Disaster risk dynamic analysis system based on unmanned aerial vehicle data

Abstract

The invention discloses a disaster risk dynamic analysis system based on unmanned aerial vehicle data. The disaster risk dynamic analysis system comprises an unmanned aerial vehicle platform, a data transmission module and a ground control center. The unmanned aerial vehicle platform comprises a data acquisition module, and the data acquisition module is used for acquiring disaster field data; the data transmission module is used for accurately transmitting data in real time and has a data encryption function to ensure data security; the ground control center comprises a data arrangement unit, a multi-source data fusion unit, a disaster prediction unit, an early warning unit and a display unit. According to the invention, various data of a disaster area can be rapidly collected through the unmanned aerial vehicle, the uneven data can be systematically processed through the data integration unit, and then the disaster can be analyzed and predicted through the prediction model constructed by the ground control center. Therefore, the disaster development prediction accuracy and efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of disaster risk analysis, and in particular to a disaster risk dynamic analysis system based on drone data. Background Art

[0002] Natural disasters refer to sudden events caused by natural factors that cause serious damage to human life, property and living environment. They are characterized by being uncontrollable, highly destructive and having a wide range of impacts.

[0003] Natural disasters such as floods, earthquakes, mudslides, and forest fires frequently wreak havoc, posing a significant threat to human life and property. Timely and accurate monitoring and early warning of natural disaster trends are crucial for developing effective disaster prevention and mitigation measures.

[0004] Traditional disaster monitoring methods rely primarily on fixed equipment, manual inspections, and specific technical platforms. These methods are limited in scope, lack of data timeliness, and difficulty integrating multi-source data when responding to disasters. They also hinder the flexible and rapid collection of on-site data for rapid risk analysis using drones.

[0005] Therefore, we propose a disaster risk dynamic analysis system based on drone data. Summary of the Invention

[0006] The purpose of the present invention is to provide a disaster risk dynamic analysis system based on drone data, which solves the problems raised in the background technology.

[0007] To achieve the above-mentioned object, the present invention provides the following technical solutions: a disaster risk dynamic analysis system based on drone data, comprising the following method steps: a drone platform, a data transmission module, and a ground control center;

[0008] The UAV platform includes a data acquisition module, which is used to collect disaster site data;

[0009] The data transmission module is used to transmit data in real time and accurately, and has a data encryption function to ensure data security;

[0010] The ground control center includes a data collation unit, a multi-source data fusion unit, a disaster prediction unit, an early warning unit, and a display unit;

[0011] The data sorting unit is used to process the received uneven data, including preliminary screening to remove obvious errors, serious missing or abnormal data points, classifying the data by source, type and accuracy, and converting data in different formats into a unified format;

[0012] The multi-source data fusion unit fuses the sorted data based on a multi-source data fusion algorithm based on deep learning. It first performs denoising, normalization, and feature extraction preprocessing on the data, uses a deep learning model to learn the association between different data sources to achieve fusion, and uses parallel computing technology to distribute the fusion task to multiple computing nodes. At the same time, it introduces a data quality assessment mechanism to monitor the data quality in real time during the fusion process and perform special processing or eliminate low-quality data.

[0013] The disaster prediction unit is used to process the fused high-precision data and quickly and accurately predict the future development trend of disasters;

[0014] The warning unit is used to automatically issue warning information containing detailed information such as warning level, disaster type, impact range, and expected occurrence time when the disaster risk reaches the warning threshold;

[0015] The display unit is used to display the disaster scene picture.

[0016] As a preferred embodiment of the present invention, the ground control center also includes a data receiving and storage unit, which is used to receive data transmitted by the drone and store it in a high-performance distributed database, using a distributed storage architecture to improve data reading and writing speed and storage capacity.

[0017] As a preferred embodiment of the present invention, the data sorting unit adopts a method based on statistical analysis and threshold judgment to remove obviously erroneous, seriously missing or abnormal data points; when classifying data, data of different sources, types and precisions are classified according to preset classification rules; in the format conversion process, a conversion algorithm is used to achieve unified format conversion for data of different formats; when supplementing missing data, a Kriging interpolation algorithm is used to supplement the data based on the spatial and temporal correlation of the data.

[0018] As a preferred embodiment of the present invention, the multi-source data fusion unit adopts advanced feature selection algorithms when performing feature extraction, including feature selection methods based on information gain, chi-square test, and mutual information, to screen out the most effective features for disaster analysis; when training the deep learning model, a convolutional neural network model structure is adopted to learn the complex correlation relationships between different data sources.

[0019] As a preferred embodiment of the present invention, the disaster prediction unit uses a numerical simulation method to establish a mathematical model based on the physical mechanism of the disaster based on the fused high-precision data, and inputs the fused data to simulate the development trend of the disaster; and uses a machine learning algorithm to train the prediction model using historical disaster data and current monitoring data to achieve rapid and accurate prediction of future development trends of disasters.

[0020] As a preferred embodiment of the present invention, the early warning unit further includes a notification module, and the notification module is used to notify relevant departments and personnel.

[0021] As a preferred embodiment of the present invention, the data acquisition module includes a five-spectral camera, a millimeter-wave radar, and a thermal infrared sensor. The five-spectral camera is used to collect spectral data in different bands to analyze vegetation coverage, soil moisture, water distribution, etc. The millimeter-wave radar is used to obtain key information such as the topography and surface deformation of the target area. The thermal infrared sensor is used to monitor the distribution of hotspots of forest fires.

[0022] As a preferred embodiment of the present invention, the display unit is combined with a geographic information system to display the disaster development situation in the form of an intuitive map.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] The present invention can quickly collect various data from disaster areas through drones, and can systematically process the uneven data through a data integration unit. Then, the prediction model built by the ground control center can analyze and predict the disaster, thereby improving the accuracy and efficiency of disaster development prediction.

[0025] The use of multivariate data fusion algorithms can enable the rapid and accurate fusion of various data, which is conducive to avoiding inaccurate fusion caused by large differences in data quality, thereby achieving rapid and accurate data fusion processing, which is conducive to further improving the accuracy of disaster line prediction and the efficiency of disaster prediction;

[0026] By using drones equipped with various data collection equipment to conduct on-site data collection, we can obtain more comprehensive information about the disaster area. At the same time, combined with the geographic information system, the disaster site can be visualized and processed, providing further support for emergency decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Other features, objects and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments with reference to the following drawings:

[0028] Figure 1 This is an operation diagram of a disaster risk dynamic analysis system based on drone data according to the present invention. DETAILED DESCRIPTION

[0029] In order to make the technical means, creative features, objectives and effects achieved by the present invention easier to understand, the present invention is further described below in conjunction with specific implementation methods.

[0030] like Figure 1As shown, the present invention proposes a disaster risk dynamic analysis system based on drone data, including the following method steps: a drone platform, a data transmission module, and a ground control center;

[0031] The UAV platform includes a data acquisition module, which is used to collect disaster site data.

[0032] The data acquisition module includes a five-spectral camera, a millimeter-wave radar, and a thermal infrared sensor. The five-spectral camera is used to collect spectral data in different bands to analyze vegetation coverage, soil moisture, water distribution, etc. The millimeter-wave radar is used to obtain key information such as the topography and surface deformation of the target area. The thermal infrared sensor is used to monitor the distribution of forest fire hotspots.

[0033] The data transmission module is used to transmit data in real time and accurately, and has a data encryption function to ensure data security.

[0034] The ground control center includes a data collation unit, a multi-source data fusion unit, a disaster prediction unit, an early warning unit, and a display unit;

[0035] The data sorting unit uses a method based on statistical analysis and threshold judgment to remove data points with obvious errors, serious missing data or abnormal data points; when classifying data, it classifies data of different sources, types and precisions according to preset classification rules; during the format conversion process, a conversion algorithm is used to achieve unified format conversion for data of different formats; when supplementing missing data, the Kriging interpolation algorithm is used to supplement the data based on the spatial and temporal correlation of the data.

[0036] The multi-source data fusion unit fuses the sorted data based on a multi-source data fusion algorithm based on deep learning; first, the data is pre-processed by denoising, normalization, and feature extraction, and the deep learning model is used to learn the correlation between different data sources to achieve fusion, and parallel computing technology is used to distribute the fusion task to multiple computing nodes; at the same time, a data quality assessment mechanism is introduced to monitor the data quality in the fusion process in real time, and low-quality data is specially processed or eliminated. When extracting features, advanced feature selection algorithms are used, including feature selection methods based on information gain, chi-square test, and mutual information, to screen out the most effective features for disaster analysis; when training the deep learning model, a convolutional neural network model structure is used to learn the complex correlation between different data sources.

[0037] The disaster prediction unit uses the fused high-precision data, applies numerical simulation methods to establish a mathematical model according to the physical mechanism of the disaster, and inputs the fused data to simulate the development trend of the disaster; uses the machine learning algorithm to train the prediction model using historical disaster data and current monitoring data to achieve rapid and accurate prediction of the future development trend of the disaster.

[0038] The early warning unit is used to automatically issue early warning information containing detailed content such as warning level, disaster type, impact range, expected time of occurrence, etc. when the disaster risk detected reaches the early warning threshold. It also includes a notification module, which is used to notify relevant departments and personnel.

[0039] The display unit is combined with a geographic information system to display the disaster development situation in the form of an intuitive map.

[0040] The ground control center also includes a data receiving and storage unit, which is used to receive data transmitted by the drone and store it in a high-performance distributed database, using a distributed storage architecture to improve data reading and writing speed and storage capacity.

[0041] Example 1

[0042] Dynamic analysis and early warning of forest fire risks

[0043] When a forest fire occurs, the drone is equipped with a five-spectral camera, thermal infrared sensor and millimeter-wave radar to collect mountain data. Then, satellite communication technology is used to ensure real-time and accurate data transmission in complex terrain. The data is encrypted with the RSA encryption algorithm to ensure data security and fed back to the ground control center. The data sorting unit performs preliminary screening, classification and format conversion on the received data, removes obvious errors, serious missing or abnormal data points, classifies the data by source, type and accuracy, and converts it into a unified format; then the multi-source data fusion unit denoises and normalizes the sorted data, uses advanced feature selection algorithms to extract features, and uses convolutional neural network models to extract features. It learns the correlation between different data sources, realizes data fusion, and monitors the data quality in the fusion process in real time. It performs special processing or eliminates low-quality data. Then, the disaster prediction unit uses the high-precision data after fusion to establish a forest fire mathematical model using numerical simulation methods, and inputs the fused data to simulate the fire development trend. At the same time, it uses the random forest algorithm to train the prediction model using historical fire data and current monitoring data to achieve rapid and accurate prediction of future fire development trends. When the fire risk reaches the warning threshold, the warning unit automatically issues a warning message and notifies relevant departments and personnel through the notification module. The display unit combines the geographic information system to intuitively display the fire scene, development status, etc. in the form of a map, making it convenient for rescue personnel to formulate rescue plans.

[0044] Example 2

[0045] Dynamic analysis and early warning of landslide disaster risks

[0046] When a landslide occurs, the drone flies over the mountain, and the five-spectral camera, millimeter radar and thermal infrared sensor collect mountain-related data in real time and transmit the data to the data transmission module. The data transmission module transmits the encrypted data to the data receiving and storage unit of the ground control center via satellite communication and stores it in the distributed file system. The data sorting unit then performs a preliminary screening of the received data, removes obvious errors, serious omissions or abnormal data points, and classifies data of different sources, types and accuracy according to preset classification rules. It also uses conversion algorithms to achieve unified format conversion for data in different formats, and then uses the multi-source data fusion unit to denoise and normalize the sorted data, and uses advanced feature selection algorithms to screen out features that are relevant to landslide classification. The most effective features are analyzed, and the convolutional neural network model is used to learn the complex correlation between different data sources to achieve data fusion and monitor data quality in real time. The disaster prediction unit uses the numerical simulation method to establish a mathematical model of landslide based on the fused high-precision data, and inputs the fused data to simulate the development trend of the landslide. At the same time, the support vector machine algorithm is used to train the prediction model using historical landslide data and current monitoring data to achieve rapid and accurate prediction of the future development trend of the landslide. When the landslide risk reaches the warning threshold, the warning unit automatically issues a warning message and notifies the relevant departments and personnel through the notification module. The display unit combines with the geographic information system to intuitively display the landslide scene and development status in the form of a map to provide decision support for rescue personnel.

[0047] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention. Therefore, the embodiments should be regarded as illustrative and non-restrictive in all respects. The scope of the present invention is defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the claims are intended to be included therein.

[0048] In addition, it should be understood that although this specification is described in terms of implementation methods, not every implementation method contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

Claims

1. A disaster risk dynamic analysis system based on drone data, characterized by: The method comprises the following steps: a UAV platform, a data transmission module, and a ground control center; The UAV platform includes a data acquisition module, which is used to collect disaster site data; The data transmission module is used to transmit data in real time and accurately, and has a data encryption function to ensure data security; The ground control center includes a data collation unit, a multi-source data fusion unit, a disaster prediction unit, an early warning unit, and a display unit; The data sorting unit is used to process the received uneven data, including preliminary screening to remove obvious errors, serious missing or abnormal data points, classifying the data by source, type and accuracy, and converting data in different formats into a unified format; The multi-source data fusion unit fuses the sorted data based on a multi-source data fusion algorithm based on deep learning. It first performs denoising, normalization, and feature extraction preprocessing on the data, uses a deep learning model to learn the association between different data sources to achieve fusion, and uses parallel computing technology to distribute the fusion task to multiple computing nodes. At the same time, it introduces a data quality assessment mechanism to monitor the data quality in real time during the fusion process and perform special processing or eliminate low-quality data. The disaster prediction unit is used to process the fused high-precision data and quickly and accurately predict the future development trend of disasters; The warning unit is used to automatically issue warning information containing detailed information such as warning level, disaster type, impact range, and expected occurrence time when the disaster risk reaches the warning threshold; The display unit is used to display the disaster scene picture.

2. The disaster risk dynamic analysis system based on drone data according to claim 1, characterized in that: The ground control center also includes a data receiving and storage unit, which is used to receive data transmitted by the drone and store it in a high-performance distributed database, using a distributed storage architecture to improve data reading and writing speed and storage capacity.

3. The disaster risk dynamic analysis system based on drone data according to claim 1, characterized in that: The data sorting unit uses a method based on statistical analysis and threshold judgment to remove data points with obvious errors, serious missing data or abnormal data points; when classifying data, it classifies data of different sources, types and precisions according to preset classification rules; during the format conversion process, a conversion algorithm is used to achieve unified format conversion for data of different formats; when supplementing missing data, the Kriging interpolation algorithm is used to supplement the data based on the spatial and temporal correlation of the data.

4. The disaster risk dynamic analysis system based on drone data according to claim 1, characterized in that: When performing feature extraction, the multi-source data fusion unit adopts advanced feature selection algorithms, including feature selection methods based on information gain, chi-square test, and mutual information, to screen out the most effective features for disaster analysis; when training the deep learning model, the convolutional neural network model structure is adopted to learn the complex correlation relationships between different data sources.

5. The disaster risk dynamic analysis system based on drone data according to claim 1, characterized in that: The disaster prediction unit uses the fused high-precision data to establish a mathematical model based on the physical mechanism of the disaster using a numerical simulation method, and inputs the fused data to simulate the development trend of the disaster; Machine learning algorithms are used to train prediction models using historical disaster data and current monitoring data to achieve rapid and accurate predictions of future disaster trends.

6. The disaster risk dynamic analysis system based on drone data according to claim 1, characterized in that: The early warning unit further comprises a notification module, and the notification module is used to notify relevant departments and personnel.

7. The disaster risk dynamic analysis system based on drone data according to claim 1, characterized in that: The data acquisition module includes a five-spectral camera, a millimeter-wave radar, and a thermal infrared sensor. The five-spectral camera is used to collect spectral data in different bands to analyze vegetation coverage, soil moisture, water distribution, etc. The millimeter-wave radar is used to obtain key information such as the topography and surface deformation of the target area. The thermal infrared sensor is used to monitor the distribution of forest fire hotspots.

8. The disaster risk dynamic analysis system based on drone data according to claim 1, characterized in that: The display unit is combined with a geographic information system to display the disaster development situation in the form of an intuitive map.

Images