Urban greening state monitoring method and system based on digital mirror image information

Abstract

The invention relates to the technical field of greening management, in particular to an urban greening state monitoring method and system based on digital mirror image information, and the method comprises the steps: collecting the real-time live-action data of a designated city, combining the real-time operation rule of the city, and generating a virtual city update mirror image through a digital twinborn technology; extracting digital mirror image information in a specified time period from the virtual city updated mirror image to obtain greening state mirror image data, and generating a virtual city greening state by analyzing the mirror image data; further detecting the greening state of the virtual city, generating greening state early warning information, and feeding back the greening state early warning information to the real city to determine a monitoring area and a monitoring index; and performing live-action monitoring on the designated area of the real city, collecting and analyzing monitoring data to obtain an actual monitoring result of the alarm area, and synchronizing the actual monitoring result to the virtual city update mirror image. Dynamic monitoring and bidirectional feedback of the greening state are realized, and technical support is provided for fine management and scientific decision-making of urban greening.

Description

Technical Field

[0001] The present invention relates to the technical field of urban greening management, and in particular to an urban greening status monitoring method and system based on digital mirror information. Background Art

[0002] In urban planning and management, real-time monitoring of greening status is crucial for urban environmental optimization, ecological protection and sustainable development. At present, the monitoring methods of urban greening status mainly include vegetation index (such as NDVI) analysis based on remote sensing images, image recognition and analysis of drone images, and monitoring based on ground sensors. These methods use different data collection methods and technical means to comprehensively monitor urban greening. Although these methods have achieved certain results in greening monitoring, they still face some technical limitations.

[0003] In the field of urban greening monitoring, the existing technology has a huge amount of urban data and it is difficult to efficiently collect greening monitoring indicators, resulting in excessive irrelevant data, diluting effective information, and affecting the efficiency and accuracy of data processing. In addition, greening monitoring data is dynamic, and existing methods are difficult to achieve real-time monitoring and accurate supervision, which restricts the effectiveness of greening management. The dynamic changes in greening areas make it impossible for existing monitoring models to adapt in a timely manner and provide clear decision-making basis for urban managers. At the same time, the redundancy of monitoring data and the low processing efficiency further limit the management department's ability to supervise and optimize greening resources.

[0004] Therefore, a method and system for monitoring urban greening status based on digital mirror information are proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for monitoring the status of urban greening based on digital mirror information. By collecting real-time real-scene data of a specified city, combining real-time operation rules, using digital twin technology to generate a virtual city renewal mirror, extracting digital mirror information related to greening, dynamically analyzing and predicting the changing trends of greening coverage, vegetation health, soil moisture and related environmental factors, it provides accurate support for real-time monitoring and management of greening status. In addition, early warning information is generated by detecting the greening status of the virtual city, and it is fed back to the real city to guide real-scene monitoring and status evaluation, and optimize greening management decisions. This method effectively improves the timeliness, accuracy and intelligence level of greening monitoring, improves the efficiency of data collection and processing, and provides technical support for the dynamic management and ecological protection of urban greening.

[0006] To achieve the above object, the present invention provides the following technical solutions: A method for monitoring urban greening status based on digital mirror information, comprising: Collect real-time scene data of designated cities, including real-time GIS data, real-time urban 3D modeling data, real-time remote sensing image data, real-time meteorological data, real-time ground sensor data, real-time population data and real-time environmental data; Obtaining the real-time operation rules of the designated city, including real-time urban spatial layout rules, real-time green coverage rules, real-time vegetation growth cycle rules, real-time climate change rules, real-time pollutant diffusion rules, real-time traffic flow rules and real-time population activity rules; Mapping the real-time scene data and the real-time operation rules to a virtual city image through digital twin technology to generate a virtual city update image; Extracting digital image information within a specified time period from the virtual city renewal image to obtain greening status image data; Analyze the greening state mirror data to obtain the greening state of the virtual city; Performing greening status detection on the greening status of the virtual city to obtain greening status early warning information; Feedback the greening status warning information to the designated city to obtain monitoring areas and monitoring indicators; Monitor the designated city according to the monitoring area and the monitoring indicators, and collect real-scene monitoring data; Analyze the real-scene monitoring data to obtain the greening status of the alarm area; Performing greening status monitoring on the greening status of the alarm area to obtain actual monitoring results; The actual monitoring results are synchronized to the virtual city renewal image.

[0007] Furthermore, when collecting the real-time scene data of the designated city, an incremental data transmission method is adopted to collect and transmit the real-time GIS data, the real-time city three-dimensional modeling data, the real-time remote sensing image data, the real-time meteorological data, the real-time ground sensor data, the real-time population data and the real-time environmental data at different time intervals.

[0008] Furthermore, the construction process of the virtual city image includes: collecting historical real-scene data of the designated city to obtain historical city real-scene data, wherein the historical city real-scene data includes historical GIS data, historical city three-dimensional modeling data, historical remote sensing image data, historical meteorological data, historical ground sensor data, historical population data and historical environmental data; preprocessing the historical city real-scene data to obtain historical city real-scene standard data; constructing a preliminary virtual city image based on the historical city real-scene standard data; acquiring the historical operating laws of the designated city to obtain the historical operating laws of the city; mapping the historical operating laws of the city to the preliminary virtual city image to obtain the virtual city image.

[0009] Furthermore, obtaining the real-time operation law of the designated city includes: obtaining urban spatial distribution data and green space layout information from a GIS database, and extracting the real-time urban spatial layout law through spatial analysis technology; obtaining real-time remote sensing images from a remote sensing image database, calculating the normalized vegetation index and vegetation coverage change trend, analyzing the green space distribution in combination with three-dimensional modeling data, and extracting the real-time green coverage law; obtaining vegetation coverage change and environmental factor data from remote sensing images and meteorological databases, extracting the growth cycle and seasonal changes of vegetation using time series analysis, and obtaining the real-time vegetation growth cycle law; obtaining real-time meteorological data from a meteorological database, analyzing the distribution law of climate change in different time and space in combination with spatial statistical technology, and extracting the real-time climate change law; obtaining pollutant concentration and diffusion data from an environmental monitoring database, building a diffusion simulation model in combination with wind speed and direction data, and extracting the real-time pollutant diffusion law; obtaining real-time vehicle flow, speed and road occupancy data from a traffic database, and extracting the real-time traffic flow law in combination with dynamic traffic simulation technology; obtaining real-time population location and population trajectory data from a population activity database, extracting population flow trends and aggregation areas through a spatiotemporal clustering algorithm, and obtaining the real-time population activity law.

[0010] Further, extracting the greening state image data within the specified time period from the virtual city renewal image includes: In the virtual city renewal image, a causal network analysis technology is used to construct a causal model of the greening status of the virtual city, and an indicator pool is listed; the indicator pool is composed of candidate indicators, and the candidate indicators include greening status monitoring indicators and pest and disease risk factors; the pest and disease risk factors include disease index, insect population density, vegetation variety disease resistance and pest and disease transmission path; Performing spatiotemporal correlation analysis on the candidate indicators to obtain significant candidate indicators; using the Sobol sensitivity analysis method to calculate the sensitivity index of each of the significant candidate indicators to the green coverage rate, calculating the sensitivity contribution rate according to the sensitivity index, screening out the significant candidate indicators whose sensitivity contribution rate is higher than a preset threshold, and obtaining key indicators; The weight of each key indicator on the prediction of green coverage rate is evaluated by random forest feature importance analysis, and the key indicators with weights higher than the preset threshold are screened out to obtain important indicators; the multicollinearity indicators in the important indicators are regularized by LASSO regression model to obtain optimized indicators; Dynamically map the optimization index with the spatiotemporal data recorded in the virtual city renewal image to obtain spatiotemporal feature data; extract the spatiotemporal feature data, dynamically simulate the pest risk factor in combination with the pest transmission model, reconstruct and predict the spatiotemporal feature data, and generate the digital image information within the specified time period. The greening status image data includes: a greening coverage rate change trend graph, a vegetation health status distribution graph, a soil moisture dynamic distribution graph, a pollutant diffusion path graph, a meteorological factor change trend graph, and a pest transmission risk graph.

[0011] A city greening status monitoring system based on digital mirror information includes: a real-scene data acquisition module, which is used to collect real-time real-scene data of a specified city, including real-time GIS data, real-time urban three-dimensional modeling data, real-time remote sensing image data, real-time meteorological data, real-time ground sensor data, real-time population data and real-time environmental data; an operation law acquisition module, which is used to obtain the real-time operation law of the specified city, including real-time urban spatial layout law, real-time green coverage law, real-time vegetation growth cycle law, real-time climate change law, real-time pollutant diffusion law, real-time traffic flow law and real-time population activity law; a virtual city construction and update module, which is used to map the real-time real-scene data and the real-time operation law to a virtual city mirror through digital twin technology to generate a virtual city update mirror; a digital mirror information extraction module, which is used to extract a specified time period from the virtual city update mirror The digital mirror information in the greening state is obtained to obtain the greening state mirror data; the greening state analysis module is used to analyze the greening state mirror data to obtain the greening state of the virtual city; the greening state detection and early warning module is used to perform greening state detection on the greening state of the virtual city to obtain greening state early warning information; the early warning feedback analysis module is used to feed back the greening state early warning information to the designated city to obtain the monitoring area and monitoring indicators; the field monitoring module is used to monitor the designated city according to the monitoring area and the monitoring indicators and collect real-scene monitoring data; the alarm area greening state analysis module is used to analyze the real-scene monitoring data to obtain the greening state of the alarm area; the actual monitoring result generation module is used to perform greening state monitoring on the greening state of the alarm area to obtain the actual monitoring result; the virtual city synchronization module is used to synchronize the actual monitoring result to the virtual city update mirror.

[0012] Furthermore, when collecting the real-time scene data of the designated city, an incremental data transmission method is adopted to collect and transmit the real-time GIS data, the real-time city three-dimensional modeling data, the real-time remote sensing image data, the real-time meteorological data, the real-time ground sensor data, the real-time population data and the real-time environmental data at different time intervals.

[0013] Furthermore, the construction process of the virtual city image includes: collecting historical real-scene data of the designated city to obtain historical city real-scene data, including historical GIS data, historical city three-dimensional modeling data, historical remote sensing image data, historical meteorological data, historical ground sensor data, historical population data and historical environmental data; preprocessing the historical city real-scene data, including denoising, format conversion, data calibration and data completion, to obtain historical city real-scene standard data; constructing a preliminary virtual city image based on the historical city real-scene standard data; obtaining the historical operating laws of the designated city to obtain the historical operating laws of the city, including historical urban spatial layout laws, historical green coverage laws, historical vegetation growth cycle laws, historical climate change laws, historical pollutant diffusion laws, historical traffic flow laws and historical population activity laws; mapping the historical operating laws of the city to the preliminary virtual city image to obtain the virtual city image.

[0014] Furthermore, obtaining the real-time operation law of the designated city includes: obtaining urban spatial distribution data and green space layout information from a GIS database, and extracting the real-time urban spatial layout law through spatial analysis technology; obtaining real-time remote sensing images from a remote sensing image database, calculating the normalized vegetation index and vegetation coverage change trend, analyzing the green space distribution in combination with three-dimensional modeling data, and extracting the real-time green coverage law; obtaining vegetation coverage change and environmental factor data from remote sensing images and meteorological databases, extracting the growth cycle and seasonal changes of vegetation using time series analysis, and obtaining the real-time vegetation growth cycle law; obtaining real-time meteorological data from a meteorological database, analyzing the distribution law of climate change in different time and space in combination with spatial statistical technology, and extracting the real-time climate change law; obtaining pollutant concentration and diffusion data from an environmental monitoring database, building a diffusion simulation model in combination with wind speed and direction data, and extracting the real-time pollutant diffusion law; obtaining real-time vehicle flow, speed and road occupancy data from a traffic database, and extracting the real-time traffic flow law in combination with dynamic traffic simulation technology; obtaining real-time population location and population trajectory data from a population activity database, extracting population flow trends and aggregation areas through a spatiotemporal clustering algorithm, and obtaining the real-time population activity law.

[0015] Further, extracting the greening state image data within the specified time period from the virtual city renewal image includes: In the virtual city renewal image, a causal network analysis technology is used to construct a causal model of the greening status of the virtual city, and an indicator pool is listed; the indicator pool is composed of candidate indicators, and the candidate indicators include greening status monitoring indicators and pest and disease risk factors; the pest and disease risk factors include disease index, insect population density, vegetation variety disease resistance and pest and disease transmission path; Performing spatiotemporal correlation analysis on the candidate indicators to obtain significant candidate indicators; using the Sobol sensitivity analysis method to calculate the sensitivity index of each of the significant candidate indicators to the green coverage rate, calculating the sensitivity contribution rate according to the sensitivity index, screening out the significant candidate indicators whose sensitivity contribution rate is higher than a preset threshold, and obtaining key indicators; The weight of each key indicator on the prediction of green coverage rate is evaluated by random forest feature importance analysis, and the key indicators with weights higher than the preset threshold are screened out to obtain important indicators; the multicollinearity indicators in the important indicators are regularized by LASSO regression model to obtain optimized indicators; Dynamically map the optimization index with the spatiotemporal data recorded in the virtual city renewal image to obtain spatiotemporal feature data; extract the spatiotemporal feature data, dynamically simulate the pest risk factor in combination with the pest transmission model, reconstruct and predict the spatiotemporal feature data, and generate the digital image information within the specified time period. The greening status image data includes: a greening coverage rate change trend graph, a vegetation health status distribution graph, a soil moisture dynamic distribution graph, a pollutant diffusion path graph, a meteorological factor change trend graph, and a pest transmission risk graph.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention proposes a method for monitoring the status of urban greening based on digital mirror information, which combines digital twin technology and efficient data processing mechanism to achieve comprehensive supervision and dynamic management of the status of urban greening. By collecting and processing real-time real-scene data and combining it with real-time operation law analysis, it can quickly and accurately reflect the status of urban greening, significantly improving monitoring efficiency and information accuracy; based on the dynamic update mechanism of the virtual city renewal mirror, it provides an intuitive and visual presentation of the greening status, providing management departments with clear decision-making support tools. At the same time, it can generate early warning information in a timely manner, laying the foundation for the greening status monitoring of real cities; by synchronizing the actual monitoring results to the virtual city renewal mirror, a two-way feedback mechanism between virtual and reality is established, dynamically optimizing the accuracy and real-time performance of the virtual model, and ensuring the long-term stable operation of the monitoring system.

[0017] 2. The present invention adopts an incremental data transmission method. When collecting real-time real-scene data of a specified city, it collects and transmits various types of real-time data at different time intervals, which significantly reduces the redundancy of data transmission and bandwidth occupancy. Compared with the traditional full data collection method, this method can flexibly adjust the collection cycle according to the characteristics of different data update frequencies, so as to achieve efficient and refined data collection. At the same time, through the incremental transmission mechanism, only the changed or newly added data is transmitted, which reduces the computational pressure of data processing, speeds up the response speed of the real-time monitoring system, and ensures that the monitoring information of the greening status can more timely and accurately reflect the dynamic changes of the urban environment. This method not only improves the efficiency of data collection and transmission, but also greatly reduces the system operation cost, providing solid technical support for the real-time monitoring and dynamic management of the urban greening status.

[0018] 3. The present invention realizes a comprehensive analysis and accurate prediction of the urban greening status by extracting digital image information within a specified time period from the virtual urban renewal image, combining causal network analysis technology and multidimensional data processing methods. By constructing a causal model of the virtual urban greening status, key indicators and optimization indicators are gradually screened out, and combined with methods such as pest and disease transmission model, Sobol sensitivity analysis and random forest feature importance analysis, complex influencing factors are quantified and their changing laws are dynamically simulated. At the same time, the extracted spatiotemporal feature data are reconstructed and trend predicted to generate dynamic information such as changes in greening coverage, distribution of vegetation health status and risks of pest and disease transmission. Unlike traditional methods, the present invention not only accurately extracts monitoring areas and monitoring indicators related to greening status through virtual urban renewal images, but also provides a scientific basis and guidance for greening monitoring in real cities. Through the identification and quantitative analysis of key areas and key indicators, the pertinence and effectiveness of monitoring work are significantly improved, and data redundancy and resource waste are avoided. At the same time, the present invention can predict the changing trend and potential risks of future greening status in advance, providing a more scientific decision-making reference for management departments. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flow chart of a method for monitoring urban greening status based on digital mirror information provided by an embodiment of the present invention; Figure 2 A structural diagram of an urban greening status monitoring system based on digital mirror information provided by an embodiment of the present invention; Figure 3 A flowchart of the process of analyzing the spatiotemporal correlation of candidate indicators provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0021] "Virtual City Mirror" simulates, predicts and optimizes the city's operation process through the "digital twin" method. This method combines real physical objects with the virtual environment, enabling managers to make scientific decisions in the virtual space and optimally allocate resources.

[0022] See also Figures 1 to 3 The present invention provides a method and system for monitoring the status of urban greening based on digital mirror information. The technical solution is as follows:

[0023] Embodiment 1 City A previously used a method based on regular remote sensing image analysis to monitor the status of urban greening. It evaluated the overall greening status of the city through regular collection and manual analysis of satellite remote sensing images. However, this method has problems such as long data update cycle, limited spatial resolution, and insufficient sensitivity to dynamic changes, resulting in a lag in the supervision of greening status, and cannot meet the needs of real-time monitoring and precise management of greening status in the context of rapid urban development.

[0024] In order to improve the timeliness, accuracy and dynamic analysis capabilities of greening status monitoring, City A introduced a method for urban greening status monitoring based on digital mirror information, including: See also Figure 1 S1 in the above code collects real-time scene data of a designated city, including real-time GIS data, real-time urban 3D modeling data, real-time remote sensing image data, real-time meteorological data, real-time ground sensor data, real-time population data and real-time environmental data. Furthermore, when collecting real-time scene data of a designated city, an incremental data transmission method is adopted to collect and transmit real-time GIS data, real-time urban three-dimensional modeling data, real-time remote sensing image data, real-time meteorological data, real-time ground sensor data, real-time population data and real-time environmental data at different time intervals.

[0025] The specific data collection and update methods are shown in Table 1.

[0026] Table 1 Data table of urban real-time scene data collection and transmission mechanism

[0027] By adopting the incremental data transmission method to collect real-time real-scene data of a designated city, the present invention realizes efficient data update and transmission, and shows significant beneficial effects. Compared with the traditional full data transmission method, incremental data transmission only transmits new or changed data, which greatly reduces the bandwidth occupancy and computing resource consumption of data transmission, thereby significantly reducing the operating burden of the system. According to the timeliness requirements of different data types, differentiated collection and transmission time intervals are set, which not only ensures the timeliness of key data (such as real-time meteorological data and environmental monitoring data), but also optimizes the update frequency of non-real-time data (such as three-dimensional modeling data or remote sensing image data), and balances system performance and data accuracy. In addition, this method can capture the dynamic changes of urban greening status in real time, provide real-time and accurate monitoring information and abnormal warning support for management departments, thereby improving the efficiency and response speed of greening management, and realizing the refined allocation and sustainable utilization of urban resources.

[0028] See also Figure 1 S2 in the example obtains urban spatial distribution data and green space layout information from the GIS database, and extracts real-time urban spatial layout rules through spatial analysis technology; obtains real-time remote sensing images from the remote sensing image database, calculates the normalized vegetation index and vegetation coverage change trend, analyzes green space distribution in combination with three-dimensional modeling data, and extracts real-time green coverage rules; obtains vegetation coverage change and environmental factor data from remote sensing images and meteorological databases, extracts vegetation growth cycles and seasonal changes using time series analysis, and obtains real-time vegetation growth cycle rules; obtains real-time meteorological data from the meteorological database, and analyzes the distribution of climate change in different time and space in combination with spatial statistical technology, and extracts real-time climate change rules; obtains pollutant concentration and diffusion data from the environmental monitoring database, and constructs a diffusion simulation model in combination with wind speed and direction data to extract real-time pollutant diffusion rules; obtains real-time vehicle flow, speed and road occupancy data from the traffic database, and extracts real-time traffic flow rules in combination with dynamic traffic simulation technology; obtains real-time population location and population trajectory data from the population activity database, extracts population flow trends and aggregation areas through spatiotemporal clustering algorithms, and obtains real-time population activity rules.

[0029] By extracting the real-time operation rules of a designated city, the present invention accurately captures the dynamic characteristics of the city based on multi-source data and multiple analysis techniques, providing comprehensive support for greening status monitoring. It can dynamically analyze the changes in greening coverage, vegetation growth rules, climate impact, pollution diffusion, and the comprehensive effects of traffic and population activities on greening of the virtual city mirror, and reflect the operation status of the virtual city and its impact on greening in real time. It not only improves the dynamics and accuracy of monitoring, but also can quickly detect anomalies, provide scientific decision-making support for greening management, and effectively improve resource utilization efficiency and response speed.

[0030] See also Figure 1S3 in the system uses digital twin technology to map real-time scene data and real-time operation rules to a virtual city image, generating a virtual city renewal image. Furthermore, the construction process of the virtual city image includes: collecting historical real-scene data of the designated city to obtain historical city real-scene data, including historical GIS data, historical city three-dimensional modeling data, historical remote sensing image data, historical meteorological data, historical ground sensor data, historical population data and historical environmental data; preprocessing the historical city real-scene data, including denoising, format conversion, data calibration and data completion, to obtain historical city real-scene standard data; constructing a preliminary virtual city image based on the historical city real-scene standard data; obtaining the historical operating laws of the designated city to obtain the historical operating laws of the city, including historical urban spatial layout laws, historical green coverage laws, historical vegetation growth cycle laws, historical climate change laws, historical pollutant diffusion laws, historical traffic flow laws and historical population activity laws; mapping the historical operating laws of the city to the preliminary virtual city image to obtain a virtual city image.

[0031] In addition, Figure 1 In step S3 shown, the preprocessing of the real-time scene data and the real-time operation rules is consistent with the above-mentioned historical data preprocessing method.

[0032] By collecting and preprocessing historical real-life data (including denoising, format conversion, data calibration and completion), the noise and incompleteness in the historical data are eliminated, and the accuracy and consistency of the data are improved; the preliminary virtual city image constructed based on standardized historical data and the mapping of historical operating laws have achieved high-precision digital reconstruction of the city's complex dynamic system. This not only provides a solid data foundation for the analysis and prediction of the current greening status, but also can explore the potential laws in historical data and provide a scientific decision-making basis for urban greening management. In addition, the mapping of historical operating laws enables the virtual city image to have the ability to simulate complex dynamic processes such as greening coverage, vegetation growth cycle, climate change and pollutant diffusion, thereby supporting more accurate greening status prediction and dynamic optimization.

[0033] See also Figure 1 S4 in which the digital image information within a specified time period is extracted from the virtual city renewal image to obtain greening status image data; Further, extracting the greening state image data within the specified time period from the virtual city renewal image includes: In the virtual city renewal mirror, the causal network analysis technology is used to construct a causal model of the virtual city's greening status and list the indicator pool; the indicator pool is composed of candidate indicators, which include greening status detection indicators and pest and disease risk factors; greening status detection indicators include: vegetation coverage, vegetation health index, soil moisture, meteorological factors, pollutant concentration and population activity distribution; pest and disease risk factors include disease index, insect population density, vegetation variety disease resistance and pest and disease transmission path; Conduct spatiotemporal correlation analysis on candidate indicators. The specific process of spatiotemporal correlation analysis includes: (1) Global correlation analysis: The global correlation between each candidate indicator and green coverage was calculated using the Pearson correlation coefficient; The significance level is calculated through the significance test. When the significance level is not less than 0.05, the candidate indicator is marked as a elimination indicator. The indicators whose significance level does not meet the requirements are removed from the candidate indicators.

[0034] (2) Local correlation analysis: For the candidate indicators that passed the Pearson correlation significance test (the candidate indicators with a significance level less than 0.05), the local correlation between each candidate indicator and the green coverage rate was calculated using geographical weighted regression (GWR) to obtain the regression coefficient and its significance level of each geographical location; Conduct significance test on the geographically weighted regression results and calculate the proportion of significant areas; When the significant area ratio is less than the preset threshold, the candidate indicator is marked as a rejection indicator and removed from the candidate indicators; When the salient area ratio is greater than or equal to a preset threshold, the candidate index is retained to obtain a salient candidate index.

[0035] The Sobol sensitivity analysis method is used to calculate the sensitivity index of each significant candidate indicator to the green coverage rate, and the sensitive contribution rate is calculated according to the sensitivity index. The significant candidate indicators with a sensitive contribution rate higher than the preset threshold are screened out to obtain the key indicators. The weight of each key indicator on the prediction of green coverage rate was evaluated through random forest feature importance analysis, and the key indicators with weights higher than the preset threshold were screened out to obtain important indicators. The LASSO regression model was used to regularize the multicollinearity indicators in the important indicators to obtain the optimized indicators. Dynamically map the optimization index with the spatiotemporal data recorded in the virtual urban renewal mirror to obtain spatiotemporal feature data; extract the spatiotemporal feature data, combine the pest and disease transmission model (the pest and disease transmission model in this embodiment is the SEIR model) to dynamically simulate the pest and disease risk factors, reconstruct and predict the spatiotemporal feature data through a multidimensional interpolation algorithm and a time series analysis method, generate a green coverage rate change trend map, a vegetation health status distribution map, a soil moisture dynamic distribution map, a pollutant diffusion path map, a meteorological factor change trend map and a pest and disease transmission risk map within a specified time period, and obtain the greening status mirror data.

[0036] By extracting digital image information within a specified time period from the virtual city renewal image, the present invention realizes high-precision dynamic prediction and multi-dimensional comprehensive analysis of the urban greening status. By using causal network analysis and spatiotemporal correlation analysis methods, combined with global and local correlation analysis, the key indicators affecting the greening coverage rate are effectively screened out, ensuring the scientificity and accuracy of data analysis; by optimizing key indicators through Sobol sensitivity analysis, random forest feature importance analysis and LASSO regression model, multicollinearity is eliminated, and the reliability and efficiency of prediction are improved; finally, combined with pest and disease transmission models and dynamic simulation technology, a distribution map covering multi-dimensional spatiotemporal changes such as greening coverage rate, vegetation health, soil moisture, pollution diffusion and meteorological factors is generated, which fully reflects the dynamic changes and potential risks of greening status. The intelligent level of urban greening monitoring and management has been significantly improved, providing strong technical support for timely detection of anomalies, formulation of targeted management measures and optimization of urban greening resource allocation.

[0037] By introducing pest and disease transmission models into the virtual city mirror, its dynamic simulation capabilities can be enhanced, the transmission process of pests and diseases under different environmental conditions can be accurately reproduced, and prediction and early warning of high-risk areas can be achieved. By combining real-time data and historical data, the model can dynamically simulate the trend of pest and disease transmission and its linkage changes with the greening status, providing intelligent decision-making support for management departments and better adapting to the complex and changing urban environment needs. The introduction of pest and disease transmission further enhances the practicality and decision-making support capabilities of the virtual city mirror, and provides a scientific basis and forward-looking guidance for real-life greening management.

[0038] See also Figure 1 S5 in which the greening status mirror data is analyzed to obtain the greening status of the virtual city; See also Figure 1 S6 in which the greening state of the virtual city is detected to obtain greening state warning information; In this embodiment, the virtual city greening status includes greening coverage, vegetation health status, soil moisture distribution, pollutant concentration distribution, pest and disease transmission risk, and greening dynamic change trend, which is used to comprehensively reflect the spatial distribution and temporal characteristics of urban greening. Greening status detection accurately detects problems such as decreased greening coverage, deteriorated vegetation health, insufficient soil moisture, excessive pollutants, and pest and disease transmission through abnormal area identification, trend analysis, risk assessment, and spatial distribution analysis of the virtual city greening status, and quantifies the severity of abnormal changes (generating graded warning information) and the scope of impact.

[0039] See also Figure 1 S7 in the above method feeds back the warning information of greening status to the designated city and obtains the monitoring area and monitoring indicators; See also Figure 1 S8 in which the designated city is monitored according to the monitoring area and monitoring indicators, and real-life monitoring data is collected; See also Figure 1 S9 in the example analyzes the real-scene monitoring data to obtain the greening status of the alarm area; See also Figure 1 S10 in which the greening status of the alarm area is monitored to obtain the actual monitoring result; See also Figure 1 S11 in the example synchronizes the actual monitoring results to the virtual city renewal image.

[0040] In order to verify the effectiveness of the urban greening status monitoring method proposed in this paper, a set of comparative experiments was designed to compare its performance with the traditional method based on remote sensing image analysis (A city historical greening status monitoring method). The urban greening status data within the same time period were selected respectively. Figure 1 The complete process in the paper monitors the urban greening status during this period. The control group uses the historical greening status monitoring method to obtain the greening status monitoring results by regularly collecting remote sensing image data and analyzing and detecting the actual image data. In the experiment, the key technical indicators of the two methods such as data update cycle, anomaly detection rate, monitoring accuracy and data processing time are quantitatively compared, and the performance difference is verified by combining significance test.

[0041] Table 2 Comparative experimental results in city A

[0042] Refer to Table 2. The comparative experimental results show that the method based on digital mirror information is significantly superior to the traditional remote sensing image analysis method in multiple key indicators. The average data update cycle is shortened from 720 hours of the traditional method to 1 hour, and the dynamic change response time is shortened from 168 hours to 0.5 hours, showing a significant improvement in real-time performance. In terms of the success rate of abnormal state warning and the accuracy of pest and disease risk assessment, they are increased from 60% and 20% to 95% and 60% respectively, greatly improving the reliability of abnormal monitoring. At the same time, the data redundancy rate is reduced from 40% to 10%, and the effective data extraction rate is increased from 60% to 98%, indicating that the digital mirror method has significant advantages in the pertinence and effectiveness of data processing. In addition, the data processing time is reduced from 24 hours to 1 hour, the green coverage monitoring accuracy is increased from 85% to 98%, and the green state abnormality detection accuracy is increased from 82% to 93%, which fully reflects the advantages of the digital mirror method in accuracy and efficiency.

[0043] The present invention provides a method for monitoring the state of urban greening based on digital mirror information. By integrating multi-source real-time data (such as GIS data, three-dimensional modeling data, remote sensing image data and meteorological data, etc.) and real-time urban operation laws (such as greening coverage laws, vegetation growth cycle laws, etc.), a virtual urban renewal mirror is constructed, and greening state mirror data is extracted and analyzed in real time, thereby realizing dynamic monitoring and early warning of the state of urban greening. Firstly, the virtual urban renewal mirror is constructed by using digital twin technology, which can not only perform high-precision and dynamic digital mapping of the real city, but also quickly analyze the trend of greening state changes in the virtual environment, effectively making up for the problems of poor real-time performance and limited scope in traditional greening monitoring methods; secondly, through the deep mining and analysis of spatiotemporal data, combined with the greening state detection and early warning functions, abnormal greening areas or potential risks can be discovered in time, providing forward-looking guidance for urban greening management; by combining real-scene monitoring data with the dynamic update of virtual city mirrors, the synchronous optimization of monitoring results and prediction models is realized, greatly improving the accuracy and decision-making efficiency of the system. Therefore, the present invention forms a closed-loop system from data collection, analysis, feedback to dynamic adjustment, which has the advantages of strong real-time performance, wide coverage, and intelligent decision support, providing technical guarantee for the refined management and sustainable development of urban greening.

[0044] Embodiment 2 City B previously used a method based on ground inspections and cloud platform analysis to monitor the status of urban greening. Through regular manual inspections of greening areas and uploading the collected inspection data to the cloud platform, the cloud platform analyzes and returns vegetation coverage, soil moisture, and pest and disease conditions, and regularly generates greening status reports. However, this method has problems such as large manpower investment, limited data collection coverage, information lag, and susceptibility to human subjective factors. It is difficult to meet the real-time monitoring needs of dynamic changes in urban greening, and it cannot provide timely and effective warnings when facing large-scale pest and disease risks.

[0045] In order to realize intelligent supervision and management of greening status, City B adopted an urban greening status monitoring system based on digital mirror information. Figure 2 ,include: The real-scene data collection module is used to collect real-time real-scene data of a specified city, including real-time GIS data, real-time urban 3D modeling data, real-time remote sensing image data, real-time meteorological data, real-time ground sensor data, real-time population data and real-time environmental data; The operation law acquisition module is used to obtain the real-time operation laws of a specified city, including real-time urban spatial layout laws, real-time green coverage laws, real-time vegetation growth cycle laws, real-time climate change laws, real-time pollutant diffusion laws, real-time traffic flow laws, and real-time population activity laws; The virtual city construction and update module is used to map real-time scene data and real-time operation rules to the virtual city image through digital twin technology to generate a virtual city update image; A digital image information extraction module is used to extract digital image information within a specified time period from the virtual city renewal image to obtain greening status image data; The greening status analysis module is used to analyze the greening status mirror data to obtain the greening status of the virtual city; The greening status detection and early warning module is used to detect the greening status of the virtual city and obtain greening status early warning information; The early warning feedback analysis module is used to feed back the greening status early warning information to the designated city to obtain the monitoring area and monitoring indicators; The field monitoring module is used to monitor the designated city according to the monitoring area and monitoring indicators and collect real-scene monitoring data; The greening status analysis module of the alarm area is used to analyze the real-scene monitoring data to obtain the greening status of the alarm area; the actual monitoring result generation module is used to monitor the greening status of the alarm area and obtain the actual monitoring results; The virtual city synchronization module is used to synchronize the actual monitoring results to the virtual city update mirror.

[0046] The system includes an electronic device and a computer program that can be run on the electronic device. When the program is run, all the steps of the method of the embodiment 1 can be implemented. Specifically, the system collects real-time real-scene data such as real-time GIS data, three-dimensional modeling data, remote sensing image data, meteorological data, ground sensor data, population data and environmental data, and combines the real-time operation law of the city to generate a virtual city renewal image through digital twin technology, and extracts greening status image data from the image. The system performs spatiotemporal correlation analysis on the extracted data, calculates the correlation between candidate indicators and greening coverage rate using Pearson correlation coefficient and geographical weighted regression, and selects significant candidate indicators through significance test; further combines Sobol sensitivity analysis to select key indicators with high sensitive contribution rate. Based on the above analysis, the system generates greening status warning information and feeds it back to the monitoring area, realizes real-time monitoring of the greening status of the alarm area through real-scene data collection and analysis, and synchronizes the monitoring results to the virtual city renewal image to dynamically update the greening status information. Through the above functions, the system can fully execute all the steps of the embodiment 1 to meet the dynamic monitoring and warning needs of the urban greening status.

[0047] Specifically, Figure 3 As shown in the figure, the temporal and spatial correlation analysis of the candidate indicators is carried out, and the specific process includes: Firstly, the Pearson correlation coefficient was used to analyze the global correlation between each candidate indicator and green coverage rate, and the significance level of the indicator was calculated through significance test.

[0048] When the significance level of a candidate indicator is not less than a preset threshold (0.05 in this embodiment), the indicator is marked as a rejection indicator. Insignificant candidate indicators are rejected.

[0049] For the candidate indicators that passed the Pearson significance test, geographically weighted regression (GWR) was further used to analyze their local correlation with green coverage rate to obtain the regression coefficient and significance level of each geographical location.

[0050] A significance test was performed on the GWR regression results, and the proportion of significant areas was calculated.

[0051] When the salient area ratio is lower than a preset threshold (set to 0.7 in this embodiment), the candidate indicator is marked as a rejection indicator. Otherwise, the indicator is retained.

[0052] Finally, significant candidate indicators are screened and used as input for subsequent analysis.

[0053] In order to verify the technical advantages of the urban greening status monitoring method based on digital mirror information over the traditional manual inspection method (the historical greening status monitoring method of City B), City B was selected as the experimental scene, and the greening status was monitored using the traditional manual inspection method and the digital mirror method. In the same time period, the two methods were used to monitor key indicators such as greening coverage, soil moisture, and the risk of pests and diseases, and quantitative comparisons were made on indicators such as data collection efficiency, coverage, warning accuracy, and anomaly detection rate.

[0054] Table 3 Comparative experimental results of city B

[0055] Referring to the experimental results in Table 3, the method based on digital mirror information is significantly superior to the traditional manual inspection method in terms of data collection efficiency, coverage, dynamic response time, success rate of pest and disease risk warning, anomaly detection rate and data accuracy. In particular, in terms of data collection coverage, the digital mirror method has expanded its coverage by 10 times, while the data collection efficiency has increased by nearly 120 times. At the same time, the success rate of pest and disease risk warning has increased from 45% to 95%, and the data accuracy has increased from 75% to 98%. These results show that the digital mirror method can efficiently and accurately monitor the status of urban greening in real time, and is particularly suitable for large-scale, dynamic and frequently changing urban greening management scenarios.

[0056] Through mechanisms such as incremental collection, on-demand frequency modulation, real-time preprocessing and intelligent scheduling, the present invention reduces unnecessary computing and storage overhead, optimizes the entire life cycle of the data stream, and significantly improves the efficiency of data collection and processing.

[0057] By introducing digital twin technology, the present invention breaks through the reliance of traditional methods on large-scale urban data collection and analysis, and avoids the limitation of having to collect and analyze data at specific time nodes. Through the virtual mapping of real-time urban scene data and operating rules, the generated virtual urban renewal image can theoretically predict the greening status at any time point, thereby realizing the early detection of potential problems and providing targeted monitoring strategies and early warning information. It not only significantly improves the flexibility and initiative of greening status monitoring, but also reduces the need for manual inspections or regular data collection, greatly improves monitoring efficiency and accuracy, and provides more scientific and efficient technical support for urban greening management.

[0058] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for monitoring urban greening status based on digital mirror information, characterized in that: include: Collect real-time scene data of designated cities, including real-time GIS data, real-time urban 3D modeling data, real-time remote sensing image data, real-time meteorological data, real-time ground sensor data, real-time population data and real-time environmental data; Obtaining the real-time operation rules of the designated city, including real-time urban spatial layout rules, real-time green coverage rules, real-time vegetation growth cycle rules, real-time climate change rules, real-time pollutant diffusion rules, real-time traffic flow rules and real-time population activity rules; Mapping the real-time scene data and the real-time operation rules to a virtual city image through digital twin technology to generate a virtual city update image; Extracting digital image information within a specified time period from the virtual city renewal image to obtain greening status image data; Analyze the greening state mirror data to obtain the greening state of the virtual city; Performing greening status detection on the greening status of the virtual city to obtain greening status early warning information; Feedback the greening status warning information to the designated city to obtain monitoring areas and monitoring indicators; Monitor the designated city according to the monitoring area and the monitoring indicators, and collect real-life monitoring data; Analyze the real-scene monitoring data to obtain the greening status of the alarm area; Performing greening status monitoring on the greening status of the alarm area to obtain actual monitoring results; The actual monitoring results are synchronized to the virtual city renewal image.

2. The urban greening status monitoring method based on digital mirror information according to claim 1 is characterized in that: When collecting the real-time scene data of the designated city, an incremental data transmission method is adopted to collect and transmit the real-time GIS data, the real-time city three-dimensional modeling data, the real-time remote sensing image data, the real-time meteorological data, the real-time ground sensor data, the real-time population data and the real-time environmental data at different time intervals.

3. The urban greening status monitoring method based on digital mirror information according to claim 1 is characterized in that: The construction process of the virtual city image includes: collecting historical real-scene data of the designated city to obtain historical city real-scene data, wherein the historical city real-scene data includes historical GIS data, historical city three-dimensional modeling data, historical remote sensing image data, historical meteorological data, historical ground sensor data, historical population data and historical environmental data; preprocessing the historical city real-scene data to obtain historical city real-scene standard data; constructing a preliminary virtual city image according to the historical city real-scene standard data; acquiring the historical operating laws of the designated city to obtain the historical operating laws of the city; and mapping the historical operating laws of the city to the preliminary virtual city image to obtain the virtual city image.

4. The urban greening status monitoring method based on digital mirror information according to claim 1 is characterized in that: Acquiring the real-time operation law of the designated city includes: acquiring urban spatial distribution data and green space layout information from a GIS database, and extracting the real-time urban spatial layout law through spatial analysis technology; acquiring real-time remote sensing images from a remote sensing image database, calculating the normalized vegetation index and vegetation coverage change trend, analyzing the green space distribution in combination with three-dimensional modeling data, and extracting the real-time green coverage law; acquiring vegetation coverage change and environmental factor data from remote sensing images and meteorological databases, extracting the growth cycle and seasonal changes of vegetation by time series analysis, and obtaining the real-time vegetation growth cycle law; acquiring real-time meteorological data from a meteorological database, analyzing the distribution law of climate change in different time and space in combination with spatial statistical technology, and extracting the real-time climate change law; acquiring pollutant concentration and diffusion data from an environmental monitoring database, building a diffusion simulation model in combination with wind speed and direction data, and extracting the real-time pollutant diffusion law; acquiring real-time vehicle flow, speed and road occupancy data from a traffic database, and extracting the real-time traffic flow law in combination with dynamic traffic simulation technology; acquiring real-time population location and population trajectory data from a population activity database, extracting population flow trends and aggregation areas through a spatiotemporal clustering algorithm, and obtaining the real-time population activity law.

5. The urban greening status monitoring method based on digital mirror information according to claim 1 is characterized in that: Extracting the greening status image data within the specified time period from the virtual city renewal image includes: In the virtual city renewal image, a causal network analysis technology is used to construct a causal model of the greening status of the virtual city, and an indicator pool is listed; the indicator pool is composed of candidate indicators, and the candidate indicators include greening status monitoring indicators and pest and disease risk factors; the pest and disease risk factors include disease index, insect population density, vegetation variety disease resistance and pest and disease transmission path; Performing spatiotemporal correlation analysis on the candidate indicators to obtain significant candidate indicators; using the Sobol sensitivity analysis method to calculate the sensitivity index of each of the significant candidate indicators to the green coverage rate, calculating the sensitivity contribution rate according to the sensitivity index, screening out the significant candidate indicators whose sensitivity contribution rate is higher than a preset threshold, and obtaining key indicators; The weight of each key indicator on the prediction of green coverage rate is evaluated by random forest feature importance analysis, and the key indicators with weights higher than the preset threshold are screened out to obtain important indicators; the multicollinearity indicators in the important indicators are regularized by LASSO regression model to obtain optimized indicators; Dynamically map the optimization index with the spatiotemporal data recorded in the virtual city renewal image to obtain spatiotemporal feature data; extract the spatiotemporal feature data, dynamically simulate the pest risk factor in combination with the pest transmission model, reconstruct and predict the spatiotemporal feature data, and generate the digital image information within the specified time period. The greening status image data includes: a greening coverage rate change trend graph, a vegetation health status distribution graph, a soil moisture dynamic distribution graph, a pollutant diffusion path graph, a meteorological factor change trend graph, and a pest transmission risk graph.

6. An urban greening status monitoring system based on digital mirror information, characterized in that: include: A real-scene data acquisition module is used to collect real-time real-scene data of a specified city, including real-time GIS data, real-time urban three-dimensional modeling data, real-time remote sensing image data, real-time meteorological data, real-time ground sensor data, real-time population data and real-time environmental data; an operation law acquisition module is used to obtain the real-time operation law of the specified city, including real-time urban spatial layout law, real-time green coverage law, real-time vegetation growth cycle law, real-time climate change law, real-time pollutant diffusion law, real-time traffic flow law and real-time population activity law; a virtual city construction and update module is used to map the real-time real-scene data and the real-time operation law to a virtual city image through digital twin technology to generate a virtual city update image; A digital image information extraction module is used to extract digital image information within a specified time period from the virtual city renewal image to obtain greening status image data; A greening status analysis module is used to analyze the greening status mirror data to obtain the greening status of the virtual city; A greening status detection and early warning module is used to detect the greening status of the virtual city and obtain greening status early warning information; An early warning feedback analysis module is used to feed back the greening status early warning information to the designated city to obtain monitoring areas and monitoring indicators; A field monitoring module, used to monitor the designated city according to the monitoring area and the monitoring indicators, and collect real-scene monitoring data; The alarm area greening status analysis module is used to analyze the real-scene monitoring data to obtain the greening status of the alarm area; the actual monitoring result generation module is used to monitor the greening status of the alarm area to obtain the actual monitoring result; the virtual city synchronization module is used to synchronize the actual monitoring result to the virtual city renewal mirror.

7. The urban greening status monitoring system based on digital mirror information according to claim 6 is characterized in that: When collecting the real-time scene data of the designated city, an incremental data transmission method is adopted to collect and transmit the real-time GIS data, the real-time city three-dimensional modeling data, the real-time remote sensing image data, the real-time meteorological data, the real-time ground sensor data, the real-time population data and the real-time environmental data at different time intervals.

8. The urban greening status monitoring system based on digital mirror information according to claim 6 is characterized in that: The construction process of the virtual city image includes: collecting the historical real scene data of the designated city to obtain the historical city real scene data, including historical GIS data, historical city three-dimensional modeling data, historical remote sensing image data, historical meteorological data, historical ground sensor data, historical population data and historical environmental data; preprocessing the historical city real scene data, including denoising, format conversion, data calibration and data completion, to obtain the historical city real scene standard data; constructing a preliminary virtual city image according to the historical city real scene standard data; obtaining the historical operation law of the designated city to obtain the historical operation law of the city, including the historical urban spatial layout law, the historical green coverage law, the historical vegetation growth cycle law, the historical climate change law, the historical pollutant diffusion law, the historical traffic flow law and the historical population activity law; mapping the historical operation law of the city to the preliminary virtual city image to obtain the virtual city image.

9. The urban greening status monitoring system based on digital mirror information according to claim 6 is characterized in that: Acquiring the real-time operation law of the designated city includes: acquiring urban spatial distribution data and green space layout information from a GIS database, and extracting the real-time urban spatial layout law through spatial analysis technology; acquiring real-time remote sensing images from a remote sensing image database, calculating the normalized vegetation index and vegetation coverage change trend, analyzing the green space distribution in combination with three-dimensional modeling data, and extracting the real-time green coverage law; acquiring vegetation coverage change and environmental factor data from remote sensing images and meteorological databases, extracting the growth cycle and seasonal changes of vegetation by time series analysis, and obtaining the real-time vegetation growth cycle law; acquiring real-time meteorological data from a meteorological database, analyzing the distribution law of climate change in different time and space in combination with spatial statistical technology, and extracting the real-time climate change law; acquiring pollutant concentration and diffusion data from an environmental monitoring database, building a diffusion simulation model in combination with wind speed and direction data, and extracting the real-time pollutant diffusion law; acquiring real-time vehicle flow, speed and road occupancy data from a traffic database, and extracting the real-time traffic flow law in combination with dynamic traffic simulation technology; acquiring real-time population location and population trajectory data from a population activity database, extracting population flow trends and aggregation areas through a spatiotemporal clustering algorithm, and obtaining the real-time population activity law.

10. The urban greening status monitoring system based on digital mirror information according to claim 6 is characterized in that: Extracting the greening status image data within the specified time period from the virtual city renewal image includes: In the virtual city renewal image, a causal network analysis technology is used to construct a causal model of the greening status of the virtual city, and an indicator pool is listed; the indicator pool is composed of candidate indicators, and the candidate indicators include greening status monitoring indicators and pest and disease risk factors; the pest and disease risk factors include disease index, insect population density, vegetation variety disease resistance and pest and disease transmission path; Performing spatiotemporal correlation analysis on the candidate indicators to obtain significant candidate indicators; using the Sobol sensitivity analysis method to calculate the sensitivity index of each of the significant candidate indicators to the green coverage rate, calculating the sensitivity contribution rate according to the sensitivity index, screening out the significant candidate indicators whose sensitivity contribution rate is higher than a preset threshold, and obtaining key indicators; The weight of each key indicator on the prediction of green coverage rate is evaluated by random forest feature importance analysis, and the key indicators with weights higher than the preset threshold are screened out to obtain important indicators; the multicollinearity indicators in the important indicators are regularized by LASSO regression model to obtain optimized indicators; Dynamically map the optimization index with the spatiotemporal data recorded in the virtual city renewal image to obtain spatiotemporal feature data; extract the spatiotemporal feature data, dynamically simulate the pest risk factor in combination with the pest transmission model, reconstruct and predict the spatiotemporal feature data, and generate the digital image information within the specified time period. The greening status image data includes: a greening coverage rate change trend graph, a vegetation health status distribution graph, a soil moisture dynamic distribution graph, a pollutant diffusion path graph, a meteorological factor change trend graph, and a pest transmission risk graph.

Images