud map data and rainfall condition correspondence table shown in Table 2, and the corresponding rainfall forecast map and rainfall amount correspondence table shown in Table 3.

[0035] Table 1. Correspondence between radar echo data and rainfall conditions

[0036] Table 2. Correspondence between meteorological cloud map data and rainfall conditions

[0037] Table 3. Correspondence between rainfall forecast map and rainfall amount

[0038] Next, rainfall characteristics corresponding to the target combination factors and actual rainfall data are collected as data samples. The initial machine learning model is then trained based on the data samples, enabling the model to learn the mapping relationship between rainfall characteristics and actual rainfall. Based on a comparison between the trained initial machine learning model and the actual rainfall data, the initial machine learning model is iteratively adjusted to obtain a machine learning model that meets the requirements.

[0039] Furthermore, in one or more embodiments of the present specification, after inputting the rainfall characteristic data into a machine learning model corresponding to the project objective to obtain a rainfall analysis result, the method further includes the following process: First, determine whether the rainfall analysis results exceed the preset result threshold. If so, the threshold exceedance value is determined based on the rainfall analysis results and the preset result threshold. Based on the threshold exceedance value, an alert message is generated and sent to the corresponding associated device. In other words, by setting an early warning threshold, areas exceeding the threshold and the corresponding data are automatically displayed on the duty system. Furthermore, to reduce the server load caused by each regional calculation service, the regional calculation results, namely the rainfall analysis results, are cached. This facilitates subsequent querying and data comparison of historical rainfall analysis results, providing fast service.

[0040] like Figure 3 As shown in FIG, the embodiment of this specification provides a structural diagram of a rainfall analysis device based on image fusion. Figure 3It can be seen that in one or more embodiments of this specification, a rainfall analysis device based on image fusion includes: An overlay unit is configured to access multiple types of image data based on a preset database interface to perform overlay processing on the multiple types of image data to obtain image data to be analyzed; wherein the multiple types of images include: radar images from the meteorological bureau, satellite cloud images, and rainfall forecast results data; an extraction unit, configured to divide the image data to be analyzed into administrative regions, so as to extract required region data corresponding to each administrative region in the image data to be analyzed; an acquisition unit, configured to generate chart data corresponding to the required area data, and acquire rainfall characteristic data of the chart data based on a project target corresponding to the administrative area; An analysis unit is used to input the rainfall characteristic data into a machine learning model corresponding to the project goal to obtain a rainfall analysis result.

[0041] like Figure 4 As shown in FIG, the embodiment of this specification provides a structural diagram of a rainfall analysis device based on image fusion. Figure 4 It can be seen that in one or more embodiments of this specification, a rainfall analysis device based on image fusion includes: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: perform any of the above methods.

[0042] like Figure 5 As shown in FIG, the embodiment of this specification provides a structural diagram of a non-volatile storage medium. Figure 5 It can be seen that in one or more embodiments of this specification, a non-volatile storage medium stores computer-executable instructions, and the computer-executable instructions can: execute any of the methods described above.

[0043] The various embodiments in this specification are described in a progressive manner. Similar portions between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from the other embodiments. In particular, the device, apparatus, and non-volatile computer storage medium embodiments are generally similar to the method embodiments, so their descriptions are relatively simplified. For relevant details, refer to the descriptions of the method embodiments.

[0044] The foregoing description of this specification describes specific embodiments. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0045] The foregoing description is merely one or more embodiments of this specification and is not intended to limit this specification. It will be apparent to those skilled in the art that various modifications and variations may be made to one or more embodiments of this specification. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of one or more embodiments of this specification are intended to be within the scope of the claims of this specification.

Claims

1. A rainfall analysis method based on image fusion, characterized in that: The method comprises: Access multiple types of image data based on a preset database interface to perform superposition processing on the multiple types of image data to obtain image data to be analyzed; wherein the multiple types of images include: meteorological bureau radar images, satellite cloud images and rainfall forecast results data; Dividing the image data to be analyzed into administrative regions to extract required regional data corresponding to each administrative region in the image data to be analyzed; Generate chart data corresponding to the required area data, and obtain rainfall characteristic data of the chart data based on the project target corresponding to the administrative area; The rainfall characteristic data is input into a machine learning model corresponding to the project objective to obtain a rainfall analysis result.

2. The rainfall analysis method based on image fusion according to claim 1, characterized in that: The multiple types of image data are superimposed to obtain image data to be analyzed, specifically including: Based on the timestamps and time resolutions corresponding to the various types of image data, the various types of image data are time-aligned to obtain multiple types of initial image data to be analyzed; Determining the number of registration points of the initial image data to be analyzed based on a preset accuracy, so as to obtain point coordinates of the number of registration points on each type of initial image data to be analyzed; Based on the point coordinates and the corresponding map coordinates, spatially registering the various types of initial image data to be analyzed to achieve spatial alignment of the various types of initial image data to be analyzed, and obtaining aligned initial image data to be analyzed; The aligned various initial image data to be analyzed are superimposed to obtain the image data to be analyzed.

3. The rainfall analysis method based on image fusion according to claim 1, characterized in that: Dividing the image data to be analyzed into administrative regions specifically includes: Determining regional boundary data of the image data to be analyzed based on the scale corresponding to the project goal and the public data corresponding to the current administrative region; Converting the region boundary data based on the data format and coordinate system corresponding to the image data to be analyzed to obtain processed region boundary data; The processed region boundary data is used as a mask to segment the image data to be analyzed based on the mask to obtain sub-image data to be analyzed of multiple administrative regions.

4. The rainfall analysis method based on image fusion according to claim 3, characterized in that: Extracting the required regional data corresponding to each administrative region in the image data to be analyzed, specifically including: Performing statistics on the sub-image data to be analyzed corresponding to each grid in the administrative region to obtain regional data corresponding to each administrative region; wherein the regional data includes: spatial data and numerical data; Based on the data information associated with the current project target, data matching is performed on each of the sub-image data to be analyzed to extract the required regional data corresponding to each of the administrative regions.

5. The rainfall analysis method based on image fusion according to claim 1, characterized in that: Generating chart data corresponding to the required area data, and obtaining rainfall characteristic data of the chart data based on the project target corresponding to the administrative area, specifically comprising: Based on the historical chart generation process corresponding to each required area data and the preset corresponding table, each required area data is processed to generate chart data corresponding to each required area data; Obtaining a data range of interest corresponding to the project target corresponding to the administrative area, and determining a target combination factor corresponding to the project target based on rainfall factors corresponding to each data within the data range of interest; Data corresponding to the target combination factor in the chart data is obtained as rainfall characteristic data.

6. The rainfall analysis method based on image fusion according to claim 1, characterized in that: Before inputting the rainfall characteristic data into a machine learning model corresponding to the project objective to obtain a rainfall analysis result, the method further includes: Based on the data volume corresponding to each rainfall characteristic data and the accuracy data corresponding to the project goal, a corresponding initial machine learning model is matched; wherein the initial machine learning model is embedded with a preset threshold rule; Collecting rainfall characteristics and actual rainfall data corresponding to the target combination factor as data samples, so as to train the initial machine learning model based on the data samples; Based on the comparison between the trained initial machine learning model and the actual rainfall data, the initial machine learning model is iteratively adjusted to obtain a machine learning model that meets the requirements.

7. The rainfall analysis method based on image fusion according to claim 1, characterized in that: After inputting the rainfall characteristic data into a machine learning model corresponding to the project objective to obtain a rainfall analysis result, the method further includes: Determining whether the rainfall analysis result exceeds a preset result threshold; If so, determining a threshold exceeding value based on the rainfall analysis result and the preset result threshold; Generate reminder information based on the threshold exceeding value, and send the reminder information to the corresponding associated device.

8. A rainfall analysis device based on image fusion, characterized in that: The device comprises: An overlay unit is configured to access multiple types of image data based on a preset database interface to perform overlay processing on the multiple types of image data to obtain image data to be analyzed; wherein the multiple types of images include: radar images from the meteorological bureau, satellite cloud images, and rainfall forecast results data; an extraction unit, configured to divide the image data to be analyzed into administrative regions, so as to extract required region data corresponding to each administrative region in the image data to be analyzed; an acquisition unit, configured to generate chart data corresponding to the required area data, and acquire rainfall characteristic data of the chart data based on a project target corresponding to the administrative area; An analysis unit is used to input the rainfall characteristic data into a machine learning model corresponding to the project goal to obtain a rainfall analysis result.

9. A rainfall analysis device based on image fusion, characterized in that: The device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: perform any of the above methods.

10. A non-volatile storage medium storing computer-executable instructions, characterized in that: The computer executable instructions can: execute any of the above methods.

Images