Satellite remote sensing monitoring, early warning and forecasting methods for geological disasters

Abstract

The present invention relates to the field of disaster early warning technology, specifically to a method for early warning and forecasting of geological disasters by satellite remote sensing monitoring, which includes the following steps: based on ground remote sensing images, edge information of images is extracted frame by frame, pixel difference values ​​and spatial displacement information are identified, and the surface area where changes occur is marked to obtain basic anomaly detection results. In the present invention, by analyzing the ground remote sensing images frame by frame to extract edge information and compare the pixel differences between the previous and next frames, the slight changes in the surface can be monitored more finely, and by matching and analyzing historical data, potential anomaly types can be accurately identified, and targeted early warnings can be given for different types of geological disasters. By performing time series analysis on the trends of abnormal areas, the development trend and potential impact of disasters can be predicted, providing data support for disaster prevention and mitigation, improving the response speed and accuracy of the early warning system, and reducing the losses caused by untimely disaster response.

Description

Technical Field

[0001] The present invention relates to the field of disaster early warning technology, and in particular to a method for satellite remote sensing monitoring, early warning and forecasting of geological disasters. Background Art

[0002] The field of disaster warning technology focuses on the development and implementation of methods for predicting natural disasters and reducing their negative impacts. This includes monitoring and early warning of a variety of natural events such as earthquakes, floods, typhoons, droughts and geological disasters. Key technologies involve meteorology, geographic information systems, remote sensing technology, and data analysis and modeling. Innovative advances in this field include using satellite and ground data to improve the accuracy and response speed of early warning systems, while integrating machine learning and artificial intelligence technologies to improve data processing and analysis capabilities, thereby achieving more effective disaster prediction and management.

[0003] Among them, the satellite remote sensing monitoring, early warning and forecasting method of geological disasters is a method of using satellite technology to monitor geological disasters, such as landslides, ground subsidence, collapse, etc., and provide early warning information to predict the occurrence of these events. This method uses remote sensing technology to collect information on the surface and underground, and predicts potential disasters by analyzing indicators of surface deformation and other geological activities. The use of this technology can effectively warn the government and the public in the relevant areas in advance, so that appropriate preventive measures can be taken to reduce the loss of life and property. The implementation of this method is particularly important in areas where geological disasters occur frequently, and it is a key technical approach to improve public safety and reduce the impact of natural disasters.

[0004] Traditional forecasting methods are not accurate enough in the detailed analysis of surface changes, especially in the extraction of the morphology and boundaries of abnormal surface areas. They are unable to effectively identify and predict small-scale or early geological abnormal changes. Traditional methods are not sufficient to fully utilize historical geological data when integrating and analyzing surface change characteristics. They lack effective data comparison and pattern matching, resulting in insufficient specificity or timeliness of early warning information. In actual operations, this leads to slow response to certain disaster types or high false alarm rates, affecting the timeliness and effectiveness of disaster response measures. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and propose a method for satellite remote sensing monitoring, early warning and forecasting of geological disasters.

[0006] In order to achieve the above object, the present invention adopts the following technical scheme: a method for satellite remote sensing monitoring, early warning and forecasting of geological disasters, comprising the following steps:

[0007] S1: Based on the ground remote sensing image, the edge information of the image is extracted frame by frame, the pixel difference value and spatial displacement information are identified, the changed surface area is marked, and the basic anomaly detection result is obtained;

[0008] S2: Based on the basic anomaly detection result, the marked surface area is gridded, and the abnormal area is separated from the background image through edge detection to obtain segmented abnormal area data;

[0009] S3: Based on the segmented abnormal region data, by comparing the brightness values ​​and color distribution of pixels in the abnormal region with the statistical characteristics of the surrounding area, quantifying the surface change characteristics of the abnormal region, and obtaining surface abnormality characteristic data;

[0010] S4: Based on the surface abnormality feature data, the surface feature value range in historical geological disaster events is called, the abnormal features are matched and analyzed, the abnormal type corresponding to the abnormal area is identified, and the abnormal type identification result is obtained;

[0011] S5: Based on the abnormal type identification result, the change trend analysis of the surface characteristic parameters of the potential disaster area is performed to evaluate the development trend range and intensity of the potential disaster, and obtain disaster development trend prediction data;

[0012] S6: Based on the disaster development trend prediction data, according to the combined geological, climatic and environmental parameters of the target area, the abnormal threshold of the target area is adjusted, the geological anomaly level in the target area is evaluated, and an early warning is pushed to relevant departments to obtain an abnormal notification result.

[0013] As a further solution of the present invention, the basic anomaly detection results include the marked change area position, pixel change value and boundary range of the change area, the segmented abnormal area data include boundary information after area division and the separated abnormal area image, the surface abnormality feature data include brightness distribution characteristics, color change parameters within the area and spatial distribution consistency information, the anomaly type identification results include the classified anomaly type, the associated historical feature value range and the verification result of the anomaly classification, the disaster development trend prediction data include the predicted regional expansion range, intensity level and time series change trend value, and the anomaly notification results include regional graded notification information, triggered graded notification list and associated real-time push channel information.

[0014] As a further solution of the present invention, based on the ground remote sensing image, the edge information of the image is extracted frame by frame, the pixel difference value and spatial displacement information are identified, and the changed surface area is marked. The steps of obtaining the basic anomaly detection result are specifically as follows:

[0015] S101: Based on the ground remote sensing image, read the image data frame by frame, apply edge enhancement filtering processing, highlight the surface edge in each frame, optimize the edge clarity and continuity, and obtain an edge enhanced image;

[0016] S102: Analyze the continuous image frames based on the edge enhanced image, identify the pixel displacement and change degree by measuring the edge position change and difference between adjacent frames, and obtain pixel change data;

[0017] S103: Based on the pixel change data, a change threshold is set, and areas exceeding the change threshold are screened out, and the areas are marked as locations where the surface changes significantly, to obtain basic anomaly detection results.

[0018] As a further solution of the present invention, based on the basic anomaly detection result, the marked surface area is gridded, and the abnormal area is separated from the background image by edge detection. The specific steps of obtaining the segmented abnormal area data are as follows:

[0019] S201: Based on the basic anomaly detection result, each detected abnormal area is gridded, and the abnormal area is divided according to the size and boundary of each grid, so as to obtain gridded area data;

[0020] S202: Based on the gridded area data, edge tracking technology is applied to the data in each grid to extract edge lines and morphological features of each grid, divide the boundaries between abnormal and non-abnormal areas, and obtain geometric boundary data;

[0021] S203: Based on the geometric boundary data, spatially separate the image in the abnormal region from the background image, isolate each abnormal region, and obtain segmented abnormal region data.

[0022] As a further solution of the present invention, based on the segmented abnormal region data, by comparing the brightness value and color distribution of pixels in the abnormal region with the statistical characteristics of the surrounding area, the surface change characteristics of the abnormal region are quantified, and the steps of obtaining the surface abnormal characteristic data are specifically as follows:

[0023] S301: Based on the segmented abnormal region data, using pixel scanning technology, analyzing the brightness values ​​of pixels in the abnormal region one by one, and collecting color information of each pixel to obtain brightness and color analysis data;

[0024] S302: Based on the brightness and color analysis data, compare the pixel index of the abnormal area with the pixel index of the surrounding non-abnormal area, evaluate the statistical difference between the images, identify the significant features of the abnormal area, and obtain significant feature data;

[0025] S303: Based on the significant feature data, by comparing with a set parameter threshold, the features representing surface changes are identified to obtain surface abnormality feature data.

[0026] As a further solution of the present invention, based on the surface abnormal feature data, the surface feature value range in historical geological disaster events is called, the abnormal features are matched and analyzed, and the abnormal type corresponding to the abnormal area is identified. The specific steps of obtaining the abnormal type identification result are:

[0027] S401: Based on the surface abnormality feature data, access the historical geological disaster database, extract the historical data range of similar surface features, collect historical surface feature values, and obtain a historical surface feature reference data set;

[0028] S402: Based on the historical surface feature reference data set, the surface features of the current abnormal area are compared and analyzed with the historical data, and a statistical matching technique is used to identify matching feature combinations to obtain matching surface feature data;

[0029] S403: Based on the matching surface feature data, the association between each surface feature and the known geological disaster type is analyzed, the matching anomaly type is identified, and an anomaly type identification result is obtained.

[0030] As a further solution of the present invention, based on the abnormal type identification result, the change trend analysis of the surface characteristic parameters of the potential disaster area is performed to evaluate the development trend range and intensity of the potential disaster, and the steps of obtaining the disaster development trend prediction data are specifically as follows:

[0031] S501: Based on the anomaly type identification result, extract the surface characteristic parameters of the potential disaster area, and use the time series analysis method to extract the displacement value, morphological evolution value and expansion rate in the target area within a period of time to obtain the surface change time series data;

[0032] S502: Based on the surface change time series data, combined with the geological, meteorological and environmental parameters of the region, trend analysis is performed by adjusting parameter weights to obtain trend analysis results;

[0033] S503: Based on the trend analysis results, evaluate the development trend scope and intensity of potential disasters, identify the disaster impact scope and the probability of disaster occurrence, and obtain disaster development trend prediction data.

[0034] As a further solution of the present invention, the formula for trend analysis is:

[0035]

[0036] Among them, D t represents the disaster development index at the target time point t, M t represents the displacement value at time point t, E t represents the morphological evolution value at time point t, S t represents the expansion rate at time point t, Vt Represents the current environment variable, V max is the maximum recorded value of the environment variable, and a, b, c, and d are weight coefficients.

[0037] As a further solution of the present invention, based on the disaster development trend prediction data, according to the combined geological, climatic and environmental parameters of the target area, the abnormal threshold of the target area is adjusted, the geological abnormality level in the target area is evaluated, and an early warning is pushed to the relevant departments. The specific steps of obtaining the abnormal notification result are:

[0038] S601: Based on the disaster development trend prediction data, the geological structure, climate conditions and environmental background of the target area are analyzed, the anomaly detection threshold of the area is adjusted, the sensitivity and accuracy of anomaly detection are optimized, and the adjusted threshold data is obtained;

[0039] S602: Based on the adjusted threshold data, the geological anomaly level in the target area is evaluated, and high-risk areas are identified and marked to obtain risk level evaluation data;

[0040] S603: Based on the risk level assessment data, trigger a corresponding notification, send warning information to relevant departments, and obtain an abnormal notification result.

[0041] As a further solution of the present invention, the formula for the abnormality detection threshold of the adjustment area is:

[0042]

[0043] Among them, T new is the adjusted anomaly detection threshold, T base represents the baseline anomaly detection threshold, P geo Represents the value of geological structural parameters, P max Represents the maximum preset value of the geological structure parameter, T climate Represents the current value of the climate condition data, W base Represents the baseline climate condition data, T env Represents the current value of the environmental background data, H base represents the baseline environmental background data, w1, w2 and w3 are the weight coefficients of geological parameters, climate conditions and environmental background for threshold adjustment respectively.

[0044] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0045] In the present invention, by analyzing the ground remote sensing images frame by frame to extract edge information and compare the pixel differences between the previous and next frames, it is possible to monitor the slight changes in the surface more finely. By processing the abnormal areas in a gridding manner and independently analyzing the brightness and color differences of each grid, the detection accuracy of the surface change characteristics is improved. By matching and analyzing with historical data, potential abnormal types can be accurately identified, and targeted early warnings can be given for different types of geological disasters. By performing time series analysis on the trends in abnormal areas, the development trend and potential impact of disasters can be predicted, providing data support for disaster prevention and mitigation, improving the response speed and accuracy of the early warning system, and reducing the losses caused by untimely response to disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0047] Figure 1 It is a schematic diagram of the main steps of the present invention.

[0048] Figure 2 This is a schematic diagram of the refinement of S1 of the present invention.

[0049] Figure 3 It is a schematic diagram of the refinement of S2 of the present invention.

[0050] Figure 4 This is a schematic diagram of the refinement of S3 of the present invention.

[0051] Figure 5 It is a schematic diagram of the refinement of S4 of the present invention.

[0052] Figure 6 It is a schematic diagram of the refinement of S5 of the present invention.

[0053] Figure 7 It is a detailed schematic diagram of S6 of the present invention. DETAILED DESCRIPTION

[0054] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0055] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0056] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.

[0057] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0058] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0059] See also Figure 1 The present invention provides a technical solution: a method for satellite remote sensing monitoring, early warning and forecasting of geological disasters, comprising the following steps:

[0060] S1: Based on the ground remote sensing image, the edge information of the image is extracted frame by frame, and the extracted edge information is matched and compared with the corresponding areas in the previous and next frame images, the pixel difference value and spatial displacement information are identified, the surface area where the change occurs is marked, and the basic anomaly detection result is obtained;

[0061] S2: Based on the basic anomaly detection results, the marked surface area is gridded, and the boundary and geometric morphology information of the abnormal area is extracted through edge detection, and the abnormal area is separated from the background image to obtain the segmented abnormal area data;

[0062] S3: Based on the segmented abnormal area data, by comparing the brightness value and color distribution of pixels in the abnormal area with the statistical characteristics of the surrounding area, the surface change characteristics of the abnormal area are quantified to obtain the surface abnormal characteristic data;

[0063] S4: Based on the surface anomaly feature data, the surface feature value range in historical geological disaster events is called, the anomaly features are matched and analyzed, the anomaly type corresponding to the anomaly area is identified, and the anomaly type identification result is obtained;

[0064] S5: Based on the results of anomaly type identification, the change trend of surface characteristic parameters in the potential disaster area is analyzed, and the displacement value, morphological evolution value and regional expansion rate are extracted according to the time series. Combined with the geological, meteorological and environmental parameters of the target area, the development trend range and intensity of potential disasters are evaluated to obtain disaster development trend prediction data;

[0065] S6: Based on the disaster development trend prediction data, according to the combined geological, climatic and environmental parameters of the target area, adjust the abnormal threshold of the target area, evaluate the geological anomaly level in the target area, trigger the corresponding notification, and push early warnings to relevant departments to obtain abnormal notification results.

[0066] The basic anomaly detection results include the marked change area location, pixel change value and boundary range of the change area. The segmented abnormal area data include the boundary information after area division and the separated abnormal area image. The surface abnormality feature data include brightness distribution characteristics, color change parameters within the area and spatial distribution consistency information. The anomaly type identification results include the classified anomaly type, the associated historical feature value range and the verification result of the anomaly classification. The disaster development trend prediction data include the predicted regional expansion range, intensity level and time series change trend value. The abnormal notification results include regional graded notification information, the triggered graded notification list and the associated real-time push channel information.

[0067] See also Figure 2 Based on the ground remote sensing image, the edge information of the image is extracted frame by frame, and the extracted edge information is matched and compared with the corresponding areas in the previous and next frame images, the pixel difference value and spatial displacement information are identified, and the surface area where the change occurs is marked. The specific steps to obtain the basic anomaly detection results are as follows:

[0068] S101: Based on the ground remote sensing image, read the image data frame by frame, apply edge enhancement filtering processing, highlight the surface edge in each frame, optimize the edge clarity and continuity, and obtain an edge enhanced image;

[0069] Based on the ground remote sensing image, preprocessing is performed on the read ground remote sensing image, including denoising and color correction, to ensure the effect and accuracy of subsequent edge enhancement, and then an edge enhancement filter, such as the Sobel operator or the Laplace operator, is applied to highlight the edges in the image. Through the convolution operation of the operator, the gradient strength and direction of the image are calculated to enhance the edge features of the surface. The operations are performed on separate frames without involving comparison between frames, ensuring that the surface edges in each frame are clearly highlighted. Through the enhanced edges, the structure and boundaries of the surface can be seen more clearly to obtain an edge-enhanced image.

[0070] S102: Analyze the continuous image frames based on the edge enhanced image, identify the pixel displacement and change degree by measuring the edge position change and difference between adjacent frames, and obtain pixel change data;

[0071] Based on edge-enhanced images, continuous image frames are analyzed. The edge positions in continuous frames are accurately tracked through image processing software or algorithms, such as optical flow or frame difference methods, and the relative displacement of edges between adjacent frames is calculated, including position comparison of edge pixels and determination of motion vectors. These data can be used to quantify the movement speed and direction of each edge pixel between continuous frames. By comparing and analyzing the statistical data of these edge movements, it is possible to identify which pixels have undergone significant position changes. These position change data are recorded and summarized for anomaly detection and analysis.

[0072] S103: Based on the pixel change data, a change threshold is set, and areas exceeding the change threshold are screened out, and the areas are marked as locations where the surface changes significantly, thereby obtaining basic anomaly detection results;

[0073] Based on the pixel change data, the collected pixel change data is screened through programming methods or using the functions in the image processing software. The threshold is set based on previous experimental data or statistical analysis to determine which changes exceed the normal range of surface changes, and the pixels exceeding the threshold are marked as significant change areas. By comparing the change data and the threshold, the locations on the surface with significant changes can be automatically identified and marked. These marked areas are highlighted in the image or coded with different colors, so that the areas of significant changes are clear at a glance, and basic anomaly detection results are obtained.

[0074] See also Figure 3 ,Based on the basic anomaly detection results, the marked surface area is gridded, and the boundary and geometric information of the abnormal area are extracted through edge detection, and the abnormal area is separated from the background image. The specific steps to obtain the segmented abnormal area data are as follows:

[0075] S201: Based on the basic anomaly detection result, each detected abnormal area is gridded, and the abnormal area is divided according to the size and boundary of each grid to obtain gridded area data;

[0076] Based on the basic anomaly detection results, the identified anomaly areas are analyzed and gridded using geographic information system software, including selecting appropriate grid size and shape, usually square or rectangular. The size of each grid is set according to the resolution of the actual monitoring area and the range of the anomaly area. The process is carried out through the automation tools of GIS software to ensure that each anomaly area is evenly divided into multiple grids to facilitate subsequent analysis and processing. After the gridding process is completed, the generated data includes the location, size and anomaly information contained in each grid. The data is stored as gridded area data for subsequent analysis.

[0077] S202: Based on the gridded area data, edge tracking technology is applied to the data in each grid to extract the edge lines and morphological features of each grid, divide the boundaries between abnormal and non-abnormal areas, and obtain geometric boundary data;

[0078] Based on the gridded area data, edge tracing processing is performed on the image data within each grid. The edge detection algorithm in the image processing software, such as the Canny edge detector, is used to identify and extract the edge lines in each grid. This includes calculating the gradient of the image and applying a threshold to determine which areas belong to the edge. Then, through morphological operations such as dilation and erosion, the edges are clarified to make the boundaries more obvious. These edge lines are analyzed to distinguish the boundaries between abnormal and non-abnormal areas. The geometric shape and features of each grid are calculated through the edge data, and geometric boundary data containing these edges are generated. These data can clearly show the exact location and shape of the abnormal area.

[0079] S203: based on the geometric boundary data, spatially separate the image in the abnormal area from the background image, isolate each abnormal area, and obtain segmented abnormal area data;

[0080] Based on the geometric boundary data, image segmentation technology is used to spatially separate each marked abnormal area from the surrounding background image. Using an image segmentation algorithm based on boundary data, such as the level set method or the region growing-based algorithm, the algorithm can determine the segmentation line according to the geometric boundary, isolate the abnormal area, and make each abnormal area independent in the image. The segmented data includes the image data of each independent abnormal area. The data is refined and stored as segmented abnormal area data for subsequent detailed analysis and recording.

[0081] See also Figure 4 Based on the segmented abnormal area data, by comparing the brightness value and color distribution of pixels in the abnormal area with the statistical characteristics of the surrounding area, the surface change characteristics of the abnormal area are quantified. The specific steps for obtaining the surface abnormal characteristic data are as follows:

[0082] S301: Based on the segmented abnormal region data, pixel scanning technology is used to analyze the brightness values ​​of pixels in the abnormal region one by one, and color information of each pixel is collected to obtain brightness and color analysis data;

[0083] Based on the data of the segmented abnormal area, start the image processing software to perform pixel scanning. For each segmented abnormal area, scan each pixel one by one, use image processing algorithms such as grayscale conversion algorithm to analyze the brightness value of each pixel, and use color extraction technology to collect the RGB color information of each pixel. The information includes the values ​​of the red, green, and blue color channels. Through this method, the detailed brightness and color data of the pixels in each abnormal area can be accurately recorded. After integrating the data, it is stored as brightness and color analysis data. The data reflects the visual characteristics inside the abnormal area and provides a basis for subsequent comparative analysis.

[0084] S302: Based on the brightness and color analysis data, compare the pixel index of the abnormal area with the pixel index of the surrounding non-abnormal area, evaluate the statistical difference between the images, identify the significant features of the abnormal area, and obtain significant feature data;

[0085] Based on the brightness and color analysis data, a comparative analysis of pixel indicators is performed. Statistical analysis software is used to compare the pixel indicators in the abnormal area with those in the surrounding non-abnormal areas. The analysis includes brightness differences and color differences. By calculating statistical differences, such as mean values ​​and standard deviations, the significant differences between the two are evaluated. Statistical data help identify the characteristics of abnormal areas, such as specific color deviations or brightness anomalies. Based on the statistical analysis results, the prominent feature data are sorted out to obtain significant feature data, which reflects the significant visual attributes of each abnormal area.

[0086] S303: Based on the significant feature data, by comparing with the set parameter threshold, the features representing the surface changes are identified to obtain surface abnormal feature data;

[0087] Based on the significant feature data, feature comparison is performed through the set parameter threshold. Specific thresholds are set using image processing and data analysis software. The thresholds are determined based on previous studies or statistical analysis to distinguish normal from abnormal surface features. Each indicator in the significant feature data is checked and compared with the threshold. Features that exceed the threshold are identified. Features that exceed the threshold are considered to be key indicators of surface changes. The key features are summarized to generate surface anomaly feature data. The data describes in detail the key visual features representing surface changes in each abnormal area.

[0088] See also Figure 5 Based on the surface anomaly feature data, the surface feature value range in historical geological disaster events is called, the anomaly features are matched and analyzed, and the anomaly type corresponding to the abnormal area is identified. The specific steps for obtaining the anomaly type identification result are as follows:

[0089] S401: Based on the surface anomaly feature data, access the historical geological disaster database, extract the historical data range of similar surface features, collect historical surface feature values, and obtain a historical surface feature reference data set;

[0090] Based on the surface anomaly feature data, the data access process is initialized and connected to a dedicated historical geological disaster database. Using a database query language such as SQL, a query is constructed based on the currently identified surface anomaly feature data to extract the feature data of the historical geological events that match it, including the historical numerical range of surface color, texture, shape and other features. Data retrieval and extraction are performed through a database management system to ensure that the collected data specifically covers the characteristic information of the corresponding geological period. The historical surface feature values ​​are summarized and formed into a reference data set for subsequent comparative analysis. The historical surface feature reference data set collected in this way records in detail the parameters of historical geological events similar to the current anomaly features.

[0091] S402: Based on the historical surface feature reference data set, the surface features of the current abnormal area are compared and analyzed with the historical data, and the matching feature combinations are identified using statistical matching technology to obtain matching surface feature data;

[0092] Based on the historical surface feature reference data set, a data comparison and analysis process is set up, and statistical matching techniques, such as multivariate statistical analysis, are used to conduct a detailed comparison of the current abnormal area characteristics with the characteristics in the historical data set. The process involves calculating the correlation and similarity between the current surface characteristics and the historical characteristics, and evaluating their statistical significance. Through this technology, historical surface feature combinations that match the current surface anomaly can be automatically identified. These matching data reflect the relationship between the current abnormal characteristics and historical events. The obtained matching surface feature data provides accurate comparison results for subsequent analysis.

[0093] S403: Based on the matching surface feature data, the association between each surface feature and the known geological disaster type is analyzed, the matching anomaly type is identified, and an anomaly type identification result is obtained.

[0094] Based on matching surface feature data, feature association analysis is performed. Using geological analysis tools and data analysis software, the matched surface features are correlated and compared with the database of known geological disaster types. In this process, the relationship between feature data and disaster type is determined through algorithm analysis, such as logistic regression or decision tree classification, to identify which feature combinations are most common in specific types of geological disasters. This method can automatically identify the disaster type that the current abnormal area may correspond to. These analysis results are organized into abnormal type identification results, which specify in detail the geological disaster type that is most likely to correspond to each abnormal area.

[0095] See also Figure 6Based on the anomaly type identification results, the change trend of the surface characteristic parameters of the potential disaster area is analyzed, and the displacement value, morphological evolution value and regional expansion rate are extracted according to the time series. Combined with the geological, meteorological and environmental parameters of the target area, the development trend range and intensity of the potential disaster are evaluated. The specific steps for obtaining the disaster development trend prediction data are as follows:

[0096] S501: Based on the anomaly type identification result, the surface characteristic parameters of the potential disaster area are extracted, and the displacement value, morphological evolution value and expansion rate in the target area within a period of time are extracted by using the time series analysis method to obtain the surface change time series data;

[0097] Based on the anomaly type identification results, the specific potential disaster areas listed in the anomaly type identification results are determined. For these areas, remote sensing technology and ground monitoring data are used to extract surface characteristic parameters. Next, time series analysis methods, such as autoregressive moving average model or seasonal decomposition time series prediction, are applied to extract and analyze the changes in these surface characteristics over a period of time, including displacement values, morphological evolution and expansion rate. The data are converted into continuous data points through time series analysis to depict the dynamic change trend of the surface. The obtained surface change time series data exhaustively records the dynamic changes of the surface in the target area.

[0098] S502: Based on the time series data of surface changes, combined with the geological, meteorological and environmental parameters of the region, trend analysis is performed by adjusting parameter weights to obtain trend analysis results;

[0099] The formula for trend analysis is:

[0100]

[0101] Among them, D t represents the disaster development index at the target time point t, M t represents the displacement value at time point t, E t represents the morphological evolution value at time point t, S t represents the expansion rate at time point t, V t Represents the current environment variable, V max is the maximum recorded value of the environment variable, and a, b, c, and d are weight coefficients.

[0102] Parameter details and how to obtain them

[0103] D t : The disaster development index at the target time point t is obtained through time series analysis and historical trend matching, reflecting the disaster development potential predicted at that time point.

[0104] M t: The displacement value at time point t is directly measured by ground displacement sensors or remote sensing equipment during regular monitoring, indicating the relative movement of the surface at that time point.

[0105] E t : The morphological evolution value at time point t is obtained by processing the surface pictures collected from the monitoring equipment through image analysis technology, which quantifies the change rate of the surface morphology.

[0106] S t : The expansion rate at time point t is obtained by differential calculation of the displacement data at consecutive time points, showing the diffusion speed of the disaster-affected area.

[0107] V t :Current environmental variables, such as temperature or humidity, are acquired through real-time data from environmental monitoring sites to reflect the current environmental status.

[0108] V max : The maximum recorded value of the environment variable, obtained from past environmental data, is used to normalize the value of the current environment variable.

[0109] a, b, c and d: weight coefficients, estimated based on the regression model of historical data, reflecting the influence weights of displacement value, morphological evolution value, expansion rate and environmental variables in the disaster prediction model respectively.

[0110] Calculation example:

[0111] Assume that at a certain time point t, there are the following data and coefficient settings: M t =10m(displacement), E t =0.3(morphological evolution rate), S t =5m / year (expansion rate), V t =25℃(current temperature), V max =50°C (highest temperature in history), coefficients a=0.5, b=0.3, c=0.15, d=0.05.

[0112] Normalize environment variables:

[0113]

[0114] Calculation of Disaster Development Index D t :

[0115] D t =0.5·10+0.3·0.3+0.15·5+0.05·0.5

[0116] =5+0.09+0.75+0.025

[0117] =5.865

[0118] The disaster development index D t =5.865 reflects the disaster development potential of the region at time point t under given displacement, morphological evolution, expansion rate and environmental conditions. t The value indicates a higher risk of disaster development, which is mainly affected by the displacement value, followed by the small contribution of the expansion rate and environmental variables. This analysis helps decision makers and relevant departments to take preventive measures to mitigate potential disaster risks.

[0119] S503: Based on the trend analysis results, evaluate the development trend scope and intensity of potential disasters, identify the disaster impact scope and the probability of disaster occurrence, and obtain disaster development trend prediction data;

[0120] Based on the results of trend analysis, an assessment of disaster development trends is conducted. Using geological models and disaster risk assessment tools, such as tools for calculating disaster risk indices or simulating disaster development scenarios, the strength of association between each surface feature and known disaster types is analyzed, and the impact range and probability of occurrence of each potential disaster are identified. The process involves algorithm calculations and model verification to ensure the accuracy of the analysis results. The resulting disaster development trend forecast data provides decision makers with important information on the potential impact of disasters, including the possible impact range, intensity and time probability of occurrence of disasters.

[0121] See also Figure 7 Based on the disaster development trend prediction data, according to the combined geological, climatic and environmental parameters of the target area, the abnormal threshold of the target area is adjusted, the geological anomaly level in the target area is evaluated, the corresponding notification is triggered, and the warning is pushed to the relevant departments. The specific steps to obtain the abnormal notification result are:

[0122] S601: Based on the disaster development trend prediction data, the geological structure, climate conditions and environmental background of the target area are analyzed, the anomaly detection threshold of the area is adjusted, the sensitivity and accuracy of anomaly detection are optimized, and the adjusted threshold data is obtained;

[0123] The formula for adjusting the anomaly detection threshold of a region is:

[0124]

[0125] Among them, T new is the adjusted anomaly detection threshold, T base represents the baseline anomaly detection threshold, P geo Represents the value of geological structural parameters, P max Represents the maximum preset value of the geological structure parameter, T climate Represents the current value of the climate condition data, W base Represents the baseline climate condition data, T envRepresents the current value of the environmental background data, H base represents the baseline environmental background data, w1, w2 and w3 are the weight coefficients of geological parameters, climate conditions and environmental background for threshold adjustment respectively.

[0126] Parameters

[0127] T base :The baseline anomaly detection threshold is the starting threshold determined based on historical data and statistical analysis. It is obtained from historical disaster data by statistically analyzing the average anomaly index value when disasters occurred in the past.

[0128] P geo :The numerical value of geological structure parameters, real-time data obtained from geological monitoring equipment such as seismographs or geological radars, reflecting geological stability.

[0129] P max : The maximum preset value of the geological structure parameter, the maximum possible value set based on the geologist's evaluation, used to normalize the geological parameters.

[0130] T climate :The current value of climate condition data is obtained through data provided by the weather station, such as current temperature, humidity, etc. These data are obtained in real time through the public meteorological data interface.

[0131] W base :Baseline climate condition data, based on the assessment set by geologists, is generally the maximum value of regional climate condition data and is used to assess the impact of current climate condition data.

[0132] T env :The current value of environmental background data, which means the data obtained from environmental monitoring, such as vegetation coverage. These data are usually collected by environmental monitoring sites and transmitted in real time through the network.

[0133] H base :Baseline environmental background data, based on the assessment settings of geologists, is generally the maximum value of the regional environmental background data and is used to assess the impact of environmental background data.

[0134] w1, w2, w3: weight coefficients, which are set by experts based on the degree of influence of different parameters on the sensitivity of anomaly detection, and are used to adjust the proportion of each influencing factor in the total threshold calculation.

[0135] Calculation example:

[0136] Assume the specific value is: T base =100, P geo =80, P max =100, T climate =30,W base=50, T env =50, H base =80, and the weight coefficients are w1=0.3, w2=0.2, and w3=0.5 respectively.

[0137] Calculate the influence of geological structural parameters on the threshold:

[0138]

[0139] Calculate the effect of climate conditions on the threshold:

[0140]

[0141] Calculate the impact of environmental background on the threshold:

[0142]

[0143] Overall Threshold Adjustment:

[0144] T new =100+0.24+0.9+0.3125≈101.45

[0145] The calculated adjusted anomaly detection threshold is 101.45, indicating that the sensitivity and accuracy of anomaly detection are optimized to more accurately predict the likelihood of potential disasters under the given geological, climatic, and environmental backgrounds.

[0146] S602: Based on the adjusted threshold data, the geological anomaly level in the target area is evaluated, and high-risk areas are identified and marked to obtain risk level evaluation data;

[0147] Based on the adjusted threshold data, the geological anomaly level in the area is assessed, and the data of each area under the adjusted threshold is automatically analyzed to identify areas with higher risks. The identification of areas is based on the frequency and intensity of abnormal data exceeding the new threshold. Each area is automatically labeled and classified according to the risk level, such as high, medium and low risk areas. The risk level assessment data is generated by data processing software and is used to update regional safety maps and risk warnings.

[0148] S603: Based on the risk level assessment data, trigger the corresponding notification, send warning information to relevant departments, and obtain abnormal notification results;

[0149] Based on the risk level assessment data, an automatic notification system is set up to trigger an alarm, and early warning information is sent to relevant departments through integrated communication platforms such as email and SMS services. The notification includes the specific location of the risk area, risk level and emergency response recommendations. The process ensures that all key information is promptly communicated to the departments that need to take action, thereby accelerating the response process and assisting in reducing the potential impact of disasters, and obtaining abnormal notification results, which are used for records and subsequent emergency management evaluations.

[0150] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0151] In the present invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0152] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0153] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0154] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0155] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0156] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0157] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0158] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0159] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for satellite remote sensing monitoring, early warning and forecasting of geological disasters, characterized in that: The following steps are involved: Based on ground remote sensing images, the edge information of the image is extracted frame by frame, the pixel difference value and spatial displacement information are identified, the surface area where changes occur is marked, and the basic anomaly detection results are obtained; Based on the basic anomaly detection result, the marked surface area is gridded, and the anomaly area is separated from the background image through edge detection to obtain segmented anomaly area data; Based on the segmented abnormal region data, by comparing the brightness values ​​and color distribution of pixels in the abnormal region with the statistical characteristics of the surrounding area, the surface change characteristics of the abnormal region are quantified to obtain surface abnormality characteristic data; Based on the surface abnormal feature data, the surface feature value range in historical geological disaster events is called, the abnormal features are matched and analyzed, the abnormal type corresponding to the abnormal area is identified, and the abnormal type identification result is obtained; Based on the anomaly type identification results, the change trend analysis of the surface characteristic parameters of the potential disaster area is carried out to evaluate the development trend range and intensity of the potential disaster and obtain disaster development trend prediction data; Based on the disaster development trend prediction data, according to the combined geological, climatic and environmental parameters of the target area, the abnormal threshold of the target area is adjusted, the geological anomaly level in the target area is evaluated, and early warnings are pushed to relevant departments to obtain abnormal notification results.

2. The method for satellite remote sensing monitoring, early warning and forecasting of geological disasters according to claim 1 is characterized in that: The basic anomaly detection results include the marked change area position, pixel change value and boundary range of the change area; the segmented abnormal area data include boundary information after area division and abnormal area image after separation; the surface abnormality feature data include brightness distribution characteristics, color change parameters within the area and spatial distribution consistency information; the anomaly type identification results include classified anomaly types, associated historical feature value ranges and verification results of anomaly classification; the disaster development trend prediction data include predicted regional expansion range, intensity level and time series change trend value; the anomaly notification results include regional graded notification information, triggered graded notification list and associated real-time push channel information.

3. The method for satellite remote sensing monitoring, early warning and forecasting of geological disasters according to claim 1 is characterized in that: Based on the ground remote sensing image, the edge information of the image is extracted frame by frame, the pixel difference value and spatial displacement information are identified, and the changed surface area is marked. The specific steps to obtain the basic anomaly detection results are as follows: Based on ground remote sensing images, image data is read frame by frame, and edge enhancement filtering is applied to highlight the surface edge in each frame, optimize the edge clarity and continuity, and obtain edge enhanced images; Based on the edge enhanced image, continuous image frames are analyzed, and pixel displacement and change degree are identified by measuring edge position changes and differences between adjacent frames to obtain pixel change data; Based on the pixel change data, a change threshold is set, and areas exceeding the change threshold are screened out. The marked areas are locations where the surface changes significantly, and basic anomaly detection results are obtained.

4. The method for satellite remote sensing monitoring, early warning and forecasting of geological disasters according to claim 1, characterized in that: Based on the basic anomaly detection results, the marked surface area is gridded, and the abnormal area is separated from the background image by edge detection to obtain the segmented abnormal area data. Specifically, the steps are as follows: Based on the basic anomaly detection results, each detected abnormal area is gridded, and the abnormal area is divided according to the size and boundary of each grid to obtain gridded area data; Based on the gridded area data, edge tracking technology is applied to the data in each grid to extract edge lines and morphological features of each grid, divide the boundaries between abnormal and non-abnormal areas, and obtain geometric boundary data; Based on the geometric boundary data, the image in the abnormal region is spatially separated from the background image, and each abnormal region is isolated to obtain segmented abnormal region data.

5. The method for satellite remote sensing monitoring, early warning and forecasting of geological disasters according to claim 1, characterized in that: Based on the segmented abnormal area data, by comparing the brightness value and color distribution of pixels in the abnormal area with the statistical characteristics of the surrounding area, the surface change characteristics of the abnormal area are quantified, and the steps of obtaining the surface abnormal characteristic data are specifically as follows: Based on the segmented abnormal region data, pixel scanning technology is used to analyze the brightness values ​​of pixels in the abnormal region one by one, and color information of each pixel is collected to obtain brightness and color analysis data; Based on the brightness and color analysis data, the pixel index of the abnormal area is compared with the pixel index of the surrounding non-abnormal area, the statistical difference between the images is evaluated, the significant features of the abnormal area are identified, and the significant feature data is obtained; Based on the significant feature data, by comparing with the set parameter threshold, the features representing surface changes are identified to obtain surface abnormal feature data.

6. The method for satellite remote sensing monitoring, early warning and forecasting of geological disasters according to claim 1, characterized in that: Based on the surface abnormal feature data, the numerical range of surface features in historical geological disaster events is called, the abnormal features are matched and analyzed, and the abnormal type corresponding to the abnormal area is identified. The specific steps of obtaining the abnormal type identification result are as follows: Based on the surface abnormality feature data, access the historical geological disaster database, extract the historical data range of similar surface features, collect historical surface feature values, and obtain a historical surface feature reference data set; Based on the historical surface feature reference data set, the surface features of the current abnormal area are compared and analyzed with the historical data, and the matching feature combinations are identified using statistical matching technology to obtain matching surface feature data; Based on the matching surface feature data, the association between each surface feature and the known geological disaster type is analyzed, the matching anomaly type is identified, and an anomaly type identification result is obtained.

7. The method for satellite remote sensing monitoring, early warning and forecasting of geological disasters according to claim 1, characterized in that: Based on the anomaly type identification result, the change trend analysis of the surface characteristic parameters of the potential disaster area is carried out to evaluate the development trend range and intensity of the potential disaster, and the steps of obtaining the disaster development trend prediction data are as follows: Based on the anomaly type identification results, the surface characteristic parameters of the potential disaster area are extracted, and the displacement value, morphological evolution value and expansion rate in the target area over a period of time are extracted by using a time series analysis method to obtain the surface change time series data; Based on the surface change time series data, combined with the regional geological, meteorological and environmental parameters, trend analysis is performed by adjusting parameter weights to obtain trend analysis results; Based on the trend analysis results, the development trend scope and intensity of potential disasters are evaluated, the disaster impact scope and the probability of disaster occurrence are identified, and disaster development trend prediction data are obtained.

8. The method for satellite remote sensing monitoring, early warning and forecasting of geological disasters according to claim 7 is characterized in that: The formula for trend analysis is: Among them, D t represents the disaster development index at the target time point t, M t represents the displacement value at time point t, E t represents the morphological evolution value at time point t, S t represents the expansion rate at time point t, V t Represents the current environment variable, V max is the maximum recorded value of the environment variable, and a, b, c, and d are weight coefficients.

9. The method for satellite remote sensing monitoring, early warning and forecasting of geological disasters according to claim 1, characterized in that: Based on the disaster development trend prediction data, according to the combined geological, climatic and environmental parameters of the target area, the abnormal threshold of the target area is adjusted, the geological anomaly level in the target area is evaluated, and an early warning is pushed to the relevant departments. The specific steps for obtaining the abnormal notification result are: Based on the disaster development trend prediction data, the geological structure, climate conditions and environmental background of the target area are analyzed, the anomaly detection threshold of the area is adjusted, the sensitivity and accuracy of anomaly detection are optimized, and the adjusted threshold data is obtained; Based on the adjusted threshold data, the geological anomaly level in the target area is evaluated, high-risk areas are identified and marked, and risk level evaluation data is obtained; Based on the risk level assessment data, a corresponding notification is triggered, and an early warning message is sent to relevant departments to obtain an abnormal notification result.

10. The method for satellite remote sensing monitoring, early warning and forecasting of geological disasters according to claim 9, characterized in that: The formula for the abnormal detection threshold of the adjustment area is: Among them, T new is the adjusted anomaly detection threshold, T base represents the baseline anomaly detection threshold, P geo Represents the value of geological structural parameters, P max Represents the maximum preset value of the geological structure parameter, T climate Represents the current value of the climate condition data, W base Represents the baseline climate condition data, T env Represents the current value of the environmental background data, H base represents the baseline environmental background data, w1, w2 and w3 are the weight coefficients of geological parameters, climate conditions and environmental background for threshold adjustment respectively.

Images