An artificial intelligence-based dam and levee breach scene unmanned aerial vehicle monitoring method and system

Abstract

The application relates to a kind of artificial intelligence-based dam-break scene unmanned aerial vehicle monitoring method and system.The method comprises: obtaining hydrological data, meteorological data, satellite image data and topographic map data, and inputting into monitoring area identification artificial intelligence model, generating target monitoring area risk level information and target monitoring area detection range information;Based on target monitoring area risk level information and target monitoring area detection range information, the number of unmanned aerial vehicle and the preset route of each unmanned aerial vehicle are allocated;Control each unmanned aerial vehicle to fly along the corresponding preset route, obtain the target monitoring data of each unmanned aerial vehicle;The target monitoring data of each unmanned aerial vehicle is input into dam-break scene analysis artificial intelligence model, and dam-break scene analysis result is generated.The method can realize the efficiency, accuracy and real-time of dam-break monitoring through multi-source data driving, dynamic task allocation and artificial intelligence model optimization.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of flood monitoring, and particularly relates to a dam and levee breach scene unmanned aerial vehicle (UAV) monitoring method and system based on artificial intelligence. BACKGROUND

[0002] Dam and levee breach disasters have the characteristics of strong suddenness and great destructiveness, and pose a serious threat to people's life and property safety. Traditional manual inspection methods are inefficient and have limited coverage, and it is difficult to meet the real-time monitoring needs in complex scenarios. With the rapid development of UAV technology, UAVs have become an important means of dam and levee breach monitoring due to their high mobility and multi-angle observation capability. Intelligent monitoring technology based on UAVs has gradually been applied in the field of water conservancy engineering safety.

[0003] In the traditional technology, the UAV monitoring system usually adopts fixed route inspection, which cannot dynamically adjust the monitoring strategy according to the real-time risk, resulting in problems such as missed detection of high-risk areas and redundant monitoring of low-risk areas. Moreover, the traditional UAV monitoring method usually requires manual remote control operation, which requires higher skills and experience of the operators, not only increasing the labor cost and operation risk, but also making the reaction speed unable to meet the demand for rapid response in the face of unexpected situations.

[0004] In the prior art with the authorization announcement number CN110044338B, a specific dam and levee breach scene UAV monitoring scheme can be generated only after a fixed-wing UAV flies along a first preset route for preliminary detection, and it is difficult to predict the dam and levee breach that has not occurred in the dam and levee breach scene, which has defects such as slow response speed, weak end-cloud collaboration and abnormal identification capability, and poor system robustness. SUMMARY

[0005] Therefore, it is necessary to provide a dam and levee breach scene UAV monitoring method and system based on artificial intelligence which can quickly respond to the above technical problems.

[0006] In a first aspect, the application provides a dam and levee breach scene UAV monitoring method based on artificial intelligence, comprising:

[0007] Obtaining hydrological data, meteorological data, satellite image data and topographic map data, and inputting the hydrological data, meteorological data, satellite image data and topographic map data into a monitoring area identification artificial intelligence model to perform target detection area identification, and generating target monitoring area risk level information and target monitoring area detection range information;

[0008] Based on the target monitoring area risk level information and the target monitoring area detection range information, the number of UAVs and the preset flight route of each UAV are allocated;

[0009] Control each unmanned aerial vehicle to fly along the corresponding preset flight path, and obtain target monitoring data of each unmanned aerial vehicle;

[0010] Input the target monitoring data of each unmanned aerial vehicle into a dam and dam breach scene analysis artificial intelligence model, perform dam and dam breach scene analysis, and generate dam and dam breach scene analysis results, which include dam and dam breach risk information and / or dam and dam breach disaster information.

[0011] In one embodiment, the number of unmanned aerial vehicles and the preset flight path of each unmanned aerial vehicle are assigned based on the target monitoring area risk level information and the target monitoring area detection range information, which includes:

[0012] The maximum operation time of the unmanned aerial vehicle and the shooting range of the on-board monitoring device are set based on the target monitoring area risk level information;

[0013] The maximum flight time of the unmanned aerial vehicle is calculated based on the maximum distance between the initial position of the unmanned aerial vehicle and the target monitoring area detection range, and the maximum flight time is used to represent the maximum time for the unmanned aerial vehicle to reach the target monitoring area detection range from the initial position;

[0014] The preset operation area of the unmanned aerial vehicle is calculated based on the maximum operation time, the shooting range of the on-board monitoring device, and the maximum flight time;

[0015] The number of unmanned aerial vehicles is assigned based on the preset operation area and the target monitoring area detection range information;

[0016] The preset flight path of each unmanned aerial vehicle is assigned based on the number of unmanned aerial vehicles, the target monitoring area risk level information, and the target monitoring area detection range information.

[0017] In one embodiment, the number of unmanned aerial vehicles is calculated according to the following formula:

[0018]

[0019]

[0020]

[0021] In the formula, N is the number of unmanned aerial vehicles, is a rounding up function, is the total area of the target monitoring area detection range, is an operation track overlap correction factor, is the maximum operation time, is the maximum flight time, is an environmental condition speed correction factor, is the average flight speed of the unmanned aerial vehicle, a width of a shooting range of the airborne monitoring device, a fuzzy control membership function of the maximum operation time length, a risk level of the target monitoring area, a fuzzy control membership function of the shooting range of the airborne monitoring device.

[0022] In one of the embodiments, the dam / levee breach scene analysis result includes the second target monitoring area risk level information and the second target monitoring area detection range information, and the dam / levee breach scene unmanned aerial vehicle monitoring method based on artificial intelligence further includes:

[0023] calculating a risk level error between the target monitoring area risk level information and the second target monitoring area risk level information, and determining whether the risk level error exceeds a risk level deviation threshold;

[0024] calculating a detection range error between the target monitoring area detection range information and the second target monitoring area detection range information, and determining whether the detection range error exceeds a detection range deviation threshold;

[0025] if the risk level error exceeds the risk level deviation threshold and / or the detection range error exceeds the detection range deviation threshold, adjusting the number of unmanned aerial vehicles based on the second target monitoring area risk level information and the second target monitoring area detection range information, and updating the preset flight path of each unmanned aerial vehicle based on the adjusted number of unmanned aerial vehicles, the target monitoring area risk level information, the target monitoring area detection range information, and the monitored area information of the unmanned aerial vehicle; wherein the monitored area information is generated based on the target monitoring data of the unmanned aerial vehicle;

[0026] adjusting and optimizing the monitoring area identification artificial intelligence model based on the target monitoring area risk level information, the target monitoring area detection range information, the second target monitoring area risk level information, and the second target monitoring area detection range information.

[0027] In one of the embodiments, the dam / levee breach scene analysis artificial intelligence model includes a dam / levee breach risk analysis sub-model and a dam / levee breach disaster prediction sub-model, and the dam / levee breach scene analysis result includes dam / levee breach risk information and dam / levee breach disaster information. The target monitoring data of each unmanned aerial vehicle is input into the dam / levee breach scene analysis artificial intelligence model for dam / levee breach scene analysis to generate the dam / levee breach scene analysis result, including:

[0028] inputting the target monitoring data into the dam / levee breach risk analysis sub-model to generate dam / levee breach risk area information and dam / levee breach risk area information corresponding dam / levee breach risk information, and the dam / levee breach risk level label includes a dam / levee breach disaster label;

[0029] If the risk information of dam breach is a dam breach disaster label, the target monitoring data and topographic map data corresponding to the dam breach disaster label are input into the dam breach disaster prediction sub-model to perform disaster prediction analysis and generate dam breach disaster information.

[0030] In one embodiment, the dam failure risk analysis sub-model is a deep learning neural network model built based on an improved YOLOV8 model;

[0031] The sub-model for analyzing dam and dike breach disasters is a deep learning neural network model built based on a U-shaped network combined with a converter module.

[0032] In one embodiment, the AI-based drone monitoring method for dam and levee breach scenarios further includes:

[0033] The target monitoring data is input into the artificial intelligence model for identifying the edge of dam breach anomalies, and dam breach anomalies are identified to generate dam breach identification marker information, dam breach anomaly area information, and dam breach anomaly degree information.

[0034] If the identification information for dam breaches indicates that a dam breach has occurred, control the drone to hover and monitor the area corresponding to the abnormal dam breach area information.

[0035] Based on information on abnormal areas of dam breaches, information on the degree of abnormality of dam breaches, information on the risk level of the target monitoring area, and information on the detection range of the target monitoring area, the validity of the risk level information of the target monitoring area and the validity of the detection range information of the target monitoring area are determined.

[0036] If the determination of the validity of the risk level information of the target monitoring area is that the risk level information of the target monitoring area is invalid and / or the determination of the validity of the detection range information of the target monitoring area is that the detection range information of the target monitoring area is invalid, an emergency report of dam breach disaster is generated. The emergency report of dam breach disaster is used to indicate that the results output by the monitoring area identification artificial intelligence model are inconsistent with the actual disaster situation.

[0037] In one embodiment, the monitoring area identification artificial intelligence model includes a first feature extraction module, a second feature extraction module, a feature fusion module, and a feature decoding module. Hydrological data, meteorological data, satellite imagery data, and topographic map data are input into the monitoring area identification artificial intelligence model to identify target detection areas and generate target monitoring area risk level information and target monitoring area detection range information, including:

[0038] Hydrological data, meteorological data, and topographic map data are input into the first feature extraction module to generate meteorological and hydrological feature maps and corresponding meteorological and hydrological anomaly feature values.

[0039] Satellite imagery data and topographic map data are input into the second feature extraction module to generate satellite imagery feature maps and corresponding satellite imagery anomaly feature values.

[0040] If the meteorological and hydrological anomaly characteristic value exceeds the preset meteorological and hydrological anomaly threshold or the satellite image anomaly characteristic value exceeds the preset satellite image anomaly threshold, the meteorological and hydrological feature map and the satellite image feature map are input into the feature fusion module to generate a fused feature map.

[0041] The fused feature map is input into the feature decoding module to generate risk level information and detection range information of the target monitoring area.

[0042] In one embodiment, each drone is controlled to fly along a corresponding preset route to acquire target monitoring data for each drone, including:

[0043] Control each drone to fly along its corresponding preset flight path and acquire image data of the target monitoring sub-scene;

[0044] The target monitoring sub-scene image data is input into the image stitching edge artificial intelligence model to perform image stitching and generate target monitoring data.

[0045] Secondly, this application also provides an artificial intelligence-based drone monitoring system for dam and dike breach scenarios, comprising:

[0046] The monitoring area data analysis module is used to acquire hydrological data, meteorological data, satellite imagery data, and topographic map data, and input the hydrological data, meteorological data, satellite imagery data, and topographic map data into the monitoring area identification artificial intelligence model to identify the target detection area and generate target monitoring area risk level information and target monitoring area detection range information.

[0047] The UAV configuration module is used to allocate the number of UAVs and the preset flight path for each UAV based on the risk level information and detection range information of the target monitoring area.

[0048] The UAV operation control module is used to control each UAV to fly along the corresponding preset route and acquire target monitoring data for each UAV.

[0049] The target monitoring data processing module is used to input the target monitoring data of each UAV into the artificial intelligence model for dam breach scenario analysis, perform dam breach scenario analysis, and generate dam breach scenario analysis results, which include dam breach risk information and / or dam breach disaster information.

[0050] The aforementioned AI-based drone monitoring method and system for dam and levee breach scenarios integrates hydrological, meteorological, satellite imagery, and topographic map data through an AI model based on monitoring area identification. This generates risk level and detection range information for the target monitoring area, enhancing the comprehensiveness of risk assessment and significantly improving the response speed of drone monitoring in such scenarios. By intelligently allocating drone numbers and preset flight paths based on the risk level and range of the target monitoring area, it ensures focused monitoring of high-risk areas, improves drone resource utilization efficiency, and guarantees the rationality and targeting of monitoring coverage. Controlling drones to fly along preset flight paths to acquire target monitoring data achieves automated and standardized data collection, reducing human intervention and ensuring the comprehensiveness and stability of data acquisition, providing a high-quality data foundation for subsequent analysis. Furthermore, by using an AI model for dam and levee breach scenario analysis to process the target monitoring data and generate analytical results containing risk and disaster information, it helps decision-makers quickly grasp the dynamic information of dam and levee breach disasters, providing precise support for emergency response and rescue deployment, and improving the timeliness and scientific rigor of disaster response. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 A schematic diagram illustrating the application environment of an AI-based drone monitoring method for dam and levee breach scenarios, provided as an embodiment of this application;

[0053] Figure 2 A flowchart illustrating an artificial intelligence-based drone monitoring method for dam and levee breach scenarios, provided as an embodiment of this application;

[0054] Figure 3 A schematic diagram illustrating a process for allocating the number of drones and flight routes, provided as an embodiment of this application;

[0055] Figure 4 A flowchart illustrating a feedback control method based on dam failure scenario analysis results, provided as an embodiment of this application;

[0056] Figure 5 A flowchart illustrating another AI-based drone monitoring method for dam and levee breach scenarios, provided as an embodiment of this application;

[0057] Figure 6This is a schematic diagram of the structure of a dam failure risk analysis sub-model in an artificial intelligence model for analyzing dam failure scenarios, provided in one embodiment of this application.

[0058] Figure 7 This application provides a schematic diagram of the structure of a dam breach disaster analysis sub-model in an artificial intelligence model for analyzing dam breach scenarios, as an embodiment of the present application.

[0059] Figure 8 This is a schematic diagram of the structure for deploying an artificial intelligence model according to one embodiment of this application;

[0060] Figure 9 This is a schematic diagram of the structure of an artificial intelligence-based drone monitoring system for dam and levee breach scenarios, provided as an embodiment of this application. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0062] The AI-based drone monitoring method for dam and levee breach scenarios provided in this application can be applied to, for example... Figure 1 In the application environment shown, sensing device 102 and working device 103 can communicate with computing device 101 via a communication channel. A data storage system can store the data that computing device 101 needs to process. The data storage system can be integrated into computing device 101 or placed in the cloud or on other network servers. Computing device 101 can generate dam failure disaster reports and working device control commands based on environmental data of the dam failure scenario acquired by sensing device 102. Computing device 101 can control working device 103 to perform work tasks based on the working device control commands. The computing device 101 can be, but is not limited to, various servers, smart devices, and Geographic Information System (GIS) workstations. Servers can be implemented using independent servers or server clusters composed of multiple servers. Smart devices can include, but are not limited to, various personal computers, laptops, smartphones, tablets, and IoT devices. IoT devices can include edge computing devices. The sensing device 102 may include, but is not limited to, various water level gauges, piezometers, seepage sensors, sonar detection equipment, remote sensing satellites, meteorological monitoring equipment, cameras, infrared sensing equipment, lidar, satellite navigation and positioning equipment, and structural strain sensors. The working equipment 103 may include, but is not limited to, fixed-wing UAVs, multi-rotor UAVs, and rescue equipment.

[0063] In one exemplary embodiment, such as Figure 2 As shown, an artificial intelligence-based drone monitoring method for dam and levee breach scenarios is provided, which is then applied to... Figure 1 Taking computing device 101 as an example, the explanation includes the following steps S201 to S204. Wherein:

[0064] Step S201: Obtain hydrological data, meteorological data, satellite imagery data, and topographic map data, and input the hydrological data, meteorological data, satellite imagery data, and topographic map data into the monitoring area identification artificial intelligence model to identify the target detection area and generate target monitoring area risk level information and target monitoring area detection range information.

[0065] Specifically, the computing device 101 can input hydrological data, meteorological data, satellite imagery data and topographic map data obtained from the sensing device into the monitoring area recognition artificial intelligence model mounted on the computing device 101 to identify the target detection area and generate target monitoring area risk level information and target monitoring area detection range information.

[0066] Optionally, the target monitoring area detection range information may include the detection range information of each target monitoring area corresponding to the risk level of each target monitoring area.

[0067] Optionally, when the difference between the target monitoring area risk level information and the target monitoring area detection range information is small, the target monitoring area risk level information may include the overall target monitoring area risk level information used to measure the overall risk level of the target monitoring area detection range information.

[0068] Optionally, when there are significant differences in the target monitoring area risk level information among the target monitoring area detection range information, the target monitoring area risk level information may include the local target monitoring area risk level information corresponding to local target monitoring areas with the same local target monitoring area risk level information.

[0069] Furthermore, the computing device 101 can construct a contour map of the target monitoring area risk level information based on the target monitoring area risk level information and the target monitoring area detection range information generated by the monitoring area identification artificial intelligence model.

[0070] Schematic illustration: The monitoring area identification artificial intelligence model mounted on computing device 101 can analyze the temporal and / or spatial dependencies of hydrological data, meteorological data, satellite imagery data, and topographic map data based on sequence models. The sequence model may include, but is not limited to, recurrent neural networks (RNNs), long short-term memory networks (LSTMs), gated recurrent unit networks (GRUs), temporal convolutional networks (TCNs), and transformer models.

[0071] Furthermore, the monitoring area identification artificial intelligence model mounted on the computing device 101 can perform image segmentation and target detection on hydrological data, meteorological data, satellite image data and topographic map data from which temporal and / or spatial dependencies have been extracted, based on convolutional neural network (CNN) models and their variant models, to generate target monitoring area risk level information and target monitoring area detection range information.

[0072] Optionally, variant models of convolutional neural network models may include, but are not limited to, AlexNet, VGGNet, ResNet, Inception series models, U-Net, YOLO series models, and SegNet.

[0073] Step S202: Allocate the number of drones and the preset flight path for each drone based on the risk level information and detection range information of the target monitoring area.

[0074] Specifically, the computing device 101 can allocate the number of drones based on the target monitoring area risk level information and target monitoring area detection range information generated by the monitoring area recognition artificial intelligence model mounted on it, and allocate a preset flight path for each drone based on the drone allocation result combined with the target monitoring area risk level information and target monitoring area detection range information.

[0075] Optionally, the computing device 101 can allocate a preset flight path for each UAV based on an intelligent optimization algorithm, combining the UAV quantity allocation result with the target monitoring area risk level information and the target monitoring area detection range information. The intelligent optimization algorithm can include, but is not limited to, evolutionary algorithms, swarm intelligence algorithms, simulated annealing algorithms, and model-based intelligent optimization algorithms. Evolutionary algorithms can include, but are not limited to, genetic algorithms (GA) and immune algorithms (IA). Swarm intelligence algorithms can include, but are not limited to, ant colony optimization (ACO), artificial bee colony optimization (ABC), particle swarm optimization (PSO), and wolf pack optimization (WCA).

[0076] Step S203: Control each UAV to fly along the corresponding preset route and acquire target monitoring data for each UAV.

[0077] Specifically, the computing device 101 can control the preset flight path of each drone based on the generated preset flight path of each drone, acquire the detection data of each drone, and generate the target monitoring data of each drone after preprocessing the detection data of each drone. The preprocessing may include, but is not limited to, image enhancement, image filtering, image registration, and image stitching.

[0078] Optionally, the drone may include, but is not limited to, fixed-wing and rotary-wing drones. The sensors carried by the drone may include, but are not limited to, thermal infrared cameras, LiDAR, high-definition cameras, GPS / RTK modules, and multispectral cameras. Among these, thermal infrared cameras can be used to detect leakage temperature differences and search for personnel, while LiDAR can be used for precise ranging and terrain modeling.

[0079] Optionally, drones can be deployed with lightweight artificial intelligence models to enable real-time data processing.

[0080] Step S204: Input the target monitoring data of each drone into the artificial intelligence model for dam breach scenario analysis, perform dam breach scenario analysis, and generate dam breach scenario analysis results.

[0081] Specifically, the computing device 101 can input the target monitoring data of each UAV into the artificial intelligence model for dam breach scenario analysis carried on the computing device 101 to perform dam breach scenario analysis and generate dam breach scenario analysis results. The dam breach scenario analysis results may include dam breach risk information and / or dam breach disaster information.

[0082] This is an illustrative representation of the main structure of the artificial intelligence model for analyzing dam and levee breach scenarios mounted on computing device 101, which can be constructed based on convolutional neural network models and variant models of convolutional neural networks.

[0083] Optionally, when the computing device 101 does not identify any existing dike or dam breach phenomena, the dike or dam breach scenario analysis results may include dike or dam breach risk information; when the computing device 101 identifies any existing dike or dam breach phenomena, the dike or dam breach scenario analysis results may include dike or dam breach disaster information; when the computing device 101 identifies both existing and impending dike or dam breach phenomena, the dike or dam breach scenario analysis results may include both dike or dam breach risk information and dike or dam breach disaster information.

[0084] Furthermore, the computing device 101 can identify both past and future dam failures based on the artificial intelligence model for analyzing dam failure scenarios mounted on the computing device 101.

[0085] For example, the dam failure scenario analysis artificial intelligence model mounted on computing device 101 can generate dam failure disaster information that includes, but is not limited to, disaster scope information and disaster severity information. Computing device 101 can generate early warning level information based on the disaster scope information and disaster severity information.

[0086] The aforementioned AI-based drone monitoring method for dam and levee breach scenarios integrates multi-source data to comprehensively consider the impact of various factors on dam and levee breaches. By constructing an AI model for monitoring area identification, it can automatically identify target areas requiring key monitoring and generate risk level and detection range information for these areas. This allows for dynamic adjustment of the drone's monitoring strategy, improving monitoring efficiency and emergency response speed. By inputting the collected target monitoring data into the AI ​​model for dam and levee breach scenario analysis, it enables in-depth analysis of risk information and / or disaster information related to dam and levee breaches. This achieves intelligent and real-time processing throughout the entire process from data collection to analysis result generation, significantly shortening the time interval from monitoring to decision-making. This allows for rapid response in the face of sudden dam and levee breach events, buying valuable time for emergency rescue.

[0087] As an illustration, the above-mentioned AI-based drone monitoring method for dam and levee breach scenarios, through multi-source data fusion, AI model application, and drone monitoring, can achieve accurate and efficient real-time monitoring and analysis of dam and levee breach scenarios, providing strong technical support for flood control and disaster reduction efforts.

[0088] In one alternative embodiment, such as Figure 3As shown, the number of drones and the preset flight path for each drone are allocated based on the risk level information and detection range information of the target monitoring area, including:

[0089] Step S301: Set the maximum operating time of the UAV and the shooting range of the airborne monitoring equipment based on the risk level information of the target monitoring area.

[0090] Specifically, the computing device can obtain the maximum operating time of the drone and the shooting range of the onboard monitoring equipment corresponding to the risk level information of the target monitoring area based on preset drone operation parameter setting rules. Among them, the maximum operating time of the drone can be used to characterize the maximum monitoring time of the drone within the target monitoring area corresponding to the target monitoring area detection range information, and the maximum operating time of the drone can be used to measure the response speed of drone monitoring in the scenario of dam breach.

[0091] Optionally, the preset drone operation parameter setting rules can be, but are not limited to, data mapping tables, calculation formulas, and BP neural networks.

[0092] Step S302: Calculate the maximum flight time of the UAV based on the maximum distance between the initial position of the UAV and the detection range of the target monitoring area.

[0093] Specifically, the computing device can calculate the maximum flight time of the UAV based on the maximum distance between the UAV's initial position and the detection range of the target monitoring area. The maximum flight time of the UAV can be used to characterize the maximum time it takes for the UAV to travel from its initial position to the detection range of the target monitoring area. The initial position of the UAV can be set as the coordinate position of the UAV base station.

[0094] Step S303: Calculate the preset operating area of ​​the UAV based on the maximum operating time, the shooting range of the airborne monitoring equipment, and the maximum flight time.

[0095] Step S304: Allocate the number of drones based on the preset operating area area and target monitoring area detection range information.

[0096] Step S305: Assign a preset flight path to each drone based on the number of drones, the risk level information of the target monitoring area, and the detection range information of the target monitoring area.

[0097] The aforementioned AI-based drone monitoring method for dam and levee breach scenarios can dynamically adjust the number and flight paths of drones based on real-time monitoring data, thereby addressing changes in the risk level and detection range of the monitoring area and improving the flexibility and adaptability of dam and levee breach scenario monitoring.

[0098] Specifically, the aforementioned AI-based drone monitoring method for dam and levee breach scenarios achieves efficient and accurate monitoring of dam and levee breach scenarios through dynamic resource allocation, optimized operation time and shooting range, precise flight time calculation, and reasonable operation area planning, providing strong technical support for flood control and disaster reduction efforts.

[0099] In one alternative embodiment, the formula for calculating the number of drones is:

[0100]

[0101]

[0102]

[0103] In the formula, For the number of drones, It is a rounding function. The total area of ​​the target monitoring area. This is the correction factor for overlap of operational flight paths. For maximum operation time, For maximum sailing time, Environmental condition speed correction factor, The average flight speed of the drone. The width of the field of view captured by the airborne surveillance equipment. The fuzzy control membership function for the maximum operation time. The risk level of the target monitoring area, A fuzzy control membership function for the shooting range of airborne monitoring equipment.

[0104] Optional, risk level of the target monitoring area It can be obtained by weighted summation of the risk level information of local target monitoring areas corresponding to local target monitoring areas with the same risk level information.

[0105] Furthermore, the risk level of the target monitoring area. The calculation formula can be:

[0106]

[0107] In the formula, The risk level of the target monitoring area, The number of types of risk level information for local target monitoring areas. For the first The weighting of risk level information for local target monitoring areas For the first The risk level corresponding to the risk level information of the local target monitoring area For the first The area of ​​the local target monitoring area corresponding to the risk level information of the local target monitoring area. This refers to the total area of ​​the target monitoring area corresponding to the detection range information of the target monitoring area.

[0108] In one alternative embodiment, such as Figure 4 As shown, the analysis results of the dam breach scenario include risk level information and detection range information of the second target monitoring area. The AI-based UAV monitoring method for dam breach scenarios also includes:

[0109] Step S401: Calculate the risk level error between the risk level information of the target monitoring area and the risk level information of the second target monitoring area, and determine whether the risk level error exceeds the risk level deviation threshold.

[0110] Step S401: Calculate the detection range error between the detection range information of the target monitoring area and the detection range information of the second target monitoring area, and determine whether the detection range error exceeds the detection range deviation threshold.

[0111] Step S401: If the risk level error exceeds the risk level deviation threshold and / or the detection range error exceeds the detection range deviation threshold, adjust the number of drones based on the risk level information of the second target monitoring area and the detection range information of the second target monitoring area, and update the preset flight path of each drone based on the adjusted number of drones, the risk level information of the target monitoring area, the detection range information of the target monitoring area, and the monitored area information of the drones.

[0112] Specifically, if the error in determining the risk level by the calculation device exceeds the risk level deviation threshold and / or the error in determining the detection range by the calculation device exceeds the detection range deviation threshold, the calculation device can recalculate and adjust the number of drones based on the risk level information of the second target monitoring area and the detection range information of the second target monitoring area, and update the preset flight path of each drone based on the adjusted number of drones, the risk level information of the target monitoring area, the detection range information of the target monitoring area, and the monitored area information of the drones; wherein, the monitored area information can be generated based on the target monitoring data of the drones.

[0113] Step S401: Adjust and optimize the artificial intelligence model for identifying the monitoring area based on the risk level information of the target monitoring area, the detection range information of the target monitoring area, the risk level information of the second target monitoring area, and the detection range information of the second target monitoring area.

[0114] Optionally, the computing device can be based on the consistency of the distributions predicted twice by KL divergence constraints.

[0115] In the aforementioned AI-based drone monitoring method for dam and levee breach scenarios, by judging whether the risk level error and detection range error exceed preset deviation thresholds, anomalies or deviations in the monitoring results can be detected in a timely manner, ensuring the accuracy and timeliness of the monitoring data. When the risk level error or detection range error exceeds the threshold, the number of drones is dynamically adjusted based on the risk level information and detection range information of the second target monitoring area. This ensures that the number of drones can be increased in a timely manner in high-risk areas or areas with significant changes in detection range, thereby improving the coverage and accuracy of dam and levee breach monitoring. By combining the risk level information and detection range information of the target monitoring area and the second target monitoring area, the AI ​​model for monitoring area identification can be adjusted and optimized. This helps the model to continuously learn and adapt to new monitoring data and environmental changes, improving the accuracy and robustness of the AI ​​model for monitoring area identification.

[0116] In one alternative embodiment, please refer to Figure 5 The AI ​​model for analyzing dam breach scenarios includes a dam breach risk analysis sub-model and a dam breach disaster prediction sub-model. The analysis results include dam breach risk information and dam breach disaster information. Target monitoring data from each drone is input into the AI ​​model for dam breach scenario analysis to generate analysis results, including:

[0117] Step S507: Input the target monitoring data into the dam failure risk analysis sub-model to generate dam failure risk area information and dam failure risk information corresponding to the dam failure risk area information.

[0118] Specifically, the computing equipment can input target monitoring data into the dam breach risk analysis sub-model to generate dam breach risk area information and corresponding dam breach risk information. The dam breach risk level label can include a dam breach disaster label.

[0119] Step S508: If the risk information of dam breach and collapse is a dam breach and collapse disaster label, input the target monitoring data and topographic map data corresponding to the dam breach and collapse disaster label into the dam breach and collapse disaster prediction sub-model, perform disaster prediction analysis, and generate dam breach and collapse disaster information.

[0120] Optionally, information on dam or breach disasters may include information on the scope of the disaster and the level of the disaster.

[0121] The aforementioned AI-based drone monitoring method for dam and breach scenarios divides the analysis of dam and breach scenarios into two stages: risk analysis and disaster prediction. This enables refined analysis of complex scenarios and improves the real-time performance and stability of dam and breach detection.

[0122] In one optional embodiment, the dam failure risk analysis sub-model is a deep learning neural network model built based on an improved YOLOv8 model;

[0123] The sub-model for analyzing dam and dike breach disasters is a deep learning neural network model built based on a U-shaped network combined with a converter module.

[0124] In an optional embodiment, please refer to Figure 6 The dam failure risk analysis sub-model in the artificial intelligence model for dam failure scenario analysis can be a deep learning neural network model built based on an improved YOLOv8 model. The improved YOLOv8 model can be obtained by replacing the fast double convolutional partially connected (C2F) modules in the spine network with fast double convolutional partially connected dilated residual convolutional modules; replacing the neck network with a bidirectional feature pyramid structure incorporating an attention mechanism; and adding a global attention mechanism to both the spine and neck networks. Specifically, the fast double convolutional partially connected dilated residual convolutional modules can be obtained by replacing the convolutional modules in the C2F module with dilated residual convolutional modules, and the bidirectional feature pyramid network incorporating an attention mechanism can be obtained by adding a permutation attention module to the skip connection structure of the initial bidirectional feature pyramid network.

[0125] As an illustration, the detection module of the head network of the dam breach risk analysis sub-model in the artificial intelligence model for dam breach scenario analysis can include a dam breach risk area information detection head and a dam breach risk information detection head.

[0126] In an optional embodiment, please refer to Figure 7 The dam breach disaster analysis sub-model in the artificial intelligence model for dam breach scenario analysis can be a deep learning neural network model built based on a U-shaped network combined with a transformer module. This deep learning neural network model can include a disaster feature fusion module and a disaster feature decoding module. The disaster feature fusion module can be used to fuse topographic map data features and target detection data features to generate a fused disaster feature map. The disaster feature decoding module can be built based on a U-shaped network combined with a residual convolution module and a transformer module. This module can be used to parse the fused disaster feature map and generate dam breach disaster information.

[0127] Optionally, the downsampling layer of the disaster feature decoding module may include, but is not limited to, a max pooling layer.

[0128] In one alternative embodiment, please refer to Figure 5 and Figure 8 The AI-based drone monitoring method for dam and levee breach scenarios also includes:

[0129] Step S509: Input the target monitoring data into the artificial intelligence model for identifying the edge of dam breach anomalies, perform dam breach anomaly identification, and generate dam breach identification marker information, dam breach anomaly area information, and dam breach anomaly degree information.

[0130] Optionally, the edge AI model for identifying dam breach anomalies can be deployed on edge computing devices.

[0131] As an illustration, computing devices can include central computing devices and edge computing devices. Central computing devices can be used to deploy AI models for monitoring area identification and AI models for analyzing dam and levee breach scenarios.

[0132] As an illustration, edge devices can be, but are not limited to, deployed on drones.

[0133] Step S510: If the dam breach identification information indicates that a dam breach exists, control the drone to hover and monitor the area corresponding to the abnormal dam breach area information.

[0134] Specifically, if the edge computing device's AI-powered edge model for identifying dam breaches indicates a dam breach, the edge computing device can control a drone to hover and monitor the area corresponding to the anomaly. At this point, the edge computing device can generate a drone hovering report and send it to the central computing device. Upon receiving the drone hovering report, the central computing device can determine whether to dispatch additional drones. If the determination indicates that additional drones are needed, the central computing device can generate the required number of additional drones and a preset flight path.

[0135] Step S511: Based on the information on abnormal areas of dike and dam breaches, the information on the degree of abnormality of dike and dam breaches, the information on the risk level of the target monitoring area, and the information on the detection range of the target monitoring area, determine the validity of the information on the risk level of the target monitoring area and the information on the detection range of the target monitoring area.

[0136] As an illustration, the central computing device can determine the validity of the risk level information and the detection range information of the target monitoring area based on information on abnormal areas of dam breaches, information on the degree of abnormality of dam breaches, information on the risk level of the target monitoring area, and information on the detection range of the target monitoring area.

[0137] Step S512: If the determination result of the validity assessment of the risk level information of the target monitoring area is that the risk level information of the target monitoring area is invalid and / or the determination result of the validity assessment of the detection range information of the target monitoring area is that the detection range information of the target monitoring area is invalid, generate emergency report information on the dam breach disaster.

[0138] Specifically, if the determination of the validity of the risk level information of the target monitoring area is that the risk level information of the target monitoring area is invalid and / or the determination of the validity of the detection range information of the target monitoring area is that the detection range information of the target monitoring area is invalid, the central computing device can generate emergency report information on the dam breach disaster. The emergency report information on the dam breach disaster can be used to characterize the fact that the results output by the artificial intelligence model for identifying the monitoring area do not match the actual disaster situation.

[0139] The aforementioned AI-based drone monitoring method for dam and levee breach scenarios reduces the frequency and workload of manual inspections, lowers labor costs and risks, and enables rapid action after anomalies are detected, thus shortening emergency response time.

[0140] In one alternative embodiment, please refer to Figure 5 The monitoring area identification AI model includes a first feature extraction module, a second feature extraction module, a feature fusion module, and a feature decoding module. Hydrological data, meteorological data, satellite imagery data, and topographic map data are input into the monitoring area identification AI model to identify target detection areas and generate target monitoring area risk level information and target monitoring area detection range information, including:

[0141] Step S501: Input hydrological data, meteorological data and topographic map data into the first feature extraction module to generate a meteorological and hydrological feature map and the meteorological and hydrological anomaly feature values ​​corresponding to the meteorological and hydrological feature map.

[0142] Specifically, the computing device can input hydrological data, meteorological data, and topographic map data into the first feature extraction module of the monitoring area identification artificial intelligence model built into the computing device, generating meteorological and hydrological feature maps and corresponding meteorological and hydrological anomaly feature values. The first feature extraction module can be used to spatially pair and fuse the hydrological data, meteorological data, and topographic map data.

[0143] Step S502: Input satellite image data and topographic map data into the second feature extraction module to generate satellite image feature map and satellite image anomaly feature value corresponding to the satellite image feature map.

[0144] Step S503: If the meteorological and hydrological anomaly feature value exceeds the preset meteorological and hydrological anomaly threshold or the satellite image anomaly feature value exceeds the preset satellite image anomaly threshold, the meteorological and hydrological feature map and the satellite image feature map are input into the feature fusion module to generate a fused feature map.

[0145] Step S504: Input the fused feature map into the feature decoding module to generate target monitoring area risk level information and target monitoring area detection range information.

[0146] In illustrative terms, the feature decoding module can be built based on, but is not limited to, the YOLOV8 model, the U-shaped network model, and the transformer model.

[0147] In the aforementioned AI-based UAV monitoring method for dam and levee breach scenarios, by setting up a first feature extraction module and a second feature extraction module to process meteorological and hydrological data and satellite imagery data respectively, key features can be quickly extracted and corresponding abnormal feature values ​​can be generated. When the abnormal feature values ​​exceed a preset threshold, the subsequent feature fusion and decoding process can be triggered in a timely manner. This not only improves computational efficiency but also enables real-time response to potential risks, providing timely and accurate basis for emergency decision-making. The feature fusion module can effectively integrate feature maps from different sources to generate a comprehensive fused feature map, further enhancing the model's adaptability to complex environments. The feature decoding module generates risk level and detection range information of the target monitoring area based on the fused feature map, which can effectively cope with complex and ever-changing natural environments and improve the generalization ability and robustness of the AI ​​model for monitoring area identification.

[0148] In one alternative embodiment, please refer to Figure 8 Control each drone to fly along a corresponding preset route and acquire target monitoring data for each drone, including:

[0149] Control each drone to fly along its corresponding preset flight path and acquire image data of the target monitoring sub-scene;

[0150] The target monitoring sub-scene image data is input into the image stitching edge artificial intelligence model to perform image stitching and generate target monitoring data.

[0151] Optionally, the image stitching edge AI model can be mounted on an edge computing device.

[0152] In the aforementioned AI-based drone monitoring method for dam and breach scenarios, the use of an AI model for image stitching edges can improve image stitching accuracy, enhance the robustness of the dam and breach detection system, and improve the end-to-cloud collaboration efficiency of the dam and breach detection system.

[0153] In one exemplary embodiment, such as Figure 5 As shown, another method for monitoring dam and levee breach scenarios using drones based on artificial intelligence is provided, including the following steps S501 to S512. Wherein:

[0154] Step S501: Input hydrological data, meteorological data and topographic map data into the first feature extraction module to generate a meteorological and hydrological feature map and the meteorological and hydrological anomaly feature values ​​corresponding to the meteorological and hydrological feature map.

[0155] Step S502: Input satellite image data and topographic map data into the second feature extraction module to generate satellite image feature map and satellite image anomaly feature value corresponding to the satellite image feature map.

[0156] Step S503: If the meteorological and hydrological anomaly feature value exceeds the preset meteorological and hydrological anomaly threshold or the satellite image anomaly feature value exceeds the preset satellite image anomaly threshold, the meteorological and hydrological feature map and the satellite image feature map are input into the feature fusion module to generate a fused feature map.

[0157] Step S504: Input the fused feature map into the feature decoding module to generate target monitoring area risk level information and target monitoring area detection range information.

[0158] Step S505: Allocate the number of drones and the preset flight path for each drone based on the risk level information and detection range information of the target monitoring area.

[0159] Step S506: Control each UAV to fly along the corresponding preset route and acquire target monitoring data for each UAV.

[0160] Step S507: Input the target monitoring data into the dam failure risk analysis sub-model to generate dam failure risk area information and dam failure risk information corresponding to the dam failure risk area information.

[0161] Step S508: If the risk information of dam breach and collapse is a dam breach and collapse disaster label, input the target monitoring data and topographic map data corresponding to the dam breach and collapse disaster label into the dam breach and collapse disaster prediction sub-model, perform disaster prediction analysis, and generate dam breach and collapse disaster information.

[0162] Step S509: Input the target monitoring data into the artificial intelligence model for identifying the edge of dam breach anomalies, perform dam breach anomaly identification, and generate dam breach identification marker information, dam breach anomaly area information, and dam breach anomaly degree information.

[0163] Step S510: If the dam breach identification information indicates that a dam breach exists, control the drone to hover and monitor the area corresponding to the abnormal dam breach area information.

[0164] Step S511: Based on the information on abnormal areas of dike and dam breaches, the information on the degree of abnormality of dike and dam breaches, the information on the risk level of the target monitoring area, and the information on the detection range of the target monitoring area, determine the validity of the information on the risk level of the target monitoring area and the information on the detection range of the target monitoring area.

[0165] Step S512: If the determination result of the validity assessment of the risk level information of the target monitoring area is that the risk level information of the target monitoring area is invalid and / or the determination result of the validity assessment of the detection range information of the target monitoring area is that the detection range information of the target monitoring area is invalid, generate emergency report information on the dam breach disaster.

[0166] The aforementioned AI-based drone monitoring method for dam and levee breach scenarios enables efficient and accurate monitoring of dam and levee breach scenarios through multi-source data fusion, efficient feature extraction and fusion, intelligent risk assessment and disaster assessment, and real-time anomaly identification and monitoring, thereby providing strong technical support for flood control and disaster reduction efforts.

[0167] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0168] Based on the same inventive concept, this application also provides an AI-based drone monitoring system for implementing the aforementioned AI-based drone monitoring method for dam breach scenarios. The solution provided by this system is similar to the implementation described in the above method. Therefore, the specific limitations of one or more AI-based drone monitoring system embodiments provided below can be found in the limitations of the AI-based drone monitoring method for dam breach scenarios described above, and will not be repeated here.

[0169] In one exemplary embodiment, such as Figure 9 As shown, an AI-based drone monitoring system 900 for dam and levee breach scenarios is provided, comprising:

[0170] The monitoring area data analysis module 901 can be used to acquire hydrological data, meteorological data, satellite imagery data, and topographic map data, and input the hydrological data, meteorological data, satellite imagery data, and topographic map data into the monitoring area identification artificial intelligence model to identify the target detection area and generate target monitoring area risk level information and target monitoring area detection range information.

[0171] The UAV configuration module 902 can be used to allocate the number of UAVs and the preset flight path of each UAV based on the risk level information and detection range information of the target monitoring area.

[0172] The UAV operation control module 903 can be used to control each UAV to fly along a corresponding preset route and acquire target monitoring data for each UAV.

[0173] The target monitoring data processing module 904 can be used to input the target monitoring data of each UAV into the artificial intelligence model for dam breach scenario analysis, perform dam breach scenario analysis, and generate dam breach scenario analysis results, which include dam breach risk information and / or dam breach disaster information.

[0174] In an optional embodiment, the UAV operational configuration module 902 can also be used for:

[0175] Set the maximum operating time of the drone and the shooting range of the airborne monitoring equipment based on the risk level information of the target monitoring area;

[0176] The maximum flight time of the UAV is calculated based on the maximum distance between the initial position of the UAV and the detection range of the target monitoring area. The maximum flight time is used to characterize the maximum time it takes for the UAV to travel from its initial position to the detection range of the target monitoring area.

[0177] The area of ​​the drone's preset operating area is calculated based on the maximum operating time, the shooting range of the airborne monitoring equipment, and the maximum flight time.

[0178] The number of drones is allocated based on the preset operating area area and the target monitoring area detection range information;

[0179] Each drone is assigned a preset flight path based on the number of drones, the risk level information of the target monitoring area, and the detection range information of the target monitoring area.

[0180] In an optional embodiment, the AI-based drone monitoring system 900 for dam and levee breach scenarios can also be used for:

[0181] Calculate the risk level error between the risk level information of the target monitoring area and the risk level information of the second target monitoring area, and determine whether the risk level error exceeds the risk level deviation threshold;

[0182] Calculate the detection range error between the detection range information of the target monitoring area and the detection range information of the second target monitoring area, and determine whether the detection range error exceeds the detection range deviation threshold;

[0183] If the risk level error exceeds the risk level deviation threshold and / or the detection range error exceeds the detection range deviation threshold, the number of drones is adjusted based on the risk level information and detection range information of the second target monitoring area, and the preset flight path of each drone is updated based on the adjusted number of drones, the risk level information of the target monitoring area, the detection range information of the target monitoring area, and the monitored area information of the drones; wherein, the monitored area information is generated based on the target monitoring data of the drones;

[0184] The monitoring area identification artificial intelligence model is adjusted and optimized based on the risk level information of the target monitoring area, the detection range information of the target monitoring area, the risk level information of the second target monitoring area, and the detection range information of the second target monitoring area.

[0185] In an optional embodiment, the target monitoring data processing module 904 can also be used for:

[0186] Input the target monitoring data into the dam breach risk analysis sub-model to generate dam breach risk area information and dam breach risk information corresponding to the dam breach risk area information. The dam breach risk level label includes the dam breach disaster label.

[0187] If the risk information of dam breach is a dam breach disaster label, the target monitoring data and topographic map data corresponding to the dam breach disaster label are input into the dam breach disaster prediction sub-model to perform disaster prediction analysis and generate dam breach disaster information.

[0188] In an optional embodiment, the AI-based drone monitoring system 900 for dam and levee breach scenarios can also be used for:

[0189] The target monitoring data is input into the artificial intelligence model for identifying the edge of dam breach anomalies, and dam breach anomalies are identified to generate dam breach identification marker information, dam breach anomaly area information, and dam breach anomaly degree information.

[0190] If the identification information for dam breaches indicates that a dam breach has occurred, control the drone to hover and monitor the area corresponding to the abnormal dam breach area information.

[0191] Based on information on abnormal areas of dam breaches, information on the degree of abnormality of dam breaches, information on the risk level of the target monitoring area, and information on the detection range of the target monitoring area, the validity of the risk level information of the target monitoring area and the validity of the detection range information of the target monitoring area are determined.

[0192] If the determination of the validity of the risk level information of the target monitoring area is that the risk level information of the target monitoring area is invalid and / or the determination of the validity of the detection range information of the target monitoring area is that the detection range information of the target monitoring area is invalid, an emergency report of dam breach disaster is generated. The emergency report of dam breach disaster is used to indicate that the results output by the monitoring area identification artificial intelligence model are inconsistent with the actual disaster situation.

[0193] In an optional embodiment, the monitoring area data analysis module 901 can also be used for:

[0194] Hydrological data, meteorological data, and topographic map data are input into the first feature extraction module to generate meteorological and hydrological feature maps and corresponding meteorological and hydrological anomaly feature values.

[0195] Satellite imagery data and topographic map data are input into the second feature extraction module to generate satellite imagery feature maps and corresponding satellite imagery anomaly feature values.

[0196] If the meteorological and hydrological anomaly characteristic value exceeds the preset meteorological and hydrological anomaly threshold or the satellite image anomaly characteristic value exceeds the preset satellite image anomaly threshold, the meteorological and hydrological feature map and the satellite image feature map are input into the feature fusion module to generate a fused feature map.

[0197] The fused feature map is input into the feature decoding module to generate risk level information and detection range information of the target monitoring area.

[0198] In an optional embodiment, the UAV operation control module 903 can also be used for:

[0199] Control each drone to fly along its corresponding preset flight path and acquire image data of the target monitoring sub-scene;

[0200] The target monitoring sub-scene image data is input into the image stitching edge artificial intelligence model to perform image stitching and generate target monitoring data.

[0201] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the aforementioned artificial intelligence-based drone monitoring method for dam and breach scenarios.

[0202] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0203] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0204] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A drone-based monitoring method for dam and levee breach scenarios based on artificial intelligence, characterized in that, The method includes: Acquire hydrological data, meteorological data, satellite imagery data, and topographic map data, and input the hydrological data, meteorological data, satellite imagery data, and topographic map data into the monitoring area identification artificial intelligence model to identify the target detection area and generate target monitoring area risk level information and target monitoring area detection range information; The number of drones and the preset flight path of each drone are allocated based on the risk level information of the target monitoring area and the detection range information of the target monitoring area. Control each of the drones to fly along the corresponding preset route and acquire target monitoring data for each drone; The target monitoring data of each of the aforementioned UAVs is input into the artificial intelligence model for analyzing dam breach scenarios, and dam breach scenario analysis is performed to generate dam breach scenario analysis results, which include dam breach risk information and / or dam breach disaster information. The allocation of the number of drones and the preset flight path for each drone based on the risk level information and detection range information of the target monitoring area includes: The maximum operating time of the UAV and the shooting range of the airborne monitoring equipment are set based on the risk level information of the target monitoring area. The maximum flight time of the UAV is calculated based on the maximum distance between the initial position of the UAV and the detection range of the target monitoring area. The maximum flight time is used to characterize the maximum time it takes for the UAV to travel from the initial position to the detection range of the target monitoring area. The preset operating area of ​​the UAV is calculated based on the maximum operating time, the shooting range of the airborne monitoring equipment, and the maximum flight time. The number of UAVs is allocated based on the preset operating area area and the target monitoring area detection range information; The preset flight path for each drone is assigned based on the number of drones, the risk level information of the target monitoring area, and the detection range information of the target monitoring area.

2. The method according to claim 1, characterized in that, The formula for calculating the number of drones is: ； ； ； In the formula, The number of the drones. It is a rounding function. The total area of ​​the detection range of the target monitoring area. This is the correction factor for overlap of operational flight paths. For the maximum operation time, For the maximum sailing time, For environmental conditions speed correction factor, The average flight speed of the drone. The width of the field of view captured by the airborne monitoring equipment. The fuzzy control membership function for the maximum operation time is... The risk level of the target monitoring area, The fuzzy control membership function is defined for the shooting range of the airborne monitoring equipment.

3. The method according to claim 1, characterized in that, The control of each UAV to fly along the corresponding preset route and to acquire target monitoring data for each UAV includes: Control each of the aforementioned drones to fly along the corresponding preset flight path and acquire target monitoring sub-scene image data; The target monitoring sub-scene image data is input into the image stitching edge artificial intelligence model to perform image stitching and generate the target monitoring data.

4. The method according to claim 3, characterized in that, The analysis results of the dam breach scenario include risk level information of the second target monitoring area and detection range information of the second target monitoring area. The method also includes: Calculate the risk level error between the risk level information of the target monitoring area and the risk level information of the second target monitoring area, and determine whether the risk level error exceeds the risk level deviation threshold; Calculate the detection range error between the detection range information of the target monitoring area and the detection range information of the second target monitoring area, and determine whether the detection range error exceeds the detection range deviation threshold; If the risk level error exceeds the risk level deviation threshold and / or the detection range error exceeds the detection range deviation threshold, the number of drones is adjusted based on the risk level information and detection range information of the second target monitoring area, and the preset flight path of each drone is updated based on the adjusted number of drones, the risk level information of the target monitoring area, the detection range information of the target monitoring area, and the monitored area information of the drones; wherein, the monitored area information is generated based on the target monitoring data of the drones; The monitoring area identification artificial intelligence model is adjusted and optimized based on the risk level information of the target monitoring area, the detection range information of the target monitoring area, the risk level information of the second target monitoring area, and the detection range information of the second target monitoring area.

5. The method according to claim 3, characterized in that, The method further includes: The target monitoring data is input into the edge artificial intelligence model for identifying dam breach anomalies, and dam breach anomalies are identified to generate dam breach identification marker information, dam breach anomaly area information, and dam breach anomaly degree information. If the dam breach identification information indicates that a dam breach has occurred, control the drone to hover and monitor the area corresponding to the abnormal dam breach area information. Based on the information on the abnormal area of ​​the dam breach, the information on the degree of abnormality of the dam breach, the information on the risk level of the target monitoring area, and the information on the detection range of the target monitoring area, the validity of the information on the risk level of the target monitoring area and the validity of the information on the detection range of the target monitoring area are determined. If the determination result of the validity assessment of the risk level information of the target monitoring area is that the risk level information of the target monitoring area is invalid and / or the determination result of the validity assessment of the detection range information of the target monitoring area is that the detection range information of the target monitoring area is invalid, an emergency report of the dike breach disaster is generated. The emergency report of the dike breach disaster is used to indicate that the result output by the artificial intelligence model for identifying the monitoring area is inconsistent with the actual disaster situation.

6. The method according to claim 1, characterized in that, The artificial intelligence model for analyzing dam breach scenarios includes a dam breach risk analysis sub-model and a dam breach disaster prediction sub-model. The dam breach scenario analysis results include dam breach risk information and dam breach disaster information. The step involves inputting the target monitoring data from each of the drones into the dam breach scenario analysis artificial intelligence model to perform dam breach scenario analysis and generate dam breach scenario analysis results, including: The target monitoring data is input into the dam breach risk analysis sub-model to generate dam breach risk area information and dam breach risk information corresponding to the dam breach risk area information. The dam breach risk information includes dam breach disaster labels. If the risk information of dam breach is the disaster label of dam breach, the target monitoring data and the topographic map data corresponding to the disaster label of dam breach are input into the dam breach disaster prediction sub-model to perform disaster prediction analysis and generate the dam breach disaster information.

7. The method according to claim 6, characterized in that: The sub-model for analyzing the risk of dam and dike breaches is a deep learning neural network model built based on an improved YOLOV8 model. The sub-model for analyzing dam and dike breach disasters is a deep learning neural network model built based on a U-shaped network combined with a converter module.

8. The method according to claim 1, characterized in that, The monitoring area identification artificial intelligence model includes a first feature extraction module, a second feature extraction module, a feature fusion module, and a feature decoding module. The hydrological data, meteorological data, satellite imagery data, and topographic map data are input into the monitoring area identification artificial intelligence model to identify the target detection area and generate target monitoring area risk level information and target monitoring area detection range information, including: The hydrological data, meteorological data, and topographic map data are input into the first feature extraction module to generate a meteorological and hydrological feature map and meteorological and hydrological anomaly feature values ​​corresponding to the generated meteorological and hydrological feature map; The satellite image data and the topographic map data are input into the second feature extraction module to generate a satellite image feature map and satellite image anomaly feature values ​​corresponding to the satellite image feature map; If the meteorological and hydrological anomaly feature value exceeds the preset meteorological and hydrological anomaly threshold or the satellite image anomaly feature value exceeds the preset satellite image anomaly threshold, the meteorological and hydrological feature map and the satellite image feature map are input into the feature fusion module to generate a fused feature map. The fused feature map is input into the feature decoding module to generate the risk level information of the target monitoring area and the detection range information of the target monitoring area.

9. An AI-based unmanned aerial vehicle (UAV) monitoring system for dam and dike breach scenarios, characterized in that: The system includes various functional modules required to implement the method of any one of claims 1 to 8.

Images