Geological disaster activity target monitoring method based on dynamic visual tracking

Abstract

The invention relates to the technical field of geological disaster monitoring, in particular to a geological disaster activity target monitoring method based on dynamic visual tracking, which comprises the following steps of: adjusting multiple shooting parameters of a camera according to illumination intensity difference in an environment based on surface image acquisition equipment, and recording time points and geographical location information of image shooting in real time; a sequence of geo-tagged images is generated. According to the method, camera parameters are adjusted according to real-time illumination conditions, the quality of surface images is optimized, the monitoring and recognition capability of surface cracks and deformation is improved, and data support is provided for prediction and type recognition of geological disasters through multi-time-point image analysis and calculation of displacement speed and sedimentation rate, so that early warning is more accurate and targeted, and the early warning efficiency is improved. According to the method, the monitoring data and the topographic features of the target area are combined, the disaster type is accurately recognized, the development trend is predicted, dynamic monitoring and pre-judgment of geological disaster activities are achieved, the effectiveness of prevention measures is enhanced, and potential threats to human life and property are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of geological disaster monitoring, and in particular to a method for monitoring geological disaster activity targets based on dynamic visual tracking. Background Art

[0002] The field of geological disaster monitoring technology focuses on using a variety of technical means to observe and analyze physical, chemical and biological changes on the earth's surface and underground, and is applied to landslide activity monitoring, soil and rock displacement observation, groundwater level and water quality change monitoring. It aims to help scientists and engineers understand geological conditions in a timely manner, predict geological disasters, including collapse, landslide, mudslide and ground collapse, and formulate effective preventive measures and emergency response plans, and achieve resource exploration and environmental protection, including mineral resource development, geothermal energy utilization and geological environment protection.

[0003] Among them, the target monitoring method of geological disaster activities focuses on monitoring and predicting various geological disasters, including collapse, landslides and mudslides. It monitors geological activities in real time or near real time, predicts disaster events, and warns relevant areas in advance to reduce the impact of disasters on human life and property. It uses ground sensor networks, drones, cameras and a variety of remote sensing equipment to collect continuous geographic and geological data, monitor surface movement and crack expansion, analyze trends and patterns of geological changes, predict the time and place of disasters, and take measures to reduce the impact of disasters and economic losses.

[0004] Traditional geological disaster activity monitoring methods have limitations in real-time data processing and image quality adjustment. They rely on sensors and camera equipment with fixed parameters, which leads to poor data quality in environments with large changes in light intensity, affecting the accurate detection of cracks and geological changes. The lack of effective real-time image analysis technology limits the immediate identification and monitoring of surface changes, affecting the timeliness and accuracy of disaster warnings, and missing key warning information in rainy weather or at night, leading to increased disaster losses. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a method for monitoring geological disaster activity targets based on dynamic visual tracking.

[0006] In order to achieve the above object, the present invention adopts the following technical scheme: a method for monitoring geological disaster activity targets based on dynamic visual tracking, comprising the following steps:

[0007] S1: Based on the surface image acquisition device, the camera's multiple shooting parameters are adjusted according to the difference in light intensity in the environment, and the time point and geographical location information of the image shooting are recorded in real time to generate a geo-tagged image sequence;

[0008] S2: Based on the geo-tagged image sequence, by analyzing the shape, color and texture distribution in the image, identifying the surface cracks in the image, recording the width and extension direction of the cracks, and by comparing and analyzing multiple images, identifying the location of terrain changes and generating surface feature data;

[0009] S3: Based on the surface feature data, by analyzing the images of the terrain change area at multiple time points, calculating the change in surface elevation and displacement direction, and identifying the displacement speed and settlement rate, generating ground settlement rate information;

[0010] S4: Based on the ground subsidence rate information, identify various geological disaster behaviors according to the location of deformation, the change rate and the terrain characteristics of the target location, and generate a disaster type identification result;

[0011] S5: using the disaster type identification result, by performing time series analysis on the deformation data, predicting the development trend of the target disaster, and generating a disaster development trend analysis result;

[0012] S6: In combination with the above-mentioned disaster development trend analysis results, by analyzing and comparing the characteristics and impacts of various geological disasters, risk levels are calculated for multiple identified geological disasters to generate risk level assessment results.

[0013] As a further solution of the present invention, the geo-tagged image sequence includes timestamp data, geographic location coordinates, and image data acquisition results; the surface feature data includes crack length and direction information, terrain change type information, and a change area marking data set; the ground subsidence rate information includes change position coordinate data, displacement distance measurement data, and calculated subsidence rate; the disaster type identification results include disaster category information, deformation position terrain features, and disaster level prediction data; the disaster development trend analysis results include disaster development path prediction results, disaster development speed calculation results, and affected area identification; the risk level assessment results include disaster impact degree, severity assessment information, and predicted economic losses.

[0014] As a further solution of the present invention, based on the surface image acquisition device, according to the difference in light intensity in the environment, multiple shooting parameters of the camera are adjusted, and the time point and geographical location information of the image shooting are recorded in real time. The steps of generating a geographically marked image sequence are specifically as follows:

[0015] S101: Based on the surface image acquisition device, according to the real-time ambient light level, multiple camera shooting parameters, including shutter and aperture, are adjusted to optimize image clarity and generate a shooting parameter adjustment record;

[0016] S102: Based on the shooting parameter adjustment record, shoot a surface image and record the shooting time of the image to generate a surface image shooting record;

[0017] S103: Based on the surface image shooting record, geographical location information of the shooting area is collected and recorded in real time to generate a geographically marked image sequence.

[0018] As a further solution of the present invention, based on the sequence of geographically marked images, the steps of identifying surface cracks in the images by analyzing the shape, color and texture distribution in the images, recording the width and extension direction of the cracks, and identifying the location of terrain changes by comparing and analyzing multiple images to generate surface feature data are specifically as follows:

[0019] S201: Based on the geographic tag image sequence, identifying and recording surface cracks in the image by analyzing the color and texture in the image, and generating surface crack identification information;

[0020] S202: Based on the surface crack identification information, identify the width and extension direction of the crack, record the geometric shape information of the crack, and generate crack feature identification data;

[0021] S203: Based on the crack feature recognition data, a plurality of images of the target location are compared and analyzed to identify the location of terrain changes, including subsidence and uplift, and generate surface feature data.

[0022] As a further solution of the present invention, based on the surface feature data, by analyzing the images of the terrain change area at multiple time points, calculating the change in surface elevation and displacement direction, and identifying the displacement speed and settlement rate, the steps of generating ground settlement rate information are specifically as follows:

[0023] S301: Based on the surface feature data, by analyzing multiple images of the terrain change area, measuring the elevation data of the deformation position at multiple time points, and generating an elevation data change record;

[0024] S302: Based on the elevation data change record, measuring the coordinates of the surface feature points of the target position at multiple time points, identifying the displacement direction and distance, and generating displacement measurement data;

[0025] S303: Based on the displacement measurement data, the displacement speed and settlement rate of the target deformation position are calculated to generate ground settlement rate information.

[0026] As a further solution of the present invention, the specific formula for calculating the displacement velocity and the sedimentation rate of the target deformation position is:

[0027]

[0028] Wherein, V represents displacement velocity, which indicates the moving rate of the target deformation position per unit time; d represents lateral displacement distance, which is the maximum horizontal displacement measured from continuous image frames; h represents longitudinal displacement height, which is the maximum vertical height change measured from the same series of image frames; t1 represents the time point of the next moment; t0 represents the time point of the previous moment; t1-t0 represents the time interval, which is the time difference obtained by subtracting the time data in the two images; and k represents the terrain adjustment coefficient, which is used to adjust the displacement velocity calculation to adapt to different geological conditions.

[0029] As a further solution of the present invention, based on the ground subsidence rate information, according to the location of deformation, the rate of change and the terrain characteristics of the target location, a plurality of geological disaster behaviors are identified, and the steps of generating a disaster type identification result are specifically as follows:

[0030] S401: Based on the land subsidence rate information, analyzing the topographic characteristics of the surface deformation location, including terrain slope, soil type, and vegetation coverage, to generate deformation location characteristic data;

[0031] S402: Calculating the similarity between the deformation characteristics of the target area and multiple known geological disaster types based on the deformation location feature data and the degree and rate of surface deformation, and generating a similarity calculation result;

[0032] S403: According to the similarity calculation result, the disaster type of the deformation location is evaluated and identified, including landslide and mud-rock flow, and a disaster type identification result is generated.

[0033] As a further solution of the present invention, the disaster type identification result is used to predict the development trend of the target disaster by performing time series analysis on the deformation data. The steps of generating the disaster development trend analysis result are specifically as follows:

[0034] S501: Based on the disaster type identification result, analyzing the change trend of the deformation speed, combining the terrain characteristics of the target location, predicting the development speed of the disaster, and generating development speed prediction data;

[0035] S502: Based on the development speed prediction data, according to the type of disaster and the terrain characteristics of the target location, predict the diffusion path of the target geological disaster and generate a diffusion path prediction result;

[0036] S503: Based on the diffusion path prediction result, predict the impact range of the target disaster and the disaster stop time, and generate a disaster development trend analysis result.

[0037] As a further solution of the present invention, in combination with the disaster development trend analysis results, by analyzing and comparing the characteristics and impacts of multiple geological disasters, the risk level is calculated for multiple identified geological disasters, and the steps of generating the risk level assessment results are specifically as follows:

[0038] S601: Based on the disaster development trend analysis results, and based on the predicted impact range and location information of the target geological disaster, multiple risk areas are identified, including landslide-prone areas, debris flow active areas, and ground subsidence areas, and risk area identification results are generated;

[0039] S602: Analyze and predict the impact of the target disaster on personnel safety, economic activities and infrastructure using the risk area identification result, and generate disaster impact prediction information;

[0040] S603: Based on the disaster impact prediction information, taking into account the expected degree of damage, impact scope, and recovery difficulty, the risk levels of multiple risk geological disasters are calculated to generate a risk level assessment result.

[0041] As a further solution of the present invention, the specific formula for calculating the risk levels of multiple risk geological disasters is:

[0042]

[0043] Among them, R represents the comprehensive risk level, which comprehensively reflects the potential impact of the disaster and the urgency of management. D represents the expected degree of damage, which quantifies the economic losses that may be caused by the disaster and the direct impact on the human living environment. A represents the scope of impact, which covers the breadth of the affected geographical area and the population density. H represents the difficulty of recovery, which reflects the complexity of restoring infrastructure and social order after the disaster. P1 is the weight coefficient of the expected degree of damage, which is used to adjust the influence of the expected degree of damage. P2 is the weight coefficient of the scope of impact, which is used to adjust the importance of the scope of impact. P3 is the weight coefficient of the difficulty of recovery, which is used to adjust the proportion of the difficulty of recovery in the total assessment. These parameters are obtained through expert evaluation and data analysis to ensure the scientificity and adaptability of risk assessment.

[0044] Compared with the prior art, the advantages and positive effects of the present invention are:

[0045] In the present invention, by adjusting the camera parameters according to the real-time lighting conditions, the quality of the surface image is optimized, the detection and identification capabilities of surface cracks and deformations are improved, and the displacement velocity and settlement rate are calculated through multi-time point image analysis, which provides data support for the prediction and type identification of geological disasters, making the early warning more accurate and targeted. By combining the monitoring data and the terrain characteristics of the target area, the disaster type can be accurately identified and the development trend can be predicted, so as to realize the dynamic monitoring of geological disaster activities, enhance the effectiveness of preventive measures, and reduce the potential threat to human life and property. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic diagram of the main steps of the present invention;

[0047] Figure 2 It is a schematic diagram of the refinement of S1 of the present invention;

[0048] Figure 3 It is a schematic diagram of the refinement of S2 of the present invention;

[0049] Figure 4 It is a schematic diagram of the refinement of S3 of the present invention;

[0050] Figure 5 It is a schematic diagram of the refinement of S4 of the present invention;

[0051] Figure 6 It is a detailed schematic diagram of S5 of the present invention;

[0052] Figure 7 It is a detailed schematic diagram of S6 of the present invention. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0054] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0055] See also Figure 1 The present invention provides a technical solution, a method for monitoring geological disaster activity targets based on dynamic visual tracking, comprising the following steps:

[0056] S1: Based on the surface image acquisition device, the camera's multiple shooting parameters are adjusted according to the difference in light intensity in the environment, and the time point and geographical location information of the image shooting are recorded in real time to generate a geo-tagged image sequence;

[0057] S2: Based on the geo-tagged image sequence, the shape, color and texture distribution in the image are analyzed to identify the surface cracks in the image, record the width and extension direction of the cracks, and identify the location of terrain changes by comparing and analyzing multiple images to generate surface feature data;

[0058] S3: Based on the surface feature data, by analyzing the images of the terrain change area at multiple time points, the change of surface elevation and displacement direction are calculated, and the displacement speed and settlement rate are identified to generate ground settlement rate information;

[0059] S4: Based on the ground subsidence rate information, various geological disaster behaviors are identified according to the location of deformation, change rate and terrain characteristics of the target location, and the disaster type identification results are generated;

[0060] S5: Using the disaster type identification results, by performing time series analysis on deformation data, the development trend of the target disaster is predicted and the disaster development trend analysis results are generated;

[0061] S6: Combined with the results of the disaster development trend analysis, by analyzing and comparing the characteristics and impacts of various geological disasters, the risk levels of multiple identified geological hazards are calculated to generate risk level assessment results.

[0062] The geo-tagged image sequence includes timestamp data, geographic location coordinates, and image data acquisition results. The surface feature data includes crack length and direction information, terrain change type information, and change area marking data set. The ground subsidence rate information includes change position coordinate data, displacement distance measurement data, and calculated subsidence rate. The disaster type identification results include disaster category information, deformation location terrain characteristics, and disaster level prediction data. The disaster development trend analysis results include disaster development path prediction results, disaster development speed calculation results, and affected area identification. The risk level assessment results include disaster impact degree, severity assessment information, and predicted economic losses.

[0063] See also Figure 2 Based on the surface image acquisition device, according to the difference in light intensity in the environment, the camera's multiple shooting parameters are adjusted, and the time point and geographical location information of the image shooting are recorded in real time. The specific steps for generating a geo-tagged image sequence are as follows:

[0064] S101: Based on the surface image acquisition device, according to the real-time ambient light level, multiple camera shooting parameters, including shutter and aperture, are adjusted to optimize image clarity and generate a shooting parameter adjustment record;

[0065] In sub-step S101, based on the real-time monitoring data of the ambient light level, a sensor is used to measure the current light intensity, and the measured light intensity value is converted into an electrical signal. The signal is input into the central processing unit after analog-to-digital conversion. The central processing unit dynamically calculates the appropriate shutter speed and aperture size based on a preset relationship model between light intensity and shooting parameters, and uses linear regression analysis to ensure that the calculation results accurately correspond to the actual situation of the ambient light. The adjusted shooting parameters can adapt to different lighting conditions and optimize the clarity and exposure level of the image. A shooting parameter adjustment record containing all adjustment records is generated, and the record lists the specific value and adjustment time of each parameter adjustment to ensure the needs of subsequent tracking and quality control.

[0066] S102: adjusting the record based on the shooting parameters, shooting a surface image and recording the shooting time of the image, and generating a surface image shooting record;

[0067] In sub-step S102, based on the shooting parameter adjustment record, during the surface image shooting process, a suitable shutter speed and aperture setting are selected to shoot the surface image. After shooting, the camera records the specific shooting time of the image. The time record is synchronized with the clock to ensure accuracy. The generated image is stored together with the shooting time in the surface image shooting record. The record includes the image file, the image shooting setting parameters and the actual shooting time, which provides necessary data support for subsequent image processing and analysis, ensuring the accurate recording of each operation step and the integrity of the image data.

[0068] S103: Based on the surface image shooting record, geographical location information of the shooting area is collected and recorded in real time to generate a geographically marked image sequence;

[0069] In sub-step S103, based on the surface image shooting record, GPS is used to collect the geographic location information of the shooting area in real time. The GPS module receives satellite signals to determine the longitude and latitude coordinates of the current location. The target coordinate data is recorded in the geographic tagged image sequence together with the image to ensure that the geographic location of each image is accurately marked, providing key data for the geographic information integration of subsequent images. During the process, the GPS signal strength is monitored in real time to ensure the high accuracy and low error rate of the geographic location information. The generated geographic tagged image sequence is convenient for subsequent geographic analysis and map creation, providing accurate geographic data support for scientific research or commercial purposes.

[0070] See also Figure 3 ,Based on the sequence of geo-tagged images, we analyze the shape, color and texture distribution in the image, identify the surface cracks in the image, record the width and extension direction of the cracks, and identify the location of terrain changes by comparing and analyzing multiple images. The specific steps of generating surface feature data are as follows:

[0071] S201: Based on the geographically tagged image sequence, by analyzing the color and texture in the image, identifying and recording the surface cracks in the image, and generating surface crack identification information;

[0072] In sub-step S201, the geo-tagged image sequence is analyzed, and the image processing algorithm is applied to conduct an in-depth analysis of the color and texture in the image. The convolutional neural network is used to enhance the feature recognition of surface cracks in the image. The technology automatically extracts the texture and edge information of the cracks through cascaded filters. During the crack recognition process, the color segmentation technology is used to determine the areas of different colors in the image, and compared with the trained model, the location and contour of the surface cracks are accurately marked, the detailed information of the target cracks is recorded, and the surface crack identification information file is generated. The file includes the precise location, length and various related attributes of the cracks. The data provides the necessary basic information for subsequent steps to ensure the integrity and traceability of the data.

[0073] S202: Based on the surface crack identification information, identify the width and extension direction of the crack, record the geometric shape information of the crack, and generate crack feature identification data;

[0074] In sub-step S202, the crack width and extension direction measurement algorithm is used to process the surface crack identification information, including edge detection and geometric analysis technology. The Canny edge detection algorithm is used to identify the crack boundary in the image, the Hough transform is used to determine the geometric shape and direction of the crack, and the formula is used to determine the crack width and extension direction. Calculate the average width of the cracks and generate crack feature identification data, which includes the length, width, direction and shape description of each crack, providing detailed crack characteristics for geological analysis and monitoring;

[0075] Among them, w represents the crack width, n represents the number of crack boundary points in the image, (x 1i ,y 1i ) and (x 2i ,y 2i ) are the starting and ending coordinates of the crack boundary, and the average width of each crack in the image is calculated;

[0076] Detailed explanation of the formula and the process of formula calculation and derivation:

[0077] Assume that a crack is identified by 5 boundary point pairs in the image, with coordinates of (10,20) and (10,25), (15,20) and (15,25), (20,20) and (20,25), (25,20) and (25,25), (30,20) and (30,25), and substitute them into the formula for calculation:

[0078]

[0079] The results show that the average width of the target fractures is 5 units. The data reflects the average width of the fractures and is used to assess fracture development and geological risks.

[0080] S203: Based on the crack feature recognition data, by comparing and analyzing multiple images of the target location, the location of terrain changes, including subsidence and uplift, is identified, and surface feature data is generated;

[0081] In sub-step S203, crack feature recognition data is used to perform comparative analysis of multiple images of the target location, and image registration and difference analysis techniques are applied, including the use of feature matching algorithms, such as scale-invariant feature transforms or accelerated robust features, to align image sequences, analyze terrain changes, such as subsidence or uplift, and identify terrain changes by comparing the differences in image data at consecutive time points. The generated surface feature data includes the type, degree and precise location of deformation. The target data provides real-time updated surface status for geological disaster warning and geographic information, ensuring that relevant departments respond quickly and take necessary preventive measures.

[0082] See also Figure 4 Based on the surface feature data, by analyzing the images of the terrain change area at multiple time points, calculating the change in surface elevation and displacement direction, and identifying the displacement speed and settlement rate, the steps for generating ground settlement rate information are as follows:

[0083] S301: Based on the surface feature data, by analyzing multiple images of the terrain change area, measuring the elevation data of the deformation position at multiple time points, and generating an elevation data change record;

[0084] In sub-step S301, multiple images of the terrain change area are analyzed, and stereo photogrammetry technology is used to measure the elevation data of the deformation location at multiple time points. By comparing the image sets taken at the same location at different times, the common feature points in the images are used to calculate the elevation difference, and the digital surface model is used to construct the surface height model at each time point. The target model enables the changes in elevation data to be accurately recorded and quantified. The generated elevation data change record lists the time series elevation values ​​of each measuring point, providing key data for deformation analysis and an accurate numerical basis for geologists and engineers to evaluate the trend and speed of surface changes.

[0085] S302: Based on the elevation data change record, measure the coordinates of the surface feature points of the target location at multiple time points, identify the displacement direction and distance, and generate displacement measurement data;

[0086] In sub-step S302, based on the elevation data change records, the coordinates of the surface feature points are measured. By capturing the coordinates of multiple time points, the target coordinate data is analyzed to identify the displacement direction and distance. By comparing and analyzing the coordinate changes of consecutive time points, the movement trend of the surface feature points is calculated. The target data is integrated into displacement measurement data. Each data item records the direction, distance and relationship of the displacement with time, ensuring the reliability and accuracy of the monitoring data and providing key quantitative information for geological risk assessment.

[0087] S303: Calculating the displacement velocity and settlement rate of the target deformation position based on the displacement measurement data to generate ground settlement rate information;

[0088] The specific formula for calculating the displacement velocity and sedimentation rate of the target deformation position is:

[0089]

[0090] Wherein, V represents displacement velocity, which indicates the moving rate of the target deformation position per unit time; d represents lateral displacement distance, which is the maximum horizontal displacement measured from continuous image frames; h represents longitudinal displacement height, which is the maximum vertical height change measured from the same series of image frames; t1 represents the time point of the next moment; t0 represents the time point of the previous moment; t1-t0 represents the time interval, which is the time difference obtained by subtracting the time data in the two images; and k represents the terrain adjustment coefficient, which is used to adjust the displacement velocity calculation to adapt to different geological conditions.

[0091] formula:

[0092]

[0093] Detailed explanation of the formula and the process of formula calculation and derivation:

[0094] The formula is used to calculate the ground displacement velocity, which represents the moving rate of the target deformation position per unit time, and is used to assess the dynamics of geological activities and disaster risks;

[0095] Parameter meaning and setting value:

[0096] d represents the lateral displacement distance, which is assumed to be 5 meters, indicating the horizontal movement distance of the surface during the interval between two observations;

[0097] h represents the longitudinal displacement height, which is assumed to be 2 meters, indicating the vertical change of the surface during the interval between two observations;

[0098] t1-t0 represents the time interval. Assume that the two measurements are taken at 12:00 and 13:00, respectively, which means the time interval is 1 hour.

[0099] k is the terrain adjustment factor, which is assumed to be 1.2, reflecting the effect of terrain slope and soil type on velocity calculation;

[0100] Substitute the parameters into the formula for calculation:

[0101]

[0102]

[0103] V≈5.385·1.2;

[0104] V≈6.462;

[0105] Result analysis:

[0106] The result of 6.462 m / h indicates that during the observation period, the average rate of surface deformation was 6.462 meters per hour. The data is used to monitor surface movement, provide early warnings of changes in surface stability in pre-disaster monitoring, help identify surface change trends, and provide key data for disaster warning and risk management.

[0107] See also Figure 5 Based on the ground subsidence rate information, various geological disaster behaviors are identified according to the location, change rate and terrain characteristics of the target location. The specific steps for generating disaster type identification results are as follows:

[0108] S401: Based on the land subsidence rate information, the topographic characteristics of the surface deformation location are analyzed, including terrain slope, soil type, and vegetation coverage, to generate deformation location characteristic data;

[0109] In sub-step S401, basic data on the location of surface deformation is extracted from geographic information. The target data covers multiple terrain features such as terrain slope, soil type and vegetation coverage. Each factor is quantified by using digital elevation model and soil map analysis tools, including soil map parser and vegetation index analysis tools, such as normalized difference vegetation index analysis. The terrain slope is calculated by the elevation data provided by DEM. The calculation formula is the ratio of the vertical distance to the horizontal distance of adjacent elevation points. The soil type is determined by comparing with the soil map database. The vegetation coverage is determined by analyzing the NDVI value in the satellite image. Each parameter is recorded in detail and converted into numerical data that can be used for analysis. The target data is integrated to form a deformation site feature data set. The data set records every terrain feature of the deformation site in detail, which is convenient for in-depth geological disaster analysis.

[0110] S402: Calculate the similarity between the deformation characteristics of the target area and multiple known geological disaster types based on the deformation location feature data and the degree and rate of surface deformation, and generate a similarity calculation result;

[0111] In sub-step S402, based on the previously generated deformation site feature data, a similarity calculation model is used to compare the similarity between the deformation features of the target area and the known geological disaster types in the database. Based on a multi-dimensional feature matching algorithm, such as a cosine similarity algorithm, the algorithm takes into account multiple variables such as deformation degree, rate, terrain slope, soil type, and vegetation coverage. The similarity is evaluated by calculating the cosine values ​​between the variables. The process quantizes the deformation features and the feature vectors of the known disaster types, and calculates the cosine values ​​between the target feature vectors. The feature vectors of each disaster type are stored in a special database. The similarity calculation result provides a similarity score for each possible disaster type, and the target score is subsequently used for the identification and prediction of the disaster type.

[0112] S403: Evaluate and identify the disaster type of the deformation location according to the similarity calculation result, including landslide and debris flow, and generate a disaster type identification result;

[0113] In sub-step S403, based on the similarity calculation results, decision support is used to evaluate and identify disaster types, and they are sorted according to the similarity scores. The disaster type with the highest score is identified as the most likely disaster at the current deformation location. The disaster type is determined based on the similarity score and the risk factors associated with each disaster, such as the historical frequency of occurrence and the possible degree of loss. The process ensures the accuracy and reliability of the identification results. The generated disaster type identification results will list the possible disaster types, the corresponding similarity scores and risk assessments, providing a scientific basis for subsequent disaster warning and prevention strategies.

[0114] See also Figure 6 , using the disaster type identification results, by performing time series analysis on deformation data, the development trend of the target disaster is predicted. The specific steps for generating the disaster development trend analysis results are as follows:

[0115] S501: Based on the disaster type identification result, analyze the change trend of the deformation speed, combine the terrain characteristics of the target location, predict the development speed of the disaster, and generate development speed prediction data;

[0116] In sub-step S501, the disaster type identification result is received. The target result includes the specific type of disaster and related attribute information. Terrain analysis software, such as ArcGIS and QGIS, is used to analyze the terrain characteristics of the target location, including slope, soil type, vegetation status, etc. Based on the target data and disaster type, a dynamic simulation algorithm, such as a cellular automaton model, is used to simulate and analyze the changing trend of the deformation speed. The dynamic simulation takes into account the influencing factors of the terrain on the disaster diffusion rate, such as the slope affecting the debris flow speed. The algorithm calculates the deformation speed at each time point and generates development speed prediction data by integrating the target data. The target data describes in detail the expected disaster development speed and its changing trend, providing a scientific basis for disaster management departments to formulate emergency response measures.

[0117] S502: Based on the development speed prediction data, according to the type of disaster and the terrain characteristics of the target location, predict the diffusion path of the target geological disaster and generate a diffusion path prediction result;

[0118] In sub-step S502, based on the generated development speed prediction data, geographic information is used for diffusion simulation. Combined with the type of geological disasters and terrain characteristics, such as the river basin affecting the landslide diffusion path, geographic statistical models and diffusion algorithms, such as the shortest path algorithm and watershed analysis, are used to predict the disaster diffusion path. In the process, the possibility and path length of each path are calculated, and the possible diffusion path is determined. The results are saved in the diffusion path prediction data. Each path is marked with possible diffusion directions and diffusion-reached areas, providing accurate geographic data support for the formulation of evacuation plans and the setting of early warnings.

[0119] S503: Based on the diffusion path prediction results, predict the impact range of the target disaster and the disaster stop time, and generate disaster development trend analysis results;

[0120] In sub-step S503, based on the diffusion path prediction results, the possible impact range of the disaster and the expected disaster cessation time are analyzed. Statistical analysis models, such as regression analysis and time series analysis, are used to consider the dynamic changes in disaster development and the impact of terrain obstacles on disaster cessation, such as rivers and mountains that may interrupt the flow of debris flows. The target analysis is modeled based on historical disaster data and terrain conditions. The prediction results record in detail the total area that the disaster may affect, the affected population, and the expected duration of the disaster. The generated disaster development trend analysis results provide local governments and rescue organizations with critical time windows and resource allocation guidelines to maximize the efficiency of emergency response and the effectiveness of disaster mitigation measures.

[0121] See also Figure 7, combined with the results of disaster development trend analysis, by analyzing and comparing the characteristics and impacts of various geological hazards, the risk level is calculated for multiple identified geological hazards. The specific steps for generating risk level assessment results are as follows:

[0122] S601: Based on the disaster development trend analysis results, and according to the predicted impact range and location information of the target geological disaster, multiple risk areas are identified, including landslide-prone areas, debris flow active areas, and ground subsidence areas, and risk area identification results are generated;

[0123] In sub-step S601, based on the results of the disaster development trend analysis, various disaster risk areas are identified using geographic information and remote sensing data processing technology, including using digital elevation models to analyze terrain and slopes to determine landslide-prone areas, identifying debris flow active areas through hydrological models and rainfall data, and identifying ground subsidence areas through the results of analyzing long-term ground deformation data. Each identified risk area will be recorded and classified according to its geographical location information and the characteristics of potential disaster occurrence. The generated risk area identification results will serve as the basic data for subsequent disaster prevention and mitigation planning and emergency response.

[0124] S602: Using the risk area identification results, analyze and predict the impact of the target disaster on human safety, economic activities and infrastructure, and generate disaster impact prediction information;

[0125] In sub-step S602, the results of the identified risk areas are used to adopt multi-factor risk assessment models, such as the disaster risk matrix, to analyze the specific impacts that the target disaster may have on human safety, economic activities and infrastructure. Combining demographic data, regional economic conditions and infrastructure distribution maps, the impact of disasters on each risk area is predicted by calculating the number of people that may be affected, economic losses and the degree of infrastructure damage. During the process, the data of each risk area is analyzed in detail and disaster impact prediction information is generated. The target information helps local governments and relevant agencies to conduct disaster warnings and resource allocation, ensuring that quick and effective action can be taken when a disaster occurs.

[0126] S603: Based on the disaster impact prediction information, the risk levels of multiple risk geological disasters are calculated taking into account the expected degree of damage, the scope of impact, and the difficulty of recovery, and a risk level assessment result is generated;

[0127] The specific formula for calculating the risk level of multiple risk geological hazards is:

[0128]

[0129] Among them, R represents the comprehensive risk level, which comprehensively reflects the potential impact of the disaster and the urgency of management. D represents the expected degree of damage, which quantifies the economic losses that may be caused by the disaster and the direct impact on the human living environment. A represents the scope of impact, which covers the breadth of the affected geographical area and the population density. H represents the difficulty of recovery, which reflects the complexity of restoring infrastructure and social order after the disaster. P1 is the weight coefficient of the expected degree of damage, which is used to adjust the influence of the expected degree of damage. P2 is the weight coefficient of the scope of impact, which is used to adjust the importance of the scope of impact. P3 is the weight coefficient of the difficulty of recovery, which is used to adjust the proportion of the difficulty of recovery in the total assessment. These parameters are obtained through expert evaluation and data analysis to ensure the scientificity and adaptability of risk assessment.

[0130] formula:

[0131]

[0132] Detailed explanation of the formula and the formula calculation derivation process: The formula is used to calculate the comprehensive risk level of each risk geological disaster. The result is used to assess the comprehensive impact that the disaster may bring, so as to facilitate the formulation of corresponding response measures.

[0133] Parameter meaning and setting value:

[0134] D is the expected degree of damage, reflecting direct economic losses and potential ecological impacts, and is assumed to be 7, reflecting a higher degree of damage.

[0135] A is the impact area, which covers the number of people affected and the size of the area. It is assumed to be 7, indicating a medium impact area.

[0136] H is the difficulty of recovery, including the time for infrastructure restoration and ecological recovery in the disaster area. It is assumed to be 7, indicating that recovery has a certain degree of difficulty.

[0137] p1, p2, and p3 are the weight coefficients of the corresponding influencing factors, assumed to be 0.4, 0.3, and 0.3.

[0138] Substitute the parameters into the formula for calculation:

[0139]

[0140] Result 6.1 shows that the target geological hazards have an overall risk level of medium to high. The results are used to help decision makers consider resource allocation and emergency response measures to address economic and ecological impacts.

[0141] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for monitoring geological disaster activity targets based on dynamic visual tracking, characterized in that: The following steps are involved: Based on the surface image acquisition equipment, according to the difference in light intensity in the environment, the camera's multiple shooting parameters are adjusted, and the time point and geographical location information of the image shooting are recorded in real time to generate a sequence of geographically tagged images; Based on the geo-tagged image sequence, by analyzing the shape, color and texture distribution in the image, identifying the surface cracks in the image, recording the width and extension direction of the cracks, and by comparing and analyzing multiple images, identifying the location of terrain changes and generating surface feature data; Based on the surface feature data, by analyzing images of the terrain change area at multiple time points, calculating the change in surface elevation and displacement direction, and identifying the displacement speed and settlement rate, generating ground settlement rate information; Based on the ground subsidence rate information, various geological disaster behaviors are identified according to the location of the deformation, the rate of change and the terrain characteristics of the target location, and a disaster type identification result is generated; Using the disaster type identification result, by performing time series analysis on deformation data, the development trend of the target disaster is predicted, and a disaster development trend analysis result is generated; In combination with the above-mentioned disaster development trend analysis results, by analyzing and comparing the characteristics and impacts of various geological disasters, risk levels are calculated for multiple identified geological disasters to generate risk level assessment results.

2. The method for monitoring geological disaster activity targets based on dynamic visual tracking according to claim 1 is characterized in that: The geo-tagged image sequence includes timestamp data, geographic location coordinates, and image data acquisition results; the surface feature data includes crack length and direction information, terrain change type information, and a change area mark data set; the ground subsidence rate information includes change position coordinate data, displacement distance measurement data, and calculated subsidence rate; the disaster type identification result includes disaster category information, deformation position terrain features, and disaster level prediction data; the disaster development trend analysis result includes disaster development path prediction results, disaster development speed calculation results, and affected area identification; the risk level assessment result includes disaster impact degree, severity assessment information, and predicted economic losses.

3. The method for monitoring geological disaster activity targets based on dynamic visual tracking according to claim 1 is characterized in that: Based on the surface image acquisition device, according to the difference in light intensity in the environment, the camera's multiple shooting parameters are adjusted, and the time point and geographical location information of the image shooting are recorded in real time. The specific steps for generating a geo-tagged image sequence are as follows: Based on the surface image acquisition device, adjust multiple camera shooting parameters, including shutter and aperture, according to the real-time ambient light level, optimize image clarity, and generate shooting parameter adjustment records; Based on the shooting parameter adjustment record, shooting a surface image and recording the shooting time of the image to generate a surface image shooting record; Based on the surface image shooting record, the geographical location information of the shooting area is collected and recorded in real time to generate a geographically marked image sequence.

4. The method for monitoring geological disaster activity targets based on dynamic visual tracking according to claim 1, characterized in that: Based on the geo-tagged image sequence, the steps of identifying surface cracks in the image by analyzing the shape, color and texture distribution in the image, recording the width and extension direction of the cracks, and identifying the location of terrain changes by comparing and analyzing multiple images to generate surface feature data are specifically as follows: Based on the geographic tagged image sequence, identifying and recording surface cracks in the image by analyzing the color and texture in the image, and generating surface crack identification information; Based on the surface crack identification information, identify the width and extension direction of the crack, record the geometric shape information of the crack, and generate crack feature identification data; Based on the crack feature recognition data, multiple images of the target location are compared and analyzed to identify the location of terrain changes, including subsidence and uplift, and generate surface feature data.

5. The method for monitoring geological disaster activity targets based on dynamic visual tracking according to claim 1, characterized in that: Based on the surface feature data, by analyzing the images of the terrain change area at multiple time points, calculating the change in surface elevation and displacement direction, and identifying the displacement speed and settlement rate, the steps of generating the ground settlement rate information are specifically as follows: Based on the surface feature data, by analyzing multiple images of the terrain change area, measuring the elevation data of the deformation position at multiple time points, and generating an elevation data change record; Based on the elevation data change record, measuring the coordinates of the surface feature points of the target location at multiple time points, identifying the displacement direction and distance, and generating displacement measurement data; Based on the displacement measurement data, the displacement speed and settlement rate of the target deformation position are calculated to generate ground settlement rate information.

6. The method for monitoring geological disaster activity targets based on dynamic visual tracking according to claim 5 is characterized in that: The specific formula for calculating the displacement velocity and sedimentation rate of the target deformation position is: Wherein, V represents displacement velocity, which indicates the moving rate of the target deformation position per unit time; d represents lateral displacement distance, which is the maximum horizontal displacement measured from continuous image frames; h represents longitudinal displacement height, which is the maximum vertical height change measured from the same series of image frames; t1 represents the time point of the next moment; t0 represents the time point of the previous moment; t1-t0 represents the time interval, which is the time difference obtained by subtracting the time data in the two images; and k represents the terrain adjustment coefficient, which is used to adjust the displacement velocity calculation to adapt to different geological conditions.

7. The method for monitoring geological disaster activity targets based on dynamic visual tracking according to claim 1, characterized in that: Based on the ground subsidence rate information, the steps of identifying various geological disaster behaviors according to the location of deformation, the change rate and the terrain characteristics of the target location and generating the disaster type identification result are specifically as follows: Based on the land subsidence rate information, analyzing the topographic characteristics of the surface deformation location, including terrain slope, soil type, and vegetation coverage, to generate deformation location characteristic data; Based on the deformation location feature data, combined with the degree and rate of surface deformation, the similarity between the deformation characteristics of the target area and multiple known geological disaster types is calculated to generate a similarity calculation result; According to the similarity calculation result, the disaster type of the deformation location is evaluated and identified, including landslide and mud-rock flow, and a disaster type identification result is generated.

8. The method for monitoring geological disaster activity targets based on dynamic visual tracking according to claim 1, characterized in that: Using the disaster type identification result, by performing time series analysis on deformation data, the development trend of the target disaster is predicted, and the steps of generating the disaster development trend analysis result are specifically as follows: Based on the disaster type identification result, the change trend of the deformation speed is analyzed, and the development speed of the disaster is predicted in combination with the terrain characteristics of the target location, and the development speed prediction data is generated; Based on the development speed prediction data, according to the type of disaster and the terrain characteristics of the target location, the diffusion path of the target geological disaster is predicted, and a diffusion path prediction result is generated; Based on the diffusion path prediction results, the impact range of the target disaster and the disaster stop time are predicted, and the disaster development trend analysis results are generated.

9. The method for monitoring geological disaster activity targets based on dynamic visual tracking according to claim 1, characterized in that: In combination with the above-mentioned disaster development trend analysis results, by analyzing and comparing the characteristics and impacts of various geological disasters, the risk levels are calculated for multiple identified geological disasters. The specific steps for generating the risk level assessment results are as follows: Based on the disaster development trend analysis results, and according to the predicted impact range and location information of the target geological disaster, multiple risk areas are identified, including landslide-prone areas, debris flow active areas, and ground subsidence areas, and risk area identification results are generated; Utilizing the risk area identification results, analyzing and predicting the impact of target disasters on human safety, economic activities and infrastructure, and generating disaster impact prediction information; Based on the disaster impact prediction information, taking into account the expected degree of damage, scope of impact, and difficulty of recovery, the risk levels of multiple risk geological disasters are calculated to generate a risk level assessment result.

10. The method for monitoring geological disaster activity targets based on dynamic visual tracking according to claim 9, characterized in that: The specific formula for calculating the risk levels of multiple risk geological disasters is: Among them, R represents the comprehensive risk level, which comprehensively reflects the potential impact of the disaster and the urgency of management. D represents the expected degree of damage, which quantifies the economic losses that may be caused by the disaster and the direct impact on the human living environment. A represents the scope of impact, which covers the breadth of the affected geographical area and the population density. H represents the difficulty of recovery, which reflects the complexity of restoring infrastructure and social order after the disaster. P1 is the weight coefficient of the expected degree of damage, which is used to adjust the influence of the expected degree of damage. P2 is the weight coefficient of the scope of impact, which is used to adjust the importance of the scope of impact. P3 is the weight coefficient of the difficulty of recovery, which is used to adjust the proportion of the difficulty of recovery in the total assessment. These parameters are obtained through expert evaluation and data analysis to ensure the scientificity and adaptability of risk assessment.

Images