Real-time geographic information monitoring system and method thereof

Abstract

The invention discloses a real-time geographic information monitoring system and a real-time geographic information monitoring method, and relates to the technical field of geographic information. The geographic information monitoring platform is in communication connection with an image data acquisition module, a weather monitoring and correction module, an image matching and positioning module, a three-dimensional reconstruction and model generation module, a real-time dynamic monitoring module and an intelligent sensing early warning module, and all the modules are in electric signal connection. According to the invention, through integration of various data sources and a preprocessing technology, geographic space data is rapidly collected and processed, the geographic space data comprises various aspects of terrains, vegetation, buildings, meteorology and the like, spatial analysis and feature extraction are carried out through a GIS technology, the real-time state of a geographic environment can be accurately reflected, and compared with a traditional environment monitoring means, the method has the advantages of high efficiency and high reliability. The real-time monitoring system greatly shortens the data updating period and provides more timely and accurate environment information.

Description

Technical Field

[0001] The present invention relates to the field of geographic information technology, and in particular to a real-time geographic information monitoring system and method thereof. Background Art

[0002] As a system for collecting, storing, managing, analyzing and displaying geographic spatial data, Geographic Information System (GIS) plays an increasingly important role in modern society. In order to realize real-time monitoring and dynamic updating of geographic information, the geographic information system needs an efficient and accurate means of acquiring geographic information. Photogrammetry technology can meet the needs of geographic information monitoring. In addition, photogrammetry, as a method of obtaining the spatial position and shape of objects by measuring the propagation and reflection of light in space, has gone through three development stages: simulation, analysis and digitization. With the rapid development of computers and computing technology, digital photogrammetry and real-time photogrammetry technology have emerged, providing strong technical support for real-time monitoring of geographic information.

[0003] Since photogrammetry is highly dependent on weather conditions, it will be seriously affected in bad weather, making it difficult to obtain clear images, which in turn affects the accuracy of geographic information monitoring. Therefore, how to weaken the impact of weather factors on photographic images and thus improve measurement accuracy is the problem we need to solve. To this end, a real-time geographic information monitoring system and method are proposed. Summary of the invention

[0004] The present invention aims to provide a real-time geographic information monitoring system and method thereof to solve the problems raised in the above-mentioned background technology.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: In a first aspect, a real-time geographic information monitoring system includes a geographic information monitoring platform, wherein the geographic information monitoring platform is communicatively connected to an image data acquisition module, a weather monitoring and correction module, an image matching and positioning module, a three-dimensional reconstruction and model generation module, a real-time dynamic monitoring module, and an intelligent perception warning module, wherein electrical signals are connected between the modules; The image data acquisition module uses drone aerial photography to acquire geographic space image data, and pre-processes the acquired raw image data to improve the usability of the image and provide accurate data for subsequent image matching; The weather monitoring and correction module monitors weather conditions in real time and corrects the pre-processed image data according to weather changes, thereby reducing the impact of weather changes on image quality by dynamically correcting the acquired image data; The image matching and positioning module extracts feature points from the corrected image data and uses feature point matching technology to match and locate the images to establish spatial relationships, improve the accuracy of spatial measurement, and reduce positioning errors caused by weather changes; The 3D reconstruction and model generation module performs 3D reconstruction based on image matching and positioning results, generates a digital elevation model and a 3D model, provides 3D spatial information, and provides intuitive and accurate data support for geographic information monitoring; The real-time dynamic monitoring module is used to analyze the three-dimensional model in real time, monitor the changes in geographic spatial data, update the geographic information database, realize continuous monitoring of the geographical environment, and reduce monitoring gaps caused by weather changes; The intelligent perception warning module combines the analysis results of geographic spatial data and uses machine learning technology to monitor the data for further analysis, clarify abnormal changes in the geographic environment, and perform corresponding warning operations to improve the response speed of the system to ensure the accuracy of geographic information monitoring.

[0006] A further improvement of the technical solution of the present invention is that in the image data acquisition module, the image data acquisition and preprocessing process is as follows: Analyze the target area, formulate the flight plan of the drone according to the geographical location, topography, and weather conditions of the target area, and carry out route planning and flight preparation, including determining the shooting area, the starting and ending points of the route, and the flight altitude and speed of the drone, and checking the flight stability, endurance, and communication equipment of the drone to ensure flight safety and continuity of data collection; Carry out flight missions according to the established flight plan, use the camera on the drone to take aerial photos of the target area, collect geographic space image data in real time, and pay close attention to the flight status and data collection of the drone during the flight to ensure the smooth progress of the flight mission; The collected original image data is preprocessed by image grayscale and image enhancement. The component method is used to perform image grayscale processing operations, and the collected color image data is converted into grayscale images to simplify data processing and reduce storage space. The spatial or frequency domain filter is used to remove noise in the image, and the image enhancement operation is performed to improve the signal-to-noise ratio of the data; The preprocessed image data is cropped, the region of interest is selected, and the image data is formatted and converted into a format suitable for subsequent processing and analysis.

[0007] A further improvement of the technical solution of the present invention is that in the weather monitoring and correction module, the process of correcting the image data is: The weather data of the target area is collected in real time through the meteorological observation equipment of ground-based observation, including the temperature, humidity, wind speed, wind direction, and precipitation indicators of the target area. The meteorological observation equipment includes a temperature sensor, a humidity sensor, an anemometer, a wind vane, and a rain gauge; Compare the real-time collected weather data with the preset historical standard weather data, and compare the real-time image data with the historical standard image data. Combine the collected weather data to analyze the real-time image data in real time, preset the interference threshold, and calculate the image interference index to evaluate the degree of interference of weather conditions on the image data quality and determine whether it is within the allowable interference range. If the weather conditions are within the permissible interference range for the image data, the image data can be directly applied to the subsequent processing flow. If the weather conditions are not within the permissible interference range for the image data, the image data is corrected, and the correction process includes the steps of atmospheric correction, geometric correction and color correction. Perform quality assessment on the corrected image data to ensure that the correction effect meets the requirements. If the correction effect is not good, readjust the correction strategy and perform correction again until the required image data standard is met.

[0008] A further improvement of the technical solution of the present invention is that the calculation expression of the image interference index is: ; in, is the image interference index, It is the measured value of meteorological data of various meteorological indicators (temperature, humidity, wind speed, wind direction, precipitation). It is the historical standard meteorological data benchmark value of each meteorological index. and are the historical maximum and benchmark minimum values ​​of each meteorological index, is a weight factor, which indicates the influence of different meteorological indicators on image interference. , is the temperature weight, is the humidity weight, is the wind speed weight, is the wind direction weight, is the precipitation weight, The value range is between 0 and 1. The smaller the value, the less interference the weather conditions have on the image data, that is, the higher the image quality.

[0009] A further improvement of the technical solution of the present invention is that in the image matching and positioning module, the process of matching and positioning the image is as follows: The feature point detection algorithm (SIFT) is used to detect stable feature points from the rectified image, and a descriptor is generated for each detected feature point. The feature point is usually a corner point, edge, or area with significant texture in the image. It has a unique description in the image and is robust to rotation, scale change, and illumination change. The fast approximate nearest neighbor search matching algorithm is used to match the feature points in multiple images. The matching process determines which feature points are corresponding points based on the similarity of the descriptors, and matches the feature point descriptors in the image to be matched with the feature point descriptors in the reference image. During the matching process, the similarity between the descriptors is calculated to find the matching pair with the highest similarity. The fast approximate nearest neighbor search (FLANN) algorithm accelerates the matching process and improves the matching efficiency by building an index structure. After finding the matching pairs with the highest similarity, the Lowe's ratio test method is used to screen the feature points of the preliminary match. The matching pairs are screened by comparing the ratio of the nearest neighbor distance to the next nearest neighbor distance. Only the matching pairs with a ratio less than the set threshold are valid matching pairs, and the wrong matching pairs are removed. According to the matched feature point pairs, the geometric transformation model between the image to be matched and the reference image is estimated, the spatial relationship between the image to be matched and the reference image is analyzed, and the parameters of the geometric transformation model are optimized through an iterative optimization algorithm (RANSAC algorithm). The RANSAC algorithm estimates the most likely geometric transformation model from the feature point set containing the wrong matching pairs through multiple random sampling and verification, and removes the wrong matching pairs; After determining the geometric transformation model, the geometric transformation model is used to locate the image to be matched in the coordinate system of the reference image to achieve spatial alignment between the images. Through image positioning, the spatial relationship between the image to be matched and the reference image is established, which is helpful for subsequent spatial measurement and analysis.

[0010] A further improvement of the technical solution of the present invention is that in the three-dimensional reconstruction and model generation module, the generation process of the digital elevation model and the three-dimensional model is: Extract image data processed by the image matching and positioning module, including the corrected image, feature point matching results, and geometric transformation model; Using the feature point matching results and the geometric transformation model, the three-dimensional sparse point cloud between the images is reconstructed through the structure from motion (SFM) method, and the reconstructed sparse point cloud is optimized to improve its accuracy and stability. Based on the sparse point cloud, the multi-view stereo vision (MVS) method is used to generate a dense 3D point cloud through multi-view image matching and depth estimation. The generated dense point cloud is filtered and denoised to remove noise points and unnecessary objects to improve the density and accuracy of the point cloud. The dense point cloud is converted into a 3D mesh using the Poisson surface reconstruction meshing algorithm to form a continuous 3D surface, which provides a basis for subsequent texture mapping. The generated 3D mesh is optimized to improve its geometric accuracy and topological structure. The texture information of the original image is mapped to the 3D mesh through UV mapping technology to generate a new realistic 3D model. The reconstructed 3D point cloud is classified into ground points and non-ground points. The ground point cloud data is used to generate the elevation values ​​of regular grid points through the inverse distance weighted interpolation method, and then the interpolated elevation values ​​are output as a digital elevation model (DEM) file, such as GeoTIFF format. The generated 3D model is optimized to improve the geometric accuracy and texture quality of the model. The optimization process includes mesh simplification, texture repair, and lighting adjustment steps, and the optimized 3D model is exported.

[0011] A further improvement of the technical solution of the present invention is that in the real-time dynamic monitoring module, the process of updating the geographic information database is: Access the 3D model data processed by the 3D reconstruction and model generation module, and access the real-time collected image data and weather data to provide more comprehensive monitoring information. The 3D model includes key geographical elements such as terrain, buildings, and vegetation. Match the real-time collected image data with the 3D model to ensure data accuracy and consistency, and then perform real-time change detection on the 3D model. By comparing the image data and 3D model at different time points, the changed areas in the model can be identified and marked, including terrain undulations, the addition or removal of buildings, and the growth or decline of vegetation; For the changed areas, key attribute information such as height, area and volume is extracted, and the attribute information is verified and validated by comparing with historical data to ensure its accuracy and reliability; Update the detected change information and extracted attribute information into the geographic information database, obtain geospatial data, ensure that the geospatial data in the database is consistent with the actual geographic environment, perform quality control and verification on the updated data, and correct or re-collect data that does not meet the requirements; Through three-dimensional visualization technology, the monitoring results are displayed intuitively, and a detailed monitoring report is generated based on the monitoring results. The report content includes monitoring time, monitoring area, change type (topography undulation, addition or demolition of buildings, growth or decline of vegetation), change amount (height change, area change, volume change, etc.), and attribute information (key attributes such as height, area, volume, etc.).

[0012] A further improvement of the technical solution of the present invention is that in the intelligent perception warning module, the analysis process of abnormal changes in the geographical environment is: Use geographic information systems to perform spatial analysis on pre-processed geospatial data, including spatial interpolation, spatial overlay, and buffer zone analysis, to extract the spatial characteristics and laws of the geographical environment, combine historical data with geospatial data obtained through real-time monitoring, analyze the dynamic change trend of the geographical environment, and mark abnormal change data in geospatial data. Among them, through spatial interpolation, the data of known points are used to infer the data of unknown points to generate more complete geospatial data. Through spatial overlay, geospatial data from different sources and different themes are superimposed to reveal the spatial relationship and interaction between them. Through buffer zone analysis, a buffer zone of a certain width is created around a specific geographical element to analyze its impact on the surrounding environment. By combining historical data and annotated abnormal change data with a neural network model, the model can learn the normal state and abnormal change patterns of the geographical environment and train the abnormal change model of the geographical environment; Based on the output results of the geographic environment abnormal change model, the geographic environment alienation coefficient is calculated to quantify the degree of deviation between the geographic environment and the normal state; Preset geographic environment deviation threshold , the calculated geographical environment alienation coefficient is compared with the preset deviation threshold. When the geographical environment alienation coefficient exceeds the threshold, it is considered that the geographical environment has undergone abnormal changes; When an abnormal change is confirmed, the early warning mechanism is triggered and an early warning message is sent to the user, including detailed information on the location, type, and degree of the abnormal change. The abnormal change data is saved in the database, and corresponding response measures are taken based on the early warning information, such as activating emergency plans, dispatching personnel for on-site verification, etc.

[0013] A further improvement of the technical solution of the present invention is that the calculation expression of the geographical environment alienation coefficient is: ; in, is the geographical environment alienation coefficient, is the value of the geospatial data acquired through real-time monitoring. is the reference value of the corresponding position in the historical data, and All The maximum and minimum values ​​in the value, is the sample size, is the exponent in the exponential function, a positive real number, used to adjust the curve shape of the exponential function. Represents the sum of squared deviations of all sample points, after normalization. The exponential decay term representing the sample point deviation is used to capture the nonlinear deviation of the geographical environment. The value range is between 0 and 1. The smaller the value, the smaller the deviation of the geographical environment from the historical normal state, that is, the more stable the geographical environment.

[0014] In a second aspect, a real-time geographic information monitoring method is implemented based on a real-time geographic information monitoring system, comprising the following steps: Step 1, obtaining geographic space image data by drone or aerial photography, and preprocessing and correcting the image data; Step 2: extract key feature points from the preprocessed data, match and locate the images, and perform three-dimensional reconstruction to generate a digital elevation model and a three-dimensional model; Step 3: Use geographic information system to perform spatial analysis on the preprocessed data, and combine historical data and real-time monitoring data to analyze the dynamic change trend of the geographical environment and mark abnormal change data in the geographic spatial data; Step 4: Compare data at different time points to identify areas of terrain, buildings, and vegetation changes, use historical data and abnormal change data to train the neural network model, and learn normal states and abnormal change patterns; Step 5, calculating the geographical environment alienation coefficient based on the trained model, quantifying the degree of deviation, and comparing the calculated geographical environment alienation coefficient with the preset deviation threshold to determine whether an abnormal change occurs; Step 6: When the geographical environment alienation coefficient exceeds the deviation threshold, the early warning mechanism is triggered, and an early warning message is sent to the user, and corresponding response measures are taken according to the early warning message.

[0015] Due to the adoption of the above technical solution, the present invention has the following technical advances compared with the prior art: 1. The present invention provides a real-time geographic information monitoring system and method thereof. By integrating multiple data sources and preprocessing technologies, the system can quickly collect and process geographic spatial data, including terrain, vegetation, buildings, meteorology and other aspects. By performing spatial analysis and feature extraction through GIS technology, the system can accurately reflect the real-time status of the geographic environment. Compared with traditional environmental monitoring methods, the real-time monitoring system greatly shortens the data update cycle and provides more timely and accurate environmental information.

[0016] 2. The present invention provides a real-time geographic information monitoring system and method thereof. By utilizing a machine learning model, it is capable of learning the normal state and abnormal change pattern of the geographic environment, automatically performing anomaly detection on real-time monitoring data, and promptly discovering deviations from the normal state. When the geographic environment alienation coefficient exceeds a preset deviation threshold, the system will trigger an early warning mechanism and send detailed early warning information to the user, which not only improves the accuracy and timeliness of the early warning, but also helps relevant departments to quickly take countermeasures. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0018] Figure 1 It is a module structure diagram of the present invention; Figure 2 A flow chart for correcting image data according to the present invention; Figure 3 The flowchart of the analysis of abnormal changes in geographical environment of the present invention; Figure 4 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Embodiment 1, as Figure 1 , Figure 2 As shown, the present invention provides a real-time geographic information monitoring system and method thereof, including a geographic information monitoring platform, wherein the geographic information monitoring platform is communicatively connected with an image data acquisition module, a weather monitoring and correction module, an image matching and positioning module, a three-dimensional reconstruction and model generation module, a real-time dynamic monitoring module, and an intelligent perception and early warning module, wherein the modules are connected by electrical signals; The image data acquisition module uses drone aerial photography to obtain geographic space image data, and pre-processes the collected raw image data to improve the usability of the image, provide accurate data for subsequent image matching, analyze the target area, and formulate the flight plan of the drone according to the geographical location, topography, and weather conditions of the target area. It also conducts route planning and flight preparation, including determining the shooting area, the starting and ending points of the route, and the flight altitude and speed of the drone, and checks the flight stability, endurance, and communication equipment of the drone to ensure flight safety and continuity of data collection. It executes the flight mission according to the formulated flight plan, and uses the camera on the drone to image the target area. Aerial photography is carried out in the field to collect image data of geographic space in real time. During the flight, the flight status and data collection of the UAV are closely monitored to ensure the smooth progress of the flight mission. The collected raw image data are preprocessed by image grayscale and image enhancement. The component method is used to perform image grayscale processing operations, and the collected color image data is converted into grayscale images to simplify data processing and reduce storage space. Spatial or frequency domain filters are used to remove noise in the image, and image enhancement operations are performed to improve the signal-to-noise ratio of the data. The preprocessed image data is cropped, the area of ​​interest is selected, and the image data is formatted and converted into a format suitable for subsequent processing and analysis; The weather monitoring and correction module monitors weather conditions in real time and corrects the pre-processed image data according to weather changes. The image data obtained by dynamic correction can reduce the impact of weather changes on image quality. The weather data of the target area is collected in real time through ground-based meteorological observation equipment, including the temperature, humidity, wind speed, wind direction, and precipitation indicators of the target area. Among them, the meteorological observation equipment includes temperature sensors, humidity sensors, anemometers, wind vanes, and rain gauges. The real-time collected weather data is compared with the preset historical standard weather data. At the same time, the real-time image data is compared with the historical standard image data. Combined with the collected weather data, the real-time image data is analyzed in real time, the interference threshold is preset, and the image interference index is calculated to evaluate the interference degree of weather conditions on the image data quality and determine whether it is within the allowable interference range. If the weather conditions have an impact on the image data, the image interference index is calculated. If the image data is within the allowable interference range, the image data can be directly applied to the subsequent processing flow. If the weather conditions are not within the allowable interference range for the image data, the image data is corrected. The correction process includes the steps of atmospheric correction, geometric correction and color correction. The image data is atmospherically corrected for the atmospheric scattering, absorption and other effects caused by weather changes to eliminate or reduce the impact of weather changes on the image quality. If the weather changes cause geometric distortion of the image data, geometric correction is required to restore the true shape of the image data. For color correction, color correction is performed for the color distortion caused by weather changes to make the color of the image data more realistic and natural. The quality of the corrected image data is evaluated to ensure that the correction effect meets the requirements. If the correction effect is not good, the correction strategy is readjusted and correction is performed again until the required image data standard is met. Furthermore, the calculation expression of the image interference index is: ; in, is the image interference index, It is the measured value of meteorological data of various meteorological indicators (temperature, humidity, wind speed, wind direction, precipitation). It is the historical standard meteorological data benchmark value of each meteorological index. and are the historical maximum and benchmark minimum values ​​of each meteorological index, is a weight factor, which indicates the influence of different meteorological indicators on image interference. , is the temperature weight, is the humidity weight, is the wind speed weight, is the wind direction weight, is the precipitation weight, The value range is between 0 and 1. The smaller the value of indicates, the less interference the weather conditions have on the image data, that is, the higher the image quality. When it is close to 1, it means that the weather conditions have a great impact on the image data and correction processing is needed. When it is close to 0, it means that weather conditions have little interference with image data, and the image data quality is high, and can be directly used for subsequent processing; The image matching and positioning module extracts feature points from the corrected image data, and uses feature point matching technology to match and locate the images to establish spatial relationships, improve the accuracy of spatial measurement, and reduce positioning errors caused by weather changes. The feature point detection algorithm (SIFT) is used to detect stable feature points from the corrected image, and a descriptor is generated for each detected feature point. The descriptor contains local information about the gradient direction and size of the feature point. The descriptor is a quantitative description of the neighborhood around the feature point and is used for subsequent matching processes. Among them, feature points are usually corners, edges, or areas with significant textures in the image. They have unique descriptions in the image and are robust to rotation, scale changes, and illumination changes. The fast approximate nearest neighbor search matching algorithm is used to match feature points in multiple images. The matching process determines which feature points are corresponding points based on the similarity of the descriptors, and matches the feature point descriptors in the image to be matched with the feature point descriptors in the reference image. During the matching process, the similarity between the descriptors is calculated to find the matching pair with the highest similarity, which is fast. The approximate nearest neighbor search (FLANN) algorithm accelerates the matching process and improves the matching efficiency by constructing an index structure. After finding the matching pair with the highest similarity, the Lowe's ratio test method is used to screen the feature points of the preliminary match. The matching pairs are screened by comparing the ratio of the nearest neighbor distance to the second nearest neighbor distance. Only the matching pairs with a ratio less than the set threshold are valid matching pairs, and then the wrong matching pairs are removed. According to the matched feature point pairs, the geometric transformation model between the image to be matched and the reference image is estimated, the spatial relationship between the image to be matched and the reference image is analyzed, and the parameters of the geometric transformation model are optimized through an iterative optimization algorithm (RANSAC algorithm). The RANSAC algorithm estimates the most likely geometric transformation model from the feature point set containing the wrong matching pairs through multiple random sampling and verification, and removes the wrong matching pairs. After determining the geometric transformation model, the geometric transformation model is used to locate the image to be matched in the coordinate system of the reference image to achieve spatial alignment between the images. Through image positioning, the spatial relationship between the image to be matched and the reference image is established, which is helpful for subsequent spatial measurement and analysis. The 3D reconstruction and model generation module performs 3D reconstruction based on image matching and positioning results, generates digital elevation models and 3D models, provides 3D spatial information, and provides intuitive and accurate data support for geographic information monitoring. It extracts image data processed by the image matching and positioning module, including calibrated images, feature point matching results, and geometric transformation models. It uses the feature point matching results and geometric transformation models to reconstruct the 3D sparse point cloud between images through the structure from motion (SFM) method, and optimizes the reconstructed sparse point cloud to improve its accuracy and stability. On the basis of the sparse point cloud, it uses the multi-view stereo vision (MVS) method to generate a dense 3D point cloud through matching and depth estimation of multi-view images, and filters and denoises the generated dense point cloud to remove noise points and unnecessary objects to improve the density and accuracy of the point cloud. The meshing algorithm of isson surface reconstruction converts dense point clouds into three-dimensional grids to form a continuous three-dimensional surface, which provides a basis for subsequent texture mapping. The generated three-dimensional grid is optimized to improve its geometric accuracy and topological structure. The texture information of the original image is mapped to the three-dimensional grid through UV mapping technology to generate a new realistic three-dimensional model. The reconstructed three-dimensional point cloud is classified and divided into ground points and non-ground points. The ground point cloud data is used to generate the elevation values ​​of regular grid points through the interpolation method of inverse distance weighted interpolation, and then the interpolated elevation values ​​are output as digital elevation model (DEM) files, such as GeoTIFF format. The generated three-dimensional model is optimized to improve the geometric accuracy and texture quality of the model. The optimization process includes mesh simplification, texture repair, and lighting adjustment steps, and the optimized three-dimensional model is exported. Real-time dynamic monitoring module, used to analyze the three-dimensional model in real time, monitor the changes of geographic spatial data, update the geographic information database, realize continuous monitoring of the geographical environment, and reduce monitoring gaps caused by weather changes; The intelligent perception warning module combines the analysis results of geospatial data and uses machine learning technology to monitor the data for further analysis, clarify abnormal changes in the geographical environment, and perform corresponding warning operations to improve the system's response speed and ensure the accuracy of geographic information monitoring.

[0021] Embodiment 2, as Figure 3 As shown, based on Example 1, the present invention provides a technical solution: Preferably, in the real-time dynamic monitoring module, the process of updating the geographic information database is: The 3D model data processed by the 3D reconstruction and model generation module is accessed, and the real-time collected image data and weather data are accessed to provide more comprehensive monitoring information. The 3D model contains key geographical elements such as terrain, buildings, and vegetation. The real-time collected image data is matched with the 3D model to ensure the accuracy and consistency of the data, and then the 3D model is detected in real time. By comparing the image data and 3D model at different time points, the changed areas in the model are identified and marked, including terrain undulations, the addition or removal of buildings, and the growth or decline of vegetation. For the changed areas, the key attribute information of height, area, and volume is extracted from them, and the attribute information is calibrated by comparing historical data. Verification and validation to ensure its accuracy and reliability, update the detected change information and extracted attribute information to the geographic information database, obtain geospatial data, ensure that the geospatial data in the database is consistent with the actual geographical environment, perform quality control and verification on the updated data, and revise or re-collect the data that does not meet the requirements, intuitively display the monitoring results through 3D visualization technology, and generate a detailed monitoring report based on the monitoring results. The report content includes monitoring time, monitoring area, change type (terrain undulation, addition or removal of buildings, growth or decline of vegetation), change amount (height change, area change, volume change, etc.), attribute information (key attributes such as height, area, volume, etc.); In the intelligent perception warning module, the analysis process of abnormal changes in the geographical environment is as follows: Use geographic information system to perform spatial analysis on preprocessed geospatial data, including spatial interpolation, spatial overlay, and buffer analysis, to extract spatial characteristics and laws of the geographical environment, combine historical data with geospatial data obtained by real-time monitoring, analyze dynamic change trends of the geographical environment, and mark abnormal change data in geospatial data. Among them, through spatial interpolation, the data of known points are used to infer the data of unknown points to generate more complete geospatial data. Through spatial overlay, geospatial data from different sources and different themes are superimposed to reveal the spatial relationship and interaction between them. Through buffer analysis, a buffer zone of a certain width is created around a specific geographical element to analyze its impact on the surrounding environment. Use historical data and annotated abnormal change data in combination with a neural network model to enable the model to learn the normal state and abnormal change pattern of the geographical environment, train the abnormal change model of the geographical environment, and calculate the geographical environment alienation coefficient based on the output results of the abnormal change model of the geographical environment, quantify the degree of deviation between the geographical environment and the normal state, and preset the deviation threshold of the geographical environment. , compare the calculated geographical environment alienation coefficient with the preset deviation threshold. When the geographical environment alienation coefficient exceeds the threshold, it is considered that the geographical environment has undergone abnormal changes. When the abnormal change is confirmed, the early warning mechanism is triggered, and early warning information is sent to the user, including detailed information on the location, type, and degree of the abnormal change. The abnormal change data is saved in the database, and corresponding response measures are taken according to the early warning information, such as activating emergency plans, dispatching personnel for on-site verification, etc.; Furthermore, the calculation expression of geographical environment alienation coefficient is: ; in, is the geographical environment alienation coefficient, is the value of the geospatial data acquired through real-time monitoring. is the reference value of the corresponding position in the historical data, and All The maximum and minimum values ​​in the value, is the sample size, is the exponent in the exponential function, a positive real number, used to adjust the curve shape of the exponential function. Represents the sum of squared deviations of all sample points, after normalization. The exponential decay term representing the sample point deviation is used to capture the nonlinear deviation of the geographical environment. The value range is between 0 and 1. The smaller the value of indicates the smaller the deviation of the geographical environment from the historical normal state, that is, the more stable the geographical environment is. It should be noted that when When it is close to 1, it means that the geographical environment has undergone major abnormal changes and needs attention. As the geographical environment gradually deviates from the normal state, The value will gradually increase, thus providing a basis for early warning.

[0022] Embodiment 3, as Figure 4 As shown, based on Embodiment 1-2, the present invention further provides a real-time geographic information monitoring method, which is implemented based on a real-time geographic information monitoring system and includes the following steps: Step 1, obtaining geographic space image data by drone or aerial photography, and preprocessing and correcting the image data; Step 2: extract key feature points from the preprocessed data, match and locate the images, and perform three-dimensional reconstruction to generate a digital elevation model and a three-dimensional model; Step 3: Use geographic information system to perform spatial analysis on the preprocessed data, and combine historical data and real-time monitoring data to analyze the dynamic change trend of the geographical environment and mark abnormal change data in the geographic spatial data; Step 4: Compare data at different time points to identify areas of terrain, buildings, and vegetation changes, use historical data and abnormal change data to train the neural network model, and learn normal states and abnormal change patterns; Step 5, calculating the geographical environment alienation coefficient based on the trained model, quantifying the degree of deviation, and comparing the calculated geographical environment alienation coefficient with the preset deviation threshold to determine whether an abnormal change occurs; Step 6: When the geographical environment alienation coefficient exceeds the deviation threshold, the early warning mechanism is triggered, and an early warning message is sent to the user, and corresponding response measures are taken according to the early warning message.

[0023] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A real-time geographic information monitoring system, comprising a geographic information monitoring platform, characterized in that: The geographic information monitoring platform is communicatively connected with an image data acquisition module, a weather monitoring and correction module, an image matching and positioning module, a three-dimensional reconstruction and model generation module, a real-time dynamic monitoring module and an intelligent perception warning module, wherein the electrical signals between the modules are connected; The image data acquisition module uses drone aerial photography to acquire geographic space image data and pre-processes the acquired raw image data; The weather monitoring and correction module monitors weather conditions in real time and corrects the pre-processed image data according to weather changes; The image matching and positioning module extracts feature points from the corrected image data and uses feature point matching technology to match and position the images to establish a spatial relationship; The three-dimensional reconstruction and model generation module performs three-dimensional reconstruction based on the image matching and positioning results to generate a digital elevation model and a three-dimensional model; The real-time dynamic monitoring module is used to analyze the three-dimensional model in real time, monitor the changes of geographic spatial data, update the geographic information database, and realize continuous monitoring of the geographic environment; The intelligent perception warning module combines the analysis results of geographic spatial data and uses machine learning technology to monitor the data for further analysis, clarify abnormal changes in the geographic environment, and perform corresponding warning operations.

2. A real-time geographic information monitoring system according to claim 1, characterized in that: In the image data acquisition module, the image data acquisition and preprocessing process is as follows: Analyze the target area, formulate the flight plan of the UAV according to the geographical location, topography, weather conditions and factors of the target area, and carry out route planning and flight preparation; Carry out flight missions according to the established flight plan, take aerial photos of the target area through the camera on the drone, and collect geographic spatial image data in real time; The collected original image data is preprocessed by image grayscale and image enhancement. The component method is used to perform image grayscale processing operations, the collected color image data is converted into a grayscale image, and a filter is used to remove noise in the image and perform image enhancement operations; The preprocessed image data is cropped, the region of interest is selected, and the image data is format converted.

3. A real-time geographic information monitoring system according to claim 2, characterized in that: In the weather monitoring and correction module, the process of correcting image data is as follows: The weather data of the target area is collected in real time through the meteorological observation equipment of ground-based observation, including the temperature, humidity, wind speed, wind direction, and precipitation indicators of the target area. The meteorological observation equipment includes a temperature sensor, a humidity sensor, an anemometer, a wind vane, and a rain gauge; Compare the real-time collected weather data with the preset historical standard weather data, and compare the real-time image data with the historical standard image data. Combine the collected weather data to analyze the real-time image data in real time, preset the interference threshold, and calculate the image interference index to evaluate the degree of interference of weather conditions on the image data quality and determine whether it is within the allowable interference range. If the weather conditions are within the permissible interference range for the image data, the image data can be directly applied to the subsequent processing flow. If the weather conditions are not within the permissible interference range for the image data, the image data is corrected, and the correction process includes the steps of atmospheric correction, geometric correction and color correction. The quality of the corrected image data is evaluated. If the correction effect is not good, the correction strategy is readjusted and the correction is performed again until the required image data standard is met.

4. A real-time geographic information monitoring system according to claim 3, characterized in that: The calculation expression of the image interference index is: ； in, is the image interference index, is the measured value of meteorological data of each meteorological index, It is the historical standard meteorological data benchmark value of each meteorological index. and are the historical maximum and benchmark minimum values ​​of each meteorological index, is a weight factor, which indicates the influence of different meteorological indicators on image interference. , is the temperature weight, is the humidity weight, is the wind speed weight, is the wind direction weight, is the precipitation weight, The value range is between 0 and 1.

5. A real-time geographic information monitoring system according to claim 4, characterized in that: In the image matching and positioning module, the process of matching and positioning the image is as follows: Using a feature point detection algorithm to detect feature points from the rectified image, and generating a descriptor for each detected feature point; Use the fast approximate nearest neighbor search matching algorithm to match the feature points in multiple images, and match the feature point descriptors in the image to be matched with the feature point descriptors in the reference image. During the matching process, calculate the similarity between the descriptors and find the matching pair with the highest similarity; After finding the matching pairs with the highest similarity, the Lowe's ratio test method is used to screen the feature points of the preliminary match. The matching pairs are screened by comparing the ratio of the nearest neighbor distance to the next nearest neighbor distance. Only the matching pairs with a ratio less than the set threshold are valid matching pairs, and the wrong matching pairs are removed. According to the matched feature point pairs, the geometric transformation model between the image to be matched and the reference image is estimated, the spatial relationship between the image to be matched and the reference image is analyzed, and the parameters of the geometric transformation model are optimized through an iterative optimization algorithm; After determining the geometric transformation model, the image to be matched is positioned in the coordinate system of the reference image using the geometric transformation model to achieve spatial alignment between the images. Through image positioning, the spatial relationship between the image to be matched and the reference image is established.

6. A real-time geographic information monitoring system according to claim 5, characterized in that: In the 3D reconstruction and model generation module, the generation process of the digital elevation model and the 3D model is as follows: Extract image data processed by the image matching and positioning module, including the corrected image, feature point matching results, and geometric transformation model; Using the feature point matching results and the geometric transformation model, the three-dimensional sparse point cloud between images is reconstructed through the structure from motion method; Based on the sparse point cloud, a multi-view stereo vision method is used to generate a dense 3D point cloud through multi-view image matching and depth estimation, and the generated dense point cloud is filtered and denoised. The dense point cloud is converted into a 3D mesh using the Poisson surface reconstruction meshing algorithm to form a continuous 3D surface, and the texture information of the original image is mapped onto the 3D mesh through UV mapping technology to generate a new 3D model. The reconstructed 3D point cloud is classified into ground points and non-ground points. The ground point cloud data is used to generate the elevation values ​​of regular grid points through the inverse distance weighted interpolation method, and then the interpolated elevation values ​​are output as digital elevation model files. The generated 3D model is optimized, including mesh simplification, texture repair, and lighting adjustment steps, and the optimized 3D model is exported.

7. A real-time geographic information monitoring system according to claim 6, characterized in that: In the real-time dynamic monitoring module, the process of updating the geographic information database is as follows: Access the 3D model data processed by the 3D reconstruction and model generation module, and access the real-time collected image data and weather data. The 3D model contains key geographical elements such as terrain, buildings, and vegetation. Match the real-time collected image data with the 3D model, and then perform real-time change detection on the 3D model. By comparing the image data and 3D model at different time points, the changed areas in the model can be identified and marked, including terrain undulations, the addition or removal of buildings, and the growth or decline of vegetation. For the changed areas, key attribute information such as height, area and volume is extracted, and the attribute information is verified and validated by comparing with historical data; Update the detected change information and extracted attribute information into the geographic information database, obtain geospatial data, perform quality control and verification on the updated data, and correct or re-collect the data that does not meet the requirements; Through three-dimensional visualization technology, the monitoring results are displayed intuitively, and a monitoring report is generated based on the monitoring results. The report content includes monitoring time, monitoring area, change type, change amount, and attribute information.

8. A real-time geographic information monitoring system according to claim 7, characterized in that: In the intelligent perception warning module, the analysis process of abnormal changes in the geographical environment is as follows: Use geographic information systems to perform spatial analysis on pre-processed geospatial data, combine historical data with geospatial data acquired through real-time monitoring, analyze the dynamic change trend of the geographical environment, and mark abnormal change data in geospatial data; By combining historical data and annotated abnormal change data with a neural network model, the model can learn the normal state and abnormal change patterns of the geographical environment and train the abnormal change model of the geographical environment; Based on the output results of the geographic environment abnormal change model, the geographic environment alienation coefficient is calculated to quantify the degree of deviation between the geographic environment and the normal state; Preset geographic environment deviation threshold , the calculated geographical environment alienation coefficient is compared with the preset deviation threshold. When the geographical environment alienation coefficient exceeds the threshold, it is considered that the geographical environment has undergone abnormal changes; When an abnormal change is confirmed, the early warning mechanism is triggered and an early warning message is sent to the user, including detailed information on the location, type, and degree of the abnormal change. The abnormal change data is saved in the database, and corresponding response measures are taken based on the early warning information.

9. A real-time geographic information monitoring system according to claim 8, characterized in that: The calculation expression of the geographical environment alienation coefficient is: ； in, is the geographical environment alienation coefficient, is the value of the geospatial data acquired through real-time monitoring. is the reference value of the corresponding position in the historical data, and All The maximum and minimum values ​​in the value, is the sample size, is the exponent in the exponential function, a positive real number, used to adjust the curve shape of the exponential function. The value range is between 0 and 1.

10. A real-time geographic information monitoring method, implemented based on the real-time geographic information monitoring system according to any one of claims 1 to 9, characterized in that: The following steps are involved: Step 1, obtaining geographic space image data by drone or aerial photography, and preprocessing and correcting the image data; Step 2: extract key feature points from the preprocessed data, match and locate the images, and perform three-dimensional reconstruction to generate a digital elevation model and a three-dimensional model; Step 3: Use geographic information system to perform spatial analysis on the preprocessed data, and combine historical data and real-time monitoring data to analyze the dynamic change trend of the geographical environment and mark abnormal change data in the geographic spatial data; Step 4: Compare data at different time points to identify areas of terrain, buildings, and vegetation changes, use historical data and abnormal change data to train the neural network model, and learn normal states and abnormal change patterns; Step 5, calculating the geographical environment alienation coefficient based on the trained model, quantifying the degree of deviation, and comparing the calculated geographical environment alienation coefficient with the preset deviation threshold to determine whether an abnormal change occurs; Step 6: When the geographical environment alienation coefficient exceeds the deviation threshold, the early warning mechanism is triggered, and an early warning message is sent to the user, and corresponding response measures are taken according to the early warning message.

Images