A method and system for predicting compound disaster risk

Abstract

The application provides a composite disaster risk prediction method and system, and relates to the technical field of unmanned aerial vehicles.The method comprises the following steps: collecting regional basic information of a target supervision area and performing grid division, generating data collection path data, performing image collection of the target supervision area based on an intelligent device, performing feature screening of the regional basic information in a feature matching library, performing feature identification of the image collection result, generating a disaster response division interval based on a geomorphologic feature identification result, performing geological identification of the disaster response division interval after obtaining a geological feature matching result and the regional basic information, generating a multi-level geological disaster sensitivity index, collecting rainfall prediction data, generating disaster risk early warning information, and solving the technical problem of inaccuracy of disaster early warning caused by errors in collection of information and data arrangement of a supervision area, so that accurate data collection and arrangement of the supervision area are realized, the supervision area is accurately divided, and the accuracy of early warning is improved.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) technology, and specifically to a method and system for predicting complex disaster risks. Background Technology

[0002] In recent years, major geological disasters such as landslides, mudslides, and rockfalls have continued to be frequently reported in the news. It should be noted that the causes of geological disasters are not limited to natural factors but also include human factors. Moreover, the occurrence of geological disasters follows predictable patterns. Close monitoring of geological disaster sites, scientific prediction of the probability of occurrence, and rapid early warning after a geological disaster can minimize losses for the public.

[0003] Against this backdrop, by fully utilizing modern technology and integrating theoretical knowledge from multiple disciplines, a geological disaster monitoring and early warning system was designed and developed in collaboration with multiple departments to provide 24 / 7 monitoring of key geological disaster areas. Once a potential geological disaster is detected, an early warning will be issued immediately.

[0004] Currently, errors exist in the collection and processing of information and data in the monitored areas, resulting in inaccurate disaster warnings. Summary of the Invention

[0005] This application provides a composite disaster risk prediction method and system to address the technical problem in existing technologies where inaccurate disaster early warnings are caused by errors in information collection and data processing in the monitored area.

[0006] In view of the above problems, this application provides a method and system for predicting complex disaster risks.

[0007] In a first aspect, this application provides a method for predicting composite disaster risks. The method includes: acquiring basic regional information of a target monitoring area; dividing the area into grids based on the basic regional information, and generating data acquisition path data based on the grid division results; acquiring images of the target monitoring area using the data acquisition path data via a smart device, and obtaining image acquisition results; performing feature filtering based on the basic regional information using a feature matching library, performing feature recognition of the image acquisition results based on the feature filtering results, and generating disaster response division intervals based on the geomorphological feature recognition results; obtaining geological feature matching results, performing geological identification of the disaster response division intervals based on the basic regional information and the geological feature matching results, and generating a multi-level geological disaster sensitivity index based on the identification results; acquiring rainfall forecast data, and generating disaster risk early warning information based on the rainfall forecast data and the multi-level geological disaster sensitivity index.

[0008] Secondly, this application provides a composite disaster risk prediction system, comprising: a regional basic information module for collecting regional basic information of a target monitoring area; a data acquisition path module for performing grid division based on the regional basic information and generating data acquisition path data based on the grid division results; an image acquisition module for acquiring images of the target monitoring area using the data acquisition path data via a smart device and obtaining image acquisition results; a disaster response division interval module for performing feature filtering based on a feature matching library based on the regional basic information, performing feature recognition of the image acquisition results based on the feature filtering results, and generating disaster response division intervals based on the geomorphological feature recognition results; a multi-level geological hazard sensitivity index module for obtaining geological feature matching results, performing geological identification of the disaster response division intervals based on the regional basic information and the geological feature matching results, and generating a multi-level geological hazard sensitivity index based on the identification results; and a disaster risk early warning information module for collecting rainfall forecast data and generating disaster risk early warning information based on the rainfall forecast data and the multi-level geological hazard sensitivity index.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] This application provides a composite disaster risk prediction method, which relates to the field of unmanned aerial vehicle (UAV) technology. It solves the technical problem in the prior art where errors in the collection and processing of information and data in the monitored area lead to inaccurate disaster warnings. The method achieves accurate data collection and processing of the monitored area, thereby accurately dividing the monitored area and improving the accuracy of the warning. Attached Figure Description

[0011] Figure 1 This application provides a schematic diagram of a composite disaster risk prediction method.

[0012] Figure 2 This application provides a schematic diagram of the process for generating multi-layered and geological hazard sensitivity indices in a composite disaster risk prediction method;

[0013] Figure 3 This application provides a schematic diagram of the process for generating disaster response division intervals in a composite disaster risk prediction method;

[0014] Figure 4 This application provides a schematic diagram of a composite disaster risk prediction system.

[0015] Attached figure labels: Regional basic information module 1, data acquisition path module 2, image acquisition module 3, disaster response division interval module 4, multi-level geological disaster sensitivity index module 5, disaster risk early warning information module 6. Detailed Implementation

[0016] This application provides a composite disaster risk prediction method to address the technical problem in existing technologies where inaccurate disaster warnings are caused by errors in information collection and data processing in the monitored area.

[0017] Example 1

[0018] like Figure 1 As shown in the figure, this application provides a method for predicting composite disaster risks. This method is applied to an unmanned aerial vehicle (UAV) surveying system and includes:

[0019] Step S100: Collect and obtain basic regional information of the target supervision area;

[0020] Specifically, the basic information of the target monitoring area is obtained through real-time dynamic monitoring. The basic information of the target monitoring area includes, but is not limited to, rainfall information, geological deformation information, and geomorphological feature information of geological disaster points, which serves as an important reference for obtaining disaster risk early warning information in the later stage.

[0021] Step S200: Perform grid division based on the basic information of the region, and generate data acquisition path data based on the grid division results;

[0022] Specifically, based on the acquired regional basic information, the collected area is divided into grids. Then, the image information captured by each grid block is traversed, identified, and screened. Based on the acquired regional basic information, the regional basic information is divided equally. At the same time, the first area in the regional basic information grid division is set as the starting point, that is, the first area obtained, and marked as the zero point area. Then, traversal is carried out starting from the first area, and the information obtained in each area is matched with the actual path in the map, thereby generating data collection path data, which lays a solid foundation for improving the accuracy of disaster risk early warning.

[0023] Step S300: Based on the data acquisition path data, the intelligent device performs image acquisition of the target supervision area to obtain image acquisition results;

[0024] Specifically, based on the image acquisition of the target monitoring area using intelligent devices, the data is then matched with the data acquisition path data. These intelligent devices can be drones, lidar, unmanned airships, etc., to further obtain the image acquisition results of the target monitoring area based on the data acquisition path data, thereby ensuring the acquisition of disaster risk early warning information.

[0025] Step S400: Based on the basic information of the region, perform feature filtering in the feature matching library, perform feature recognition of the image acquisition results based on the feature filtering results, and generate disaster response division intervals based on the geomorphological feature recognition results;

[0026] Specifically, based on the obtained regional basic information, features are selected from a feature matching library. This library covers different terrains (slope aspect features), different landforms, and different soil data (surface composition features). The selected features are then used to identify features from the acquired images. Furthermore, disaster response division intervals are generated based on the obtained landform feature identification results. These intervals include, but are not limited to, landforms that respond quickly to rainstorms, landforms that respond slowly to rainstorms, landforms that contribute little to rainstorms, and landforms that respond quickly to rainstorms but have poor connectivity. These intervals have a profound impact on the subsequent generation of disaster risk warning information.

[0027] Step S500: Obtain geological feature matching results, perform geological identification of the disaster response division intervals based on the regional basic information and the geological feature matching results, and generate multi-level geological disaster sensitivity indices based on the identification results;

[0028] Specifically, the geological features of the target monitoring area are extracted by regional remote sensing geological features and the geological feature matching results are obtained. The geological features include structural features, strata, oil and gas features, reservoir features, resource quantity, etc. Then, based on the obtained regional basic information and the obtained geological feature matching results, the geological labels of the disaster response division intervals are made. Based on the geological labeling results, multi-layer and geological disaster sensitivity indices are generated, and on this basis, disaster risk early warning information is obtained more accurately.

[0029] Step S600: Collect rainfall forecast data and generate disaster risk early warning information based on the rainfall forecast data and the multi-level geological disaster sensitivity index.

[0030] Specifically, based on the information collected through real-time dynamic monitoring, rainfall prediction data is generated. The basic information of the monitored target area includes, but is not limited to, rainfall information, geological deformation information, and geomorphological feature information of geological disaster points. The obtained rainfall prediction data and the aforementioned multi-layered and geological disaster sensitivity indices are further matched and integrated, thereby generating more accurate disaster risk early warning information based on the integrated data.

[0031] Furthermore, this invention provides a composite disaster risk prediction method and system, relating to the field of unmanned aerial vehicle (UAV) technology. The method includes: collecting basic regional information of a target monitoring area; dividing the area into grids based on the basic regional information; generating data acquisition path data based on the grid division results; acquiring images of the target monitoring area using intelligent devices based on the data acquisition path data; obtaining image acquisition results; performing feature filtering based on a feature matching library based on the basic regional information; performing feature recognition based on the feature filtering results of the image acquisition results; generating disaster response division intervals based on the geomorphological feature recognition results; obtaining geological feature matching results; identifying the geological characteristics of the disaster response division intervals based on the basic regional information and the geological feature matching results; generating a multi-level geological disaster sensitivity index based on the identification results; collecting rainfall forecast data; and generating disaster risk early warning information based on the rainfall forecast data and the multi-level geological disaster sensitivity index. This invention solves the technical problem in the prior art where the inability to scientifically predict geological disasters leads to significant losses after the occurrence of geological disasters. It enables all-weather monitoring of key geological disaster areas, reducing social and economic losses caused by floods and droughts.

[0032] Furthermore, such as Figure 2 As shown, step S500 of this application further includes:

[0033] Step S510: Collect activity information based on the basic information of the region to obtain the activity information collection results;

[0034] Step S520: Based on the activity information collection results, obtain activity geological impact level data, wherein the activity geological impact level data has an activity area identifier;

[0035] Step S530: Based on the activity information collection results, obtain activity geological impact level data, wherein the activity geological impact level data has an activity area identifier;

[0036] Step S540: Generate the multi-level geological hazard sensitivity index based on the adjustment results of the geological information.

[0037] Specifically, based on the obtained regional basic information, activity information is collected. This activity information refers to human engineering activities, which are influenced by factors such as the intensity of mining activities, road network density, and projected population density. The collected activity information is then analyzed to obtain activity geological impact level data. This data identifies different activity areas of human engineering activities. Further adjustments are made to the identified geological information based on this data. Since geological hazards are affected not only by human engineering activities but also by topsoil thickness, lithology, topographic slope, and landform type, the identified geological information is adjusted according to different topsoil thicknesses, lithologies, topographic slopes, and landform types after identifying human engineering activity areas. A multi-level geological hazard sensitivity index is generated based on the adjusted geological information. Each multi-level index corresponds to a specific element, and its calculation formula is as follows: In the formula: i is the sensitivity index of a specific element pp a Let be the conditional probability of a disaster point in the i-th type of element, and be the ratio of the number of disaster points in the i-th type of element to the area of ​​the i-th type of element; pp s The prior probability of a disaster point can be expressed as the ratio of the number of disasters to the target monitored area; M is the total number of disaster points in the target monitored area; A is the total area of ​​the target monitored area; m is the number of disasters within the i-th type of element; and a is the area of ​​the i-th type of element. The multi-level geological disaster sensitivity index obtained through this calculation indicates the degree of close relationship between the element and geological disasters. The larger the value, the closer the potential relationship between the element and geological disasters, and the more likely geological disasters are to occur. This achieves the technical effect of providing an important basis for obtaining disaster risk early warning information in the later stage.

[0038] Furthermore, such as Figure 3 As shown, step S400 of this application further includes:

[0039] Step S410: Based on the feature filtering results, perform slope aspect feature recognition on the image acquisition results to obtain slope aspect feature recognition results;

[0040] Step S420: Based on the feature filtering results, perform surface composition feature recognition of the image acquisition results to obtain surface composition feature recognition results;

[0041] Step S430: Generate the disaster response division interval based on the slope aspect feature identification result and the surface composition feature identification result.

[0042] Specifically, based on the obtained feature screening results, slope aspect feature recognition and surface composition feature recognition are performed respectively with the image acquisition results to obtain the corresponding slope aspect feature recognition results and surface composition feature recognition results. In the slope aspect feature recognition, the collected slope aspect data and slope data need to be extracted and integrated to obtain the slope aspect feature recognition result. In the surface composition feature recognition, the surface features that are distributed in a stepped pattern and the landmark features that are mainly mountains and plateaus need to be extracted and integrated to obtain the surface composition feature recognition result. Furthermore, based on the obtained slope aspect feature recognition results and surface composition feature recognition results, disaster response division intervals are generated, which achieves the technical effect of improving the accuracy of disaster risk early warning information.

[0043] Furthermore, step S430 of this application includes:

[0044] Step S431: Construct initial weight values ​​for slope aspect features and land surface composition features;

[0045] Step S432: Obtain slope aspect data and slope data based on the slope aspect feature identification results;

[0046] Step S433: Perform disaster response rating based on the slope aspect data, the slope data, the surface composition feature identification results, and the initial weight value, and obtain the disaster response classification results based on the disaster response rating results.

[0047] Specifically, based on the obtained slope aspect features and surface composition features, an initial weight ratio is allocated to the slope aspect features and surface composition features. The weight ratio of slope aspect features to surface composition features can be 6:4 (first influence coefficient: second influence coefficient). Further, slope aspect data and slope data are extracted based on the obtained slope aspect feature identification results. Slope aspect data represents the direction of the projection of the slope normal onto the horizontal plane, i.e., the direction from high to low. Slope data represents the inclination of the ground in the target monitoring area relative to the horizontal plane, usually expressed as a percentage, using the formula: i = h / L. A larger slope (larger i) results in a shorter horizontal distance (smaller ΔL) and denser contour lines. Then, based on the obtained slope aspect data, slope data, surface composition feature identification results, and initial weight values, the disaster impact is rated. The disaster response rating is based on the initial weight values, classifying levels according to slope aspect features and surface composition features, and the area information of the disaster response rating area. Thus, a disaster response classification result is generated based on the obtained disaster response rating results, enabling reasonable judgment of subsequent disaster risk warning information.

[0048] Furthermore, step S433 of this application also includes:

[0049] Step S4331: Obtain disaster response rating level information and disaster response rating area information based on the disaster response rating results;

[0050] Step S4332: Set the adjacent region aggregation threshold;

[0051] Step S4333: Obtain the disaster response rating difference between the first region and the second region based on the disaster response rating level information, wherein the first region and the second region are adjacent regions;

[0052] Step S4334: Obtain the area information of the first region and the area information of the second region through the disaster response rating area information;

[0053] Step S4335: Obtain the area rating difference based on the area information of the first region and the area information of the second region;

[0054] Step S4336: Calculate the aggregate value of adjacent areas based on the disaster response rating difference and the area rating difference;

[0055] Step S4337: Determine whether the adjacent region aggregation value meets the adjacent region aggregation threshold. If the adjacent region aggregation threshold is met, merge the first region and the second region, and use the disaster response rating level corresponding to the large area in the first region and the second region as the disaster response rating level of the merged region.

[0056] Specifically, based on the disaster response rating results obtained above, disaster response rating level information and disaster response rating area information are extracted respectively. The difference in disaster response rating between the first and second regions is obtained based on the disaster response rating level information, where the first and second regions are adjacent regions. The disaster response ratings for the first and second regions are then determined based on their slope aspect data, slope gradient data, surface composition feature identification results, and initial weight values ​​for slope aspect and surface composition features. The obtained disaster response rating is directly proportional to the disaster response. For example, if the first region is an east or southeast slope with a steep slope and a stepped surface composition, it will generate a lot of precipitation, increasing the probability of disasters. The disaster response rating is set at Level 1 with a weighting ratio of 6:4 for slope aspect characteristics and surface composition characteristics. If the second region is a west or southwest slope with a gentle slope and a surface composition dominated by mountains and plateaus, it will reduce precipitation and the probability of disasters. The disaster response rating is set at Level 2 with a weighting ratio of 6:4 for slope aspect characteristics and surface composition characteristics. The difference between the first and second regions is then calculated to obtain the disaster response rating difference between the first and second regions.

[0057] Further, based on the disaster response rating area information, the area information of the first area and the area information of the second area are obtained. The difference between the obtained area information of the first area and the area information of the second area is used to obtain the area rating difference value. For example, 5 square kilometers and below is set as the first area rating level, more than 5 square kilometers and less than 10 square kilometers is set as the second area rating level, and more than 10 square kilometers and less than 15 square kilometers is set as the third area rating level. If the area of ​​the first area is 3 square kilometers, it meets the first area rating level. If the area of ​​the second area is 11 square kilometers, it meets the second area rating level. Therefore, the area rating difference between the first area and the second area is 2.

[0058] An adjacent area aggregation threshold is set, which is 0-1. That is, the difference between the aggregation values ​​of two adjacent areas is 0 or 1. The adjacent area aggregation value is obtained by subtracting the difference between the disaster response rating and the area rating. When the obtained adjacent area aggregation value meets the set adjacent area aggregation threshold, the first area and the second area are merged. At the same time as merging the areas, the disaster response rating level corresponding to the larger area of ​​the first area and the second area is used as the disaster response rating level of the merged area, providing a reference for disaster risk early warning information.

[0059] Furthermore, step S4337 of this application also includes:

[0060] Step S43371: Set the merge area constraint value in the region merging process;

[0061] Step S43372: Determine whether the small areas in the first region and the second region satisfy the merged area constraint value;

[0062] Step S43373: When the small area does not meet the merged area constraint value, the first area and the second area are not merged.

[0063] Specifically, based on the aforementioned merging of the first and second regions, a merged region is formed by merging the larger areas of the first and second regions. A merging area constraint value is set, and then it is determined whether the smaller areas of the resulting first and second regions meet the merging area constraint value. If the smaller areas do not meet the set merging area constraint value, the merging of the first and second regions is not performed. For example, if the area of ​​the first region is 7 square kilometers and the area of ​​the second region is 15 square kilometers, and the merging area constraint value is set at 10 square kilometers, then the smaller area of ​​the first and second regions does not meet the merging area constraint value, and therefore the merging of the first and second regions is not performed. This provides an important basis for obtaining the disaster response rating level.

[0064] Furthermore, step S43373 of this application also includes:

[0065] Step S433731: When multiple adjacent regions exist simultaneously for merging, execute the sequential merge instruction;

[0066] Step S433732: Merge adjacent regions in descending order of their aggregated values ​​using the sequential merge instruction.

[0067] Specifically, when merging regions involves three or more adjacent regions, they are determined to be multiple adjacent regions. An execution order instruction is then generated, which sorts the adjacent regions from largest to smallest based on the obtained aggregate values. The larger the aggregate value of the adjacent regions, the smaller the difference in their grade and the larger the difference in their area. The regions are then merged from largest to smallest according to the merging order instruction, which improves the accuracy of the disaster response rating and significantly enhances the credibility of the final disaster risk warning information.

[0068] Example 2

[0069] Based on the same inventive concept as the composite disaster risk prediction method in the foregoing embodiments, such as Figure 4 As shown, this application provides a composite disaster risk prediction system, the system comprising:

[0070] Regional basic information module 1, which is used to collect and obtain regional basic information of the target supervision area;

[0071] Data acquisition path module 2 is used to perform grid division based on the regional basic information and generate data acquisition path data based on the grid division result.

[0072] Image acquisition module 3 is used to acquire images of the target supervision area based on the data acquisition path data of the intelligent device, and obtain image acquisition results;

[0073] The disaster response division interval module 4 is used to perform feature filtering based on the regional basic information, perform feature recognition based on the image acquisition results, and generate disaster response division intervals based on the geomorphological feature recognition results.

[0074] The multi-level geological hazard sensitivity index module 5 is used to obtain geological feature matching results, perform geological identification of the disaster response division intervals based on the regional basic information and the geological feature matching results, and generate a multi-level geological hazard sensitivity index based on the identification results.

[0075] The disaster risk early warning information module 6 is used to collect rainfall forecast data and generate disaster risk early warning information based on the rainfall forecast data and the multi-level geological disaster sensitivity index.

[0076] Furthermore, the system also includes:

[0077] The activity information collection module is used to collect activity information based on the basic information of the region and obtain the activity information collection results.

[0078] The activity geological impact level module is used to obtain activity geological impact level data based on the activity information collection results, wherein the activity geological impact level data has an activity area identifier;

[0079] The geological information adjustment result module is used to adjust the identified geological information based on the activity geological impact level data to obtain the geological information adjustment result.

[0080] A multi-level geological hazard sensitivity index generation module is used to generate the multi-level geological hazard sensitivity index based on the geological information adjustment results.

[0081] Furthermore, the system also includes:

[0082] The slope aspect feature recognition result module is used to perform slope aspect feature recognition on the image acquisition result based on the feature filtering result, and obtain the slope aspect feature recognition result;

[0083] The surface composition feature recognition result module is used to identify the surface composition features of the image acquisition results based on the feature filtering results, and obtain the surface composition feature recognition result.

[0084] The disaster response division interval generation module is used to generate the disaster response division interval based on the slope aspect feature identification result and the surface composition feature identification result.

[0085] Furthermore, the system also includes:

[0086] The initial weight value construction module is used to construct the initial weight values ​​for slope aspect features and land surface composition features.

[0087] The data acquisition module is used to obtain slope aspect data and slope data based on the slope aspect feature identification results.

[0088] The disaster response classification result module is used to perform disaster response rating based on the slope aspect data, the slope data, the surface composition feature identification results and the initial weight value, and obtain the disaster response classification result based on the disaster response rating result.

[0089] Furthermore, the system also includes:

[0090] The information acquisition module is used to obtain disaster response rating level information and disaster response rating area information based on the disaster response rating results.

[0091] The threshold setting module is used to set the aggregation threshold for adjacent regions.

[0092] The disaster response rating difference module is used to obtain the disaster response rating difference between the first region and the second region based on the disaster response rating level information, wherein the first region and the second region are adjacent regions.

[0093] The area information acquisition module is used to obtain the area information of the first region and the area information of the second region through the area information of the disaster response rating region.

[0094] The area rating difference module is used to obtain the area rating difference based on the area information of the first area and the area information of the second area.

[0095] The adjacent area aggregation value module is used to calculate the adjacent area aggregation value based on the disaster response rating difference and the area rating difference.

[0096] The disaster response rating module is used to determine whether the aggregate value of the adjacent areas meets the adjacent area aggregation threshold. When the adjacent area aggregation threshold is met, the first area and the second area are merged, and the disaster response rating level corresponding to the larger area in the first area and the second area is used as the disaster response rating level of the merged area.

[0097] Furthermore, the system also includes:

[0098] The merge area constraint value module is used to set the merge area constraint value in the area merging process.

[0099] The first judgment module is used to determine whether the small-area regions in the first region and the second region meet the merged area constraint value.

[0100] The second judgment module is used to prevent the merging of the first region and the second region when the small area does not meet the merging area constraint value.

[0101] Furthermore, the system also includes:

[0102] The sequential merge instruction execution module is used to execute sequential merge instructions when multiple adjacent regions exist to be merged simultaneously.

[0103] The merging module is used to merge adjacent regions in descending order of their aggregated values ​​according to the order of their values, based on the sequential merging instructions.

[0104] Through the foregoing detailed description of a method for predicting a complex disaster risk, those skilled in the art can clearly understand the method and system for predicting a complex disaster risk in this embodiment. As for the apparatus disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and relevant parts can be referred to the method section description.

[0105] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for predicting composite disaster risks, characterized in that, The method includes: Collect and obtain basic regional information of the target monitoring area; Based on the basic information of the region, a grid is divided, and data collection path data is generated based on the grid division results. Image acquisition results are obtained by acquiring images of the target supervision area based on the data acquisition path data using a smart device. Based on the regional basic information, feature selection is performed in the feature matching library; based on the feature selection results, feature recognition is performed on the image acquisition results; and based on the geomorphological feature recognition results, disaster response division intervals are generated. Obtain geological feature matching results, and based on the regional basic information and the geological feature matching results, perform geological identification of the disaster response division intervals, and generate multi-level geological disaster sensitivity indices based on the identification results; Rainfall forecast data is collected and obtained, and disaster risk early warning information is generated based on the rainfall forecast data and the multi-level geological hazard sensitivity index. The method further includes: Based on the basic information of the region, activity information is collected to obtain the activity information collection results; Based on the analysis of the activity information collection results, activity geological impact level data is obtained, wherein the activity geological impact level data has an activity area identifier; The geological information of the identifier is adjusted by using the activity geological impact level data to obtain the geological information adjustment result; The multi-level geological hazard sensitivity index is generated based on the adjustment results of the geological information. The method further includes: Based on the disaster response rating results, obtain disaster response rating level information and disaster response rating area information; Set an aggregation threshold for adjacent regions; The disaster response rating difference between the first region and the second region is obtained based on the disaster response rating information, wherein the first region and the second region are adjacent regions; The area information of the first region and the area information of the second region are obtained through the disaster response rating area information; The area rating difference is obtained based on the area information of the first region and the area information of the second region; The aggregate value of adjacent areas is calculated based on the disaster response rating difference and the area rating difference; Determine whether the aggregate value of the adjacent regions meets the aggregate threshold of the adjacent regions. If the aggregate threshold of the adjacent regions is met, then merge the first region and the second region, and use the disaster response rating level corresponding to the large area in the first region and the second region as the disaster response rating level of the merged region. The method further includes: Set the area constraint value for merging regions; Determine whether the small-area regions in the first region and the second region satisfy the merged area constraint value; If the small area does not meet the merged area constraint value, then the first area and the second area will not be merged.

2. The method as described in claim 1, characterized in that, The method further includes: Based on the feature selection results, the slope aspect features of the image acquisition results are identified to obtain the slope aspect feature identification results; Based on the feature filtering results, the surface composition features of the image acquisition results are identified to obtain the surface composition feature identification results; The disaster response division interval is generated based on the slope aspect feature identification results and the surface composition feature identification results.

3. The method as described in claim 2, characterized in that, The method further includes: Construct initial weight values ​​for slope aspect features and land surface composition features; Based on the slope aspect feature identification results, slope aspect data and slope gradient data are obtained; Disaster response rating is performed based on the slope aspect data, the slope data, the surface composition feature identification results, and the initial weight values, and the disaster response classification results are obtained based on the disaster response rating results.

4. The method as described in claim 1, characterized in that, The method further includes: When multiple adjacent regions exist simultaneously, a sequential merge instruction is executed. The sequential merge instruction merges adjacent regions in descending order of their aggregated values.

5. A composite disaster risk prediction system, employing the method described in any one of claims 1-4, characterized in that, The system includes: A regional basic information module, which is used to collect and obtain regional basic information of the target supervision area; A data acquisition path module is used to perform grid division based on the basic information of the region and generate data acquisition path data based on the grid division results. An image acquisition module is used to acquire images of the target supervision area based on the data acquisition path data using a smart device, and to obtain image acquisition results. The disaster response division interval module is used to perform feature filtering based on the regional basic information, perform feature recognition based on the image acquisition results, and generate disaster response division intervals based on the geomorphological feature recognition results. A multi-level geological hazard sensitivity index module is used to obtain geological feature matching results, perform geological identification of the disaster response division intervals based on the regional basic information and the geological feature matching results, and generate a multi-level geological hazard sensitivity index based on the identification results. The disaster risk early warning information module is used to collect rainfall forecast data and generate disaster risk early warning information based on the rainfall forecast data and the multi-level geological disaster sensitivity index.

Images