Slope deformation monitoring method and device based on multi-source data fusion

Abstract

The invention provides a slope deformation monitoring method and device based on multi-source data fusion, and relates to the technical field of dam slope safety monitoring. The method comprises the steps that a multi-source monitoring system composed of a global satellite signal receiver, a measuring robot and an interference synthetic aperture radar is arranged in a dam slope area; according to the collected slope deformation disturbance parameters, performing monitoring area selection division on a dam slope area; determining a data fusion weight corresponding to each data fusion observation area; collecting multi-source monitoring data in each data fusion observation area; and based on the determined data fusion weight, performing weighted fusion processing on the multi-source monitoring data in the selected area to generate a slope deformation monitoring result. According to the technical scheme, fusion of multi-source monitoring data is realized through region division, the reliability, comprehensiveness and environment adaptability of slope deformation monitoring data are improved, and the stability, accuracy and precision of a slope deformation monitoring result are improved.

Description

Background Art

[0002] The slope deformation monitoring in alpine canyon areas poses significant technical challenges due to their complex geological environment and extreme natural conditions. These areas usually have drastic terrain undulations, dense vegetation cover, and are significantly affected by external forces such as rainfall and weathering, resulting in a relatively high potential risk of geological disasters. As an important part of the dam project, the structural stability of the dam slope area is directly related to the overall safety and reliability of the dam. Under the long-term action of natural conditions and external loads, the dam slope area is prone to deformation accumulation, which may in turn trigger disaster events such as landslides and collapses. Therefore, continuous and accurate deformation monitoring of the dam slope area is of great significance for timely detecting potential hidden dangers and ensuring project safety. The dam slope monitoring technologies in related arts mainly rely on a single type of monitoring equipment. For example, the Global Navigation Satellite System (GNSS) is used for displacement monitoring, or the Interferometric Synthetic Aperture Radar (InSAR) is utilized to achieve large-scale surface deformation measurement. In some applications, the Robotic Total Station (RTS) is also adopted to achieve high-precision displacement observation of key areas.

[0003] However, limited by the complexity of the terrain and the performance limitations of the monitoring equipment itself, there are still certain deficiencies in the practical application of related slope deformation monitoring methods. For example, a single type of monitoring equipment is significantly affected by environmental conditions. For instance, the signal of GNSS attenuates severely in high-occlusion areas, affecting the monitoring accuracy; the interference signal-to-noise ratio of InSAR is relatively low in areas with lush vegetation or snow cover, prone to generating errors; the Robotic Total Station is limited by the site layout conditions and meteorological environment, making it difficult to achieve efficient monitoring of the entire area. Secondly, in the face of diverse geological characteristics and disturbance conditions in the dam slope area, the related monitoring strategies usually adopt unified layout and data collection standards, failing to conduct differential processing according to different regional characteristics, resulting in low utilization efficiency of monitoring resources, and it is difficult to guarantee the accuracy and stability of monitoring results in complex environments. Therefore, how to achieve efficient, accurate, stable, and reliable slope deformation monitoring in the complex and changeable dam slope area is still an issue that urgently needs attention and solution in the current technical field.

[0004] It should be noted that the information disclosed in the above Background Art section is only used to enhance the understanding of the background of the present disclosure, and thus may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the embodiments of the present disclosure is to provide a slope deformation monitoring method based on multi-source data fusion, a slope deformation monitoring device based on multi-source data fusion, an electronic device and a computer-readable storage medium, so as to fuse multi-source monitoring data, improve the reliability, comprehensiveness and environmental adaptability of slope monitoring data, and enhance the stability, accuracy and precision of slope deformation monitoring results.

[0006] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0007] According to a first aspect of an embodiment of the present disclosure, a slope deformation monitoring method based on multi-source data fusion is provided, comprising: A monitoring coverage network consisting of global satellite signal receivers, measurement robots and interferometric synthetic aperture radar is deployed in the dam slope area to be monitored to form a multi-source monitoring system; According to the collected slope deformation disturbance parameters, the dam slope area is divided into monitoring areas to obtain multiple data fusion observation areas; Determine the data fusion weight corresponding to each of the data fusion observation areas; Periodically collecting multi-source monitoring data in each of the data fusion observation areas through the multi-source monitoring system; Based on the data fusion weights of each data fusion observation area, weighted fusion processing is performed on the collected multi-source monitoring data to generate slope deformation monitoring results corresponding to each data fusion observation area.

[0008] In some example embodiments of the present disclosure, based on the aforementioned scheme, the monitoring area of the dam slope is divided according to the collected slope deformation disturbance parameters to obtain multiple data fusion observation areas, including: preliminarily dividing the dam slope area according to a preset division grid to obtain multiple monitoring sub-areas; based on the slope deformation disturbance parameters in each of the monitoring sub-areas, the monitoring sub-areas are classified, and the monitoring sub-areas belonging to the same classification are merged and fused to obtain multiple data fusion observation areas.

[0009] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the slope deformation disturbance parameters include the dam horizontal distance, slope gradient, satellite signal occlusion degree, vegetation coverage density, and potential geological hazard risk level at each position coordinate in the dam slope area; based on the slope deformation disturbance parameters in each of the monitoring sub-areas, classifying the monitoring sub-areas, and merging and fusing the monitoring sub-areas belonging to the same classification to obtain a plurality of data fusion observation areas, including: determining the quantitative scores of the dam horizontal distance, slope gradient, satellite signal occlusion degree, vegetation coverage density, and potential geological hazard risk level in each of the monitoring sub-areas; fusing the quantitative scores according to the weight ratios of the slope deformation disturbance parameters to obtain the area classification scores of each of the monitoring sub-areas; classifying the monitoring sub-areas through the area classification scores, and merging and fusing the monitoring sub-areas belonging to the same classification to obtain a plurality of data fusion observation areas.

[0010] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the method further includes: obtaining the topographic elevation data of the dam slope area, and determining the dam horizontal distance from each position coordinate in the dam slope area to the main body of the dam according to the topographic elevation data; calculating the slope gradient of each dam slope area through the topographic elevation data to determine the slope gradient of each position coordinate; obtaining the signal occlusion test results based on the unmanned aerial vehicle, and determining the satellite signal occlusion degree at each position coordinate according to the signal strength in the signal occlusion test results; calculating the normalized difference vegetation index through multi-temporal remote sensing images of the dam slope area, generating a vegetation coverage map based on the normalized difference vegetation index, and determining the vegetation coverage density at each position coordinate through the vegetation coverage map; determining the potential geological hazard risk level of each position coordinate according to the historical geological hazard data and geological exploration results corresponding to the dam slope area.

[0011] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the determining the data fusion weights corresponding to each of the data fusion observation areas includes: obtaining the environmental impact factors of each monitoring device in the multi-source monitoring system under the dam horizontal distance, slope gradient, satellite signal occlusion degree, vegetation coverage density, and potential geological hazard risk level; calculating the environmental adaptability scores of each monitoring device in the current data fusion observation area through the slope deformation disturbance parameters and the environmental impact factors in the data fusion observation area; performing normalization processing on the environmental adaptability scores of each monitoring device to obtain the data fusion weights corresponding to the data fusion observation area.

[0012] In some exemplary embodiments of the present disclosure, based on the foregoing solution, before normalizing the environmental adaptability scores of the monitoring devices to obtain the data fusion weights corresponding to the data fusion observation regions, the method further includes: if it is determined that the environmental adaptability score of the current monitoring device is less than or equal to a preset score threshold, setting the environmental adaptability score of the current monitoring device to zero.

[0013] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the periodic acquisition of multi-source monitoring data in each of the data fusion observation regions by the multi-source monitoring system includes: determining the deployment density and acquisition frequency corresponding to the data fusion observation region according to the environmental adaptability scores of the global satellite signal receiver, the surveying robot, and the interferometric synthetic aperture radar in the multi-source monitoring system in the data fusion observation region; and acquiring the multi-source monitoring data in each of the data fusion observation regions by the multi-source monitoring system at the deployment density and the acquisition frequency.

[0014] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the multi-source monitoring data includes monitoring point change monitoring data, surface deformation monitoring data, and local high-precision monitoring data. The acquisition of the multi-source monitoring data in each of the data fusion observation regions includes: acquiring the monitoring point change monitoring data in the geocentric coordinate system by the global satellite signal receiver set at each monitoring point in the data fusion observation region; acquiring the surface deformation monitoring data in the image coordinate system by the interferometric synthetic aperture radar set in the data fusion observation region; and acquiring the local high-precision monitoring data in the local coordinate system by the surveying robot set in the data fusion observation region.

[0015] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the method further includes: constructing a rotation matrix and a translation matrix based on the position coordinates of the selected reference monitoring point, and converting the monitoring point change monitoring data to the local coordinate system through the rotation matrix and the translation matrix to obtain the unified point position change monitoring data; solving a projection matrix based on the position coordinates and image coordinates of the reference monitoring point, and converting the surface deformation monitoring data to the local coordinate system through the projection matrix to obtain the unified surface deformation monitoring data; performing deviation correction and alignment on the local high-precision monitoring data based on the position coordinates of the reference monitoring point to obtain the unified local high-precision monitoring data; and performing consistency verification on the unified monitoring point change monitoring data, the unified surface deformation monitoring data, and the unified local high-precision monitoring data to obtain the multi-source monitoring data in the data fusion observation region.

[0016] According to a second aspect of the embodiments of the present disclosure, there is provided a slope deformation monitoring device based on multi-source data fusion, including: A multi-source monitoring module, configured to deploy a monitoring coverage network composed of a global satellite signal receiver, a total station, and an interferometric synthetic aperture radar in the dam slope area to be monitored, so as to form a multi-source monitoring system; A monitoring selection area division module, configured to divide the monitoring selection area of the dam slope area according to the collected slope deformation disturbance parameters, so as to obtain a plurality of data fusion observation areas; A fusion weight module, configured to determine the data fusion weights corresponding to the respective data fusion observation areas; A multi-source monitoring data acquisition module, configured to periodically acquire multi-source monitoring data in the respective data fusion observation areas through the multi-source monitoring system; A slope deformation analysis module, configured to perform weighted fusion processing on the acquired multi-source monitoring data based on the data fusion weights of the respective data fusion observation areas, so as to generate slope deformation monitoring results corresponding to the respective data fusion observation areas.

[0017] The technical solution provided by the embodiments of the present disclosure may include the following beneficial effects: The slope deformation monitoring method based on multi-source data fusion in the example embodiment of the present disclosure, on the one hand, can effectively realize the dynamic matching of the deployment of monitoring equipment and the characteristics of the slope area by deploying a multi-source monitoring system composed of a global satellite signal receiver, a measuring robot and an interferometric synthetic aperture radar in the dam slope area, and combining the regional division according to the slope deformation disturbance parameters. Moreover, since the slope deformation disturbance parameters can reflect the differences in terrain changes, geological characteristics and environmental conditions in the region, combined with the adaptability of multiple types of monitoring equipment in different environments, combined with the data fusion weights, it is possible to achieve targeted fusion of monitoring data according to the actual terrain and environmental conditions of different data fusion observation areas, thereby improving the monitoring coverage and data reliability in complex terrain, obstructed environment and high-risk areas, and making up for the monitoring blind spot problem caused by the limitation of a single monitoring means in related technologies; on the other hand, On the one hand, by weighting the data collected by the multi-source monitoring system according to the data fusion weights specific to each observation area, the advantages of data from different sources can be adaptively exerted according to the actual environmental characteristics. The determined data fusion weights not only take into account the environmental adaptability of each monitoring device under different slope deformation disturbance parameters, but also eliminate the interference of single monitoring data anomalies or deviations on the overall monitoring results through weighted fusion, thereby improving the stability and consistency of slope deformation monitoring results; on the other hand, by periodically collecting multi-source monitoring data in each observation area and fusing the data based on unified standards, the deformation process of the slope area can be continuously tracked in the time dimension, improving the time continuity and change trend recognition ability of the monitoring data, so that the system can capture abnormal signs in time at the early stage of slope deformation, which is helpful to form a complete and continuous record of slope deformation evolution. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0019] Figure 1 A schematic flowchart of a slope deformation monitoring method based on multi-source data fusion according to some embodiments of the present disclosure is schematically shown.

[0020] Figure 2 The flowchart of determining a data fusion observation area by monitoring sub-areas according to some embodiments of the present disclosure is schematically shown.

[0021] Figure 3 A schematic diagram of a process for collecting slope deformation disturbance parameters according to some embodiments of the present disclosure is schematically shown.

[0022] Figure 4 Schematically shows a flowchart of determining data fusion weights for a data fusion observation area according to some embodiments of the present disclosure.

[0023] Figure 5 Schematically shows a scenario diagram of dividing a data fusion observation area in a dam slope area according to some embodiments of the present disclosure.

[0024] Figure 6 Schematically shows a schematic diagram of a slope deformation monitoring device based on multi-source data fusion according to some embodiments of the present disclosure.

[0025] Figure 7 Schematically shows a schematic diagram of a computer system of an electronic device according to some embodiments of the present disclosure.

[0026] Figure 8 Schematically shows a schematic diagram of a computer-readable storage medium according to some embodiments of the present disclosure.

[0027] In the drawings, the same or corresponding reference numerals indicate the same or corresponding parts. Detailed Description of Specific Embodiments

[0028] Here, exemplary embodiments will be described in detail, and examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this specification. On the contrary, they are merely examples of devices and methods consistent with some aspects of this specification as detailed in the appended claims.

[0029] In addition, the drawings are only schematic diagrams and are not necessarily drawn to scale. The block diagrams shown in the drawings are only functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0030] In the present exemplary embodiment, first, a slope deformation monitoring method based on multi-source data fusion is provided. The slope deformation monitoring method based on multi-source data fusion can be applied to a terminal device or a server. This embodiment does not make special limitations in this regard. Subsequently, the case where the server executes this method will be used as an example for description. Figure 1 Schematically shows a flowchart of a slope deformation monitoring method based on multi-source data fusion according to some embodiments of the present disclosure. Refer to Figure 1As shown, the slope deformation monitoring method based on multi-source data fusion may include the following steps: Step S110, deploying a monitoring coverage network consisting of global satellite signal receivers, measurement robots and interferometric synthetic aperture radar in the dam slope area to be monitored to form a multi-source monitoring system; Step S120, dividing the dam slope area into monitoring areas according to the collected slope deformation disturbance parameters to obtain multiple data fusion observation areas; Step S130, determining the data fusion weight corresponding to each of the data fusion observation areas; Step S140, periodically collecting multi-source monitoring data in each of the data fusion observation areas through the multi-source monitoring system; Step S150, based on the data fusion weights of each data fusion observation area, weighted fusion processing is performed on the collected multi-source monitoring data to generate slope deformation monitoring results corresponding to each data fusion observation area.

[0031] According to the slope deformation monitoring method based on multi-source data fusion in this example embodiment, on the one hand, by deploying a multi-source monitoring system consisting of a global satellite signal receiver, a measuring robot and an interferometric synthetic aperture radar in the dam slope area, and combining the regional division according to the slope deformation disturbance parameters, it is possible to effectively achieve dynamic matching between the deployment of monitoring equipment and the characteristics of the slope area. Moreover, since the slope deformation disturbance parameters can reflect the differences in terrain changes, geological characteristics and environmental conditions in the region, combined with the adaptability of multiple types of monitoring equipment in different environments, combined with the data fusion weights, it is possible to achieve targeted fusion of monitoring data according to the actual terrain and environmental conditions of different data fusion observation areas, thereby improving the monitoring coverage and data reliability in complex terrain, obstructed environments and high-risk areas, and making up for the monitoring blind spot problem caused by the limitation of a single monitoring method in related technologies. On the one hand, by weighting the data collected by the multi-source monitoring system according to the data fusion weights specific to each observation area, the advantages of data from different sources can be adaptively exerted according to the actual environmental characteristics. The determined data fusion weights not only take into account the environmental adaptability of each monitoring device under different slope deformation disturbance parameters, but also eliminate the interference of single monitoring data anomalies or deviations on the overall monitoring results through weighted fusion, thereby improving the stability and consistency of slope deformation monitoring results; on the other hand, by periodically collecting multi-source monitoring data in each observation area and fusing the data based on unified standards, the deformation process of the slope area can be continuously tracked in the time dimension, improving the time continuity and change trend recognition ability of the monitoring data, so that the system can capture abnormal signs in time at the early stage of slope deformation, which is helpful to form a complete and continuous record of slope deformation evolution.

[0032] Next, the slope deformation monitoring method based on multi-source data fusion in this exemplary embodiment will be further described.

[0033] In step S110, a monitoring coverage network composed of a global satellite signal receiver, a measuring robot, and an interferometric synthetic aperture radar is arranged in the dam slope area to be monitored, forming a multi-source monitoring system.

[0034] In an exemplary embodiment of the present disclosure, the multi-source monitoring system can be arranged and planned based on the topographic and geomorphic features of the dam body and its two-side slope areas. The global satellite signal receiver refers to a receiving station device configured with a high-precision GNSS module, which can receive satellite signals in real time and perform high-precision three-dimensional positioning under the support of multi-frequency and multi-systems (such as GPS, Beidou, GLONASS, Galileo). The measuring robot refers to a total station system with an automatic target recognition and tracking function, which can obtain high-precision angle and distance observation values by actively locking the reflection prism without manual operation. The interferometric synthetic aperture radar includes an orbital InSAR observation system and a ground-based InSAR system, which are used to analyze surface deformation based on radar phase interference technology and form monitoring data with a large-scale coverage.

[0035] In the actual arrangement process, the global satellite signal receiver is preferably set at an open and unobstructed elevation position to ensure the GNSS signal reception quality, and is evenly distributed in combination with the arrangement grid strategy. The spacing between the set points of the global satellite signal receiver can be set within the range of dozens of meters to hundreds of meters according to the monitoring accuracy requirements, and this embodiment does not make special limitations on this. In areas with serious local signal occlusion but active geological activities, the coverage ability can be improved by setting up relay signal modules or portable receiving nodes.

[0036] The measuring robot can be installed on a fixed base with a good visual field, usually located at the toe of the slope or a relatively stable platform area of the slope, to ensure the visual tracking ability of multiple monitoring prism targets. The prism targets are set on key slip surfaces, crack edges, or important structures, and are arranged according to the optimal observation geometry to avoid extreme elevation angles or occlusions.

[0037] An interferometric synthetic aperture radar can determine the combined use of an orbiting satellite InSAR or a ground-based InSAR based on the scope of the slope area, the terrain occlusion situation, and the vegetation distribution. Among them, the ground-based InSAR equipment can be installed at a position that can overlook the monitoring area without occlusion to ensure the maximization of the radar viewing angle and the quality of the target reflected signal. Optionally, for areas with large-scale vegetation coverage or severe signal decorrelation, artificial corner reflectors can be laid to enhance the coherence of the radar echo; for local areas of high-steep slopes, portable short-range survey robots can be used to strengthen local high-frequency monitoring. Further, under extreme environmental conditions, both the global satellite signal receiver and the survey robot can be equipped with protective covers and environmental monitoring modules to cope with the influence of natural factors such as rainfall, snow, high temperature, and strong wind, and ensure the stable operation of the equipment.

[0038] In step S120, according to the collected slope deformation disturbance parameters, the dam slope area is monitored and divided into multiple data fusion observation areas.

[0039] In an exemplary embodiment of the present disclosure, the slope deformation disturbance parameter refers to a quantitative index characterizing the environmental changes and deformation activity characteristics of the slope area. For example, the slope deformation disturbance parameter may include, but is not limited to, the horizontal distance of the dam at each position point, the slope gradient, the degree of satellite signal occlusion, the vegetation coverage density, and the potential risk level of geological disasters. Of course, there may also be other parameters, such as the current meteorological conditions, the visible area environment, etc. The type of slope deformation disturbance parameter is not specifically limited in this exemplary embodiment.

[0040] The horizontal distance of the dam refers to the horizontal straight-line distance from the monitoring point to the center of the dam body, which is used to evaluate the degree of influence of the area on the structural stability of the dam; the slope gradient can be calculated based on the Digital Elevation Model (DEM) and reflects the degree of surface inclination; the degree of satellite signal occlusion can be obtained through statistical analysis of the signal-to-noise ratio (SNR) of the GNSS signal receiver, which describes the degree of signal obstruction by obstacles; the vegetation coverage density can be calculated through the Normalized Difference Vegetation Index (NDVI) extracted from remote sensing images to evaluate the influence of vegetation on the signal quality of monitoring means (such as InSAR); the potential risk level of geological disasters is classified based on historical disaster data, lithology distribution, and the development of tectonic fractures.

[0041] After obtaining the slope deformation disturbance parameters at each location, optionally, the dam slope area can be preliminarily divided into regular grid cells according to a preset rule. The area of each cell can be set to 10 meters × 10 meters, 20 meters × 20 meters or other appropriate sizes according to engineering requirements. Subsequently, within each grid cell, a quantitative scoring process is carried out based on the corresponding slope deformation disturbance parameters. For example, in areas where the horizontal distance from the dam is closer, the slope gradient is larger, the signal occlusion is more serious, the vegetation is denser, and the risk level is higher, the disturbance score is higher. Different slope deformation disturbance parameters are assigned different weights according to their influence degrees on the reliability, timeliness, and coverage of monitoring data. For example, for GNSS monitoring, the weight of the signal occlusion degree is higher than the weight of the vegetation coverage density to reflect the sensitivity differences of different monitoring technologies to environmental characteristics.

[0042] In an alternative implementation, if there are special landforms (such as cliffs, slip zones, and trenches) within the dam slope area, the data fusion observation area can be assisted and corrected by combining the landform feature map to avoid division errors caused by sudden terrain changes. Further, around high-dynamic areas (such as landslides with frequent deformation activities), the monitoring selection area division result can be updated in real time through dynamic disturbance parameter monitoring to enhance the overall system's rapid response ability to sudden geological changes.

[0043] In step S130, determine the data fusion weights corresponding to each of the data fusion observation areas.

[0044] In an exemplary embodiment of the present disclosure, the data fusion weight refers to the contribution ratio assigned to the monitoring data from different sources during the multi-source data fusion process, used to balance the influence degrees of various types of monitoring data in the fusion calculation. For example, the data fusion weight can be determined based on the adaptability, data accuracy, and stability of each monitoring device in a specific environment. Of course, the data fusion weight can also comprehensively consider the data sampling frequency, spatial resolution, and real-time indicators of the monitoring device. This exemplary embodiment does not make a special limitation on the specific setting method of the weight ratio.

[0045] Optionally, the data fusion weight can be directly assigned through a preset rule. For example, reduce the GNSS data weight and increase the InSAR data weight in areas with severe satellite signal occlusion. It can also adopt a dynamic evaluation mechanism to adjust the weight according to real-time data quality indicators. For example, evaluate the data stability through the variance of the change amount of the monitoring point position and dynamically correct the weights of each device to adapt to environmental changes. Of course, it can also combine a weight optimization strategy based on machine learning algorithms, use historical monitoring data to train a regression model, and automatically output the optimal fusion weight configuration scheme for each area. This exemplary embodiment does not make a special limitation on the setting method of the data fusion weight.

[0046] The corresponding data fusion weight can be obtained by combining the weight setting method with the applicability and data reliability of each monitoring device within each data fusion observation area. By setting the data fusion weights that meet the terrain and environmental characteristics of each data fusion observation area, in the subsequent data processing process, the contribution of data sources with high reliability and low error in a specific environment can be highlighted, and the influence of potential noise sources can be suppressed, thereby improving the overall consistency and monitoring accuracy of the fused data.

[0047] In step S140, the multi-source monitoring data in each of the data fusion observation areas is periodically collected by the multi-source monitoring system.

[0048] In an example embodiment of the present disclosure, multi-source monitoring data refers to a set of observation data that is obtained from the observation area through different monitoring equipment and can reflect the deformation state of the slope area. For example, the multi-source monitoring data may include the displacement data of the monitoring point collected by the GNSS receiver, the three-dimensional coordinate change data measured by the surveying robot, and the surface deformation interference map data obtained by the InSAR system; of course, the multi-source monitoring data may also have other extended data types, such as remote sensing image change data, ground crack sensor monitoring data, or temperature and humidity environmental data. This embodiment does not specifically limit the source type of the multi-source monitoring data. In an optional implementation, the GNSS receiver can be set to a continuous acquisition mode to obtain high-frequency displacement data at time intervals of seconds or minutes; the surveying robot can regularly collect the spatial position information of the control point according to the set inspection cycle; the InSAR system can obtain surface deformation image data every few days based on the orbit revisit cycle.

[0049] Periodic collection refers to the regular updating and collection of data in each observation area within a preset time period according to an established plan. For example, GNSS data can be set to real-time mobile collection, measurement robot data can be updated in batches on a daily or weekly basis, and InSAR data can be downloaded and processed regularly according to the satellite orbit cycle. Of course, the collection cycle can also be dynamically adjusted according to the monitoring stage. The sampling period can be extended when the slope is in a stable state, and the sampling period can be shortened when abnormal change trends are found, so as to achieve a balance between resource optimization and sensitive capture of change processes.

[0050] It is understandable that periodic collection can be set up in layers based on regional importance, risk level and equipment availability. For example, a high-frequency sampling strategy can be adopted for high-risk areas, and a low-frequency sampling strategy can be adopted for stable areas to rationally allocate monitoring resources. Of course, the sampling frequency can also be temporarily increased during special periods (such as heavy rainfall and large-scale construction activities) to enhance the monitoring capabilities of environmental disturbances.

[0051] By periodically collecting multi-source monitoring data in each data fusion observation area through a multi-source monitoring system, a continuous spatio-temporal data chain of the slope deformation process can be established, which not only ensures the temporal continuity of the monitoring data, but also can timely capture the abnormal change trend of the slope state, improve the early identification ability of potential risk changes, and provide a complete and dynamic observation basis for subsequent data fusion processing and deformation trend analysis.

[0052] In step S150, based on the data fusion weights of each data fusion observation area, the collected multi-source monitoring data are respectively weighted and fused to generate the slope deformation monitoring results corresponding to each data fusion observation area.

[0053] In an exemplary embodiment of the present disclosure, the weighted fusion process refers to proportionally weighting and integrating data from different sources according to the data fusion weights determined for each monitoring data source in the current observation area to generate a unified monitoring result reflecting the comprehensive deformation state. For example, the weighted fusion process can adopt the weighted average method to perform superposition calculations of spatial positions or displacement amounts of different data sources according to the weight ratio; of course, the weighted fusion process can also adopt the weighted least squares method (Weighted Least Squares, WLS) to model the errors of each data source and perform optimized fusion to minimize the overall error. This embodiment does not make a special limitation on the selection of specific data fusion algorithms.

[0054] It can be understood that during the fusion process, data preprocessing can be preferentially performed, such as unifying the coordinate reference, time synchronization, anomaly rejection, etc., to ensure the basic consistency of the fusion process; of course, a consistency verification step can also be added after the fusion to judge the reliability of the fusion result by verifying the fitting residuals of different data sources, thereby further improving the accuracy and stability of the final monitoring data.

[0055] Next, the content in steps S110 to S150 will be described in detail.

[0056] In an exemplary embodiment of the present disclosure, the monitoring area selection of the dam slope area can be realized according to the collected slope deformation disturbance parameters through the following steps to obtain multiple data fusion observation areas, which specifically may include: The dam slope area can be initially divided into multiple monitoring sub-areas according to a preset division grid; based on the slope deformation disturbance parameters in each monitoring sub-area, the monitoring sub-areas are classified, and the monitoring sub-areas belonging to the same classification are merged and fused to obtain multiple data fusion observation areas.

[0057] Among them, the preset division grid refers to a grid division structure that is set in advance according to a certain spatial scale and rules for dividing the monitoring range. For example, the division grid can adopt a square grid, a rectangular grid or a hexagonal honeycomb grid to ensure the uniform division of the monitoring area. Of course, the division grid can also adaptively adjust the size of the grid cells according to the change of terrain elevation difference. For example, small-size grids are used in areas with severe terrain undulations, and large-size grids are used in areas with gentle terrain. The specific shape and size of the division grid are not particularly limited in this exemplary embodiment. Optionally, the division grid can be directly divided based on the longitude and latitude coordinate system, which is suitable for large-scale rough zoning; it can also be divided planarly based on the projection coordinate system (such as the UTM coordinate system), which is suitable for local high-precision zoning requirements. Of course, the division grid can also be automatically generated according to the dam design plan or terrain DEM (Digital Elevation Model) to adapt to the actual engineering layout.

[0058] After determining the division grid, the preliminary division of the dam slope area can be realized, that is, according to the positions of the grid nodes, the dam slope range is covered and divided into multiple independent small areas, and each small area is used as a monitoring sub-area. Each monitoring sub-area has an independent spatial boundary and number identification, and can perform data collection, parameter calculation and subsequent classification processing independently. Through the preliminary area division by the preset grid, a standardized monitoring unit system can be effectively established quickly in a large range, avoiding the subjective differences brought by manual division, and providing a good spatial basis for subsequent refinement processing and data management.

[0059] After obtaining the slope deformation disturbance parameters of each monitoring sub-area, classification is carried out based on the parameter characteristics of each monitoring sub-area. The classification can adopt a rule-based classification method based on a set threshold. For example, the sub-areas where the horizontal distance of the dam is less than a certain value and the slope is greater than a certain angle threshold are classified as high-deformation sensitive areas; unsupervised clustering algorithms (such as K-means, DBSCAN, etc.) can also be used to automatically divide areas with similar characteristics without the need to set classification criteria manually in advance. The classification method is not particularly limited in this exemplary embodiment. Of course, the classification process can also be dominated by a single main control parameter (such as slope), or can be comprehensively classified based on the combined characteristics of multiple disturbance parameters. For example, after extracting the main components of the disturbance parameters by the principal component analysis (Principal Component Analysis, PCA) method, clustering is carried out to improve the scientificity and stability of the classification.

[0060] After completing the sub-region classification, the monitoring sub-regions belonging to the same classification can be merged and fused to generate multiple data fusion observation regions. Merging and fusing means merging adjacent sub-regions with the same or similar classification results into a larger monitoring region, so that subsequent monitoring layout, data collection and fusion processing can be more targeted and regionally unified. In the process of merging and fusing, an adjacency relationship determination strategy can be adopted, that is, only adjacent or similar sub-regions of the same type in space are merged to avoid the formation of unreasonable observation regions that are spatially discontinuous or cross abnormal terrain features; of course, a fusion scale threshold can also be set. If the number of sub-regions of a certain category is less than the preset standard, the fusion range can be expanded or merged into the nearest neighbor category for supplementary optimization to ensure the rationality of the fusion result and the practicality of monitoring.

[0061] By classifying based on the slope deformation disturbance parameters in each monitoring sub-region and merging and fusing the monitoring sub-regions belonging to the same classification, it is possible to achieve refined management of different regions and differential monitoring layout according to the differences in the natural environment and potential risks within the dam slope region, and improve the adaptability, resource allocation rationality and monitoring accuracy of the entire monitoring system in a complex environment.

[0062] In an exemplary embodiment of the present disclosure, the slope deformation disturbance parameters may include the dam horizontal distance, slope gradient, satellite signal occlusion degree, vegetation coverage density, and potential geological hazard risk level at each position coordinate in the dam slope region.

[0063] Among them, the dam horizontal distance refers to the shortest horizontal distance from each position point in the slope region to the main body of the dam, which is used to measure the association strength between this position and the dam structure. For example, the dam horizontal distance can be calculated by extracting the shortest Euclidean distance between the position point coordinates and the dam center line; in complex terrain, it can also be corrected and calculated based on the projection distance or the distance along the slope path to improve the practical applicability of actual projects; of course, the dam horizontal distance can also be graded and quantified, such as the first level for 0-50 meters, the second level for 50-100 meters, and so on, for subsequent standardized scoring processing. The specific calculation model of the dam horizontal distance in this exemplary embodiment is not particularly limited.

[0064] The slope gradient refers to the degree of surface inclination at each position within the slope area, usually expressed as the angle in degrees with respect to the horizontal plane or as a percentage. For example, the slope gradient can be calculated based on the elevation difference and horizontal distance between adjacent pixels in a Digital Elevation Model (DEM). In a small local area, the slope value can also be directly obtained using triangulation based on on-site measurement data. Of course, the slope gradient can also be discretized according to slope grades, such as low slope for 0 - 15°, medium slope for 15 - 30°, and high slope for above 30°, for subsequent unified quantification. This exemplary embodiment does not particularly limit the slope calculation method.

[0065] The degree of satellite signal occlusion refers to the degree of reduction in GNSS signal reception quality at each position point within the slope area due to reasons such as terrain occlusion, vegetation, or artificial structures. For example, the degree of satellite signal occlusion can be determined by measuring the change in the Signal to Noise Ratio (SNR) using an on-site signal quality tester. It can also be based on three-dimensional terrain data to simulate the occlusion angles in different directions and calculate the number of visible satellites, and then infer the occlusion degree. Of course, the occlusion degree can be quantified as an occlusion coefficient from 0 to 1, where 0 represents no occlusion and 1 represents complete occlusion, for standardizing subsequent scoring. This exemplary embodiment does not particularly limit the occlusion measurement method.

[0066] The vegetation coverage density refers to the degree of surface vegetation coverage at each position point within the slope area, usually obtained through remote sensing image analysis. For example, the vegetation coverage density can be calculated to obtain the vegetation coverage percentage based on the Normalized Difference Vegetation Index (NDVI). It can also be based on pixel classification of high-resolution optical images or LiDAR point cloud data for vegetation height extraction and then infer the coverage degree. Of course, the vegetation coverage density can be discretized into several grades within the range of 0 - 100% coverage rate to meet different analysis requirements. This exemplary embodiment does not particularly limit the vegetation density extraction method.

[0067] The potential risk level of geological disasters refers to the disaster sensitivity index comprehensively evaluated at each position point based on historical disaster events, geological structure characteristics, and geological survey results. For example, the potential risk level of geological disasters can be classified according to the occurrence probabilities of disasters such as landslides, collapses, and debris flows. It can also be quantitatively evaluated by combining the distribution of regional in-situ stress fields, lithological characteristics, and hydrogeological conditions. Of course, the geological disaster risk level can be divided into three levels: low, medium, and high, and different quantitative weights are assigned to each level for subsequent standardization processing. This exemplary embodiment does not particularly limit the risk level classification criteria.

[0068] It can be based on Figure 2The steps in [reference] are based on the slope deformation disturbance parameters in each monitoring sub-region, classify the monitoring sub-regions, and merge and fuse the monitoring sub-regions belonging to the same classification to obtain multiple data fusion observation regions. Figure 2 As shown in [reference], it may specifically include: Step S210, determine the quantitative scores of the dam horizontal distance, the slope gradient, the satellite signal occlusion degree, the vegetation coverage density, and the potential risk level of geological disasters in each of the monitoring sub-regions; Step S220, fuse the quantitative scores according to the weight ratios of the slope deformation disturbance parameters to obtain the regional classification scores of the monitoring sub-regions; Step S230, classify the monitoring sub-regions through the regional classification scores, and merge and fuse the monitoring sub-regions belonging to the same classification to obtain multiple data fusion observation regions.

[0069] Among them, the quantitative score refers to converting each disturbance parameter into a standardized value according to a unified standard to reflect the potential influence degree of the parameter on the regional stability. For example, a relatively short dam horizontal distance can be given a higher sensitivity score, and a larger slope can be given a higher deformation risk score; of course, the quantitative scoring standard can be adjusted and optimized according to actual engineering safety assessment experience. In this example embodiment, the setting method of the quantitative scoring standard is not particularly limited. In an optional implementation manner, a linear mapping method can also be used to convert the original parameter value into a score range from 0 to 100 in proportion; a non-linear scoring function, such as the Sigmoid function, can also be used to highlight the sensitive change characteristics near a specific parameter threshold.

[0070] After completing the quantitative scoring of each disturbance parameter, fuse the quantitative scores according to the weight ratios of the slope deformation disturbance parameters to obtain the regional classification scores of the monitoring sub-regions. Fusion means weighted summing the quantitative scores according to the importance of each disturbance parameter in affecting the slope stability according to the set weights. For example, it can be determined according to experience or historical monitoring analysis that the weight of the dam horizontal distance is 0.3, the weight of the slope gradient is 0.3, the weight of the satellite signal occlusion degree is 0.15, the weight of the vegetation coverage density is 0.15, and the weight of the potential risk level of geological disasters is 0.1 for weighted calculation; the weight ratio can also be dynamically adjusted according to the actual engineering requirements of different monitoring regions. In this example embodiment, the setting method of the weights is not particularly limited.

[0071] After the regional classification scores of each monitoring sub-region are determined, the monitoring sub-regions are classified according to the regional classification scores, and the monitoring sub-regions belonging to the same classification are merged and integrated. The regional classification can be based on the classification score interval division criteria. For example, sub-regions with scores above 80 are classified as high-risk areas, those with scores between 50 and 80 are classified as medium-risk areas, and those with scores below 50 are classified as low-risk areas; or the classification thresholds can be automatically determined according to the statistical quantile method to adapt to different data distribution patterns. The present exemplary embodiment does not particularly limit the setting method of the classification criteria.

[0072] Those skilled in the art can understand that the classification can be combined with the determination of spatial adjacency, and sub-regions with high spatial connectivity and the same classification are preferentially merged to form a spatially continuous fusion observation area; of course, a minimum regional unit limit can also be set to appropriately absorb or merge isolated small sub-regions to optimize the spatial structure after fusion.

[0073] By determining the quantitative scores of the horizontal distance of the dam, slope gradient, satellite signal occlusion degree, vegetation coverage density, and potential risk level of geological disasters in each monitoring sub-region, and fusing them according to the weights of each parameter, and finally classifying and merging the monitoring sub-regions through the regional classification scores, it is possible to achieve precise zoning and differential management of the internal environmental characteristics of the dam slope area in a unified, quantitative, and scientific manner, improve the environmental adaptability, resource allocation rationality, and monitoring accuracy of the slope deformation monitoring system, and provide a reliable basis for subsequent monitoring equipment layout, data fusion processing, and abnormal change identification.

[0074] In an exemplary embodiment of the present disclosure, the acquisition of slope deformation disturbance parameters can be achieved through Figure 3 the steps in, referring to Figure 3 shown, which specifically may include: Step S310, obtaining the topographic elevation data of the dam slope area, and determining the horizontal distance of each position coordinate in the dam slope area to the main body of the dam according to the topographic elevation data; Step S320, calculating the slope of each dam slope area through the topographic elevation data to determine the slope gradient of each position coordinate; Step S330, obtaining the signal occlusion test results based on the unmanned aerial vehicle, and determining the satellite signal occlusion degree at each position coordinate according to the signal intensity in the signal occlusion test results; Step S340, calculating the normalized difference vegetation index through multi-period remote sensing images of the dam slope area, generating a vegetation coverage map based on the normalized difference vegetation index, and determining the vegetation coverage density at each position coordinate through the vegetation coverage map; Step S350, determining the potential risk level of geological disasters at each location coordinate based on the historical geological disaster data and geological exploration results corresponding to the dam slope area.

[0075] Among them, terrain elevation data refers to a set of spatially distributed data describing the elevation values of various location points in the dam slope area. For example, terrain elevation data can be derived from a digital elevation model (DEM) obtained by satellite remote sensing, such as SRTM data, ASTER GDEM data, etc.; of course, terrain elevation data can also be obtained through unmanned aerial vehicle laser radar (Light Detection and Ranging, LiDAR) scanning to provide higher resolution and higher precision elevation information. This embodiment does not specifically limit the method for obtaining elevation data.

[0076] Optionally, terrain elevation data can use a DEM with a resolution of 30 meters for preliminary analysis; in key areas, a high-precision DEM with a resolution of less than 1 meter can be used to improve monitoring accuracy; of course, the quality of elevation data can also be improved through multi-source fusion (such as combining aerial surveys with ground measurements).

[0077] After the terrain elevation data is acquired, the horizontal distance of the dam is determined by measuring the shortest horizontal distance from each location point to the center line of the dam body or the boundary contour based on the elevation information of each location coordinate. The horizontal distance can be calculated using the straight-line Euclidean distance in the projected coordinate system. For example, project the monitoring point to the nearest point on the center line of the dam and calculate the horizontal distance. Of course, in the case of complex and undulating dam slopes, the actual shortest path distance along the slope path can also be introduced for correction to improve the applicability of the project. This embodiment does not specifically limit the specific horizontal distance calculation method. In an optional implementation, the horizontal distance of the dam can be calculated in batches through a GIS (Geographic Information System) platform to improve the processing efficiency of large-scale monitoring points.

[0078] Slope refers to the rate of change of elevation of the surface in the horizontal direction, usually expressed as an angle with the horizontal plane or as a percentage. For example, the slope can be calculated by taking the inverse tangent value after dividing the elevation difference between adjacent pixels by the horizontal distance; of course, in areas where micro-topography changes dramatically, the local plane fitting method can be used to estimate the slope value based on a third-order or high-order neighborhood window to enhance the description of small-scale terrain features. This example embodiment does not specifically limit the slope calculation method. Slope calculation can be performed on DEM data of different resolutions of 5 meters, 10 meters, and 30 meters to meet monitoring needs of different scales; of course, according to the needs of slope stability analysis, slope consistency correction can be used to avoid slope distortion caused by data noise.

[0079] After the slope calculation is completed, the slope information of each position point, as one of the slope deformation disturbance parameters, directly affects the classification of the monitored sub-region and the formulation of subsequent monitoring strategies. Through the quantification of slope characteristics, steep sections, turning zones, and platform areas of the slope can be identified, providing an important basis for the analysis of deformation hazards, the layout of monitoring equipment, and the construction of early warning models.

[0080] The UAV signal occlusion test refers to using a UAV equipped with a GNSS signal receiving device to record in real time indicators such as satellite signal strength, the number of visible satellites, and the signal-to-noise ratio (SNR) at each flight position during the flight along a preset route in the dam slope area, so as to evaluate the occlusion effect of environmental factors such as terrain and vegetation on satellite signal transmission. For example, the signal occlusion test can be completed by a single UAV flying multiple flights to achieve full coverage scanning; of course, multiple UAVs can also be used to fly in coordination to improve the test efficiency. In this example embodiment, there are no special restrictions on the UAV model, flight mode, and test frequency.

[0081] It can be understood that the UAV signal occlusion test can be repeated under different meteorological conditions such as sunny days, cloudy days, and foggy days to capture the influence of weather changes on signal occlusion; of course, the test results at different altitude levels can also be combined to evaluate the variation characteristics of the occlusion mode at different flight altitudes.

[0082] After the signal occlusion test is completed, based on the collected signal strength data and combined with the pre-set occlusion intensity classification standard, for example, the area with SNR lower than a certain threshold is determined as a strong occlusion area, the area with moderately low SNR is determined as a medium occlusion area, and the area with good SNR is determined as an unoccluded area, to determine the satellite signal occlusion degree of each position coordinate. Optionally, the occlusion degree can be normalized to the 0-1 interval for subsequent quantification processing; of course, multi-level occlusion impact indicators can also be set according to the sensitivity requirements of different monitoring devices for signals to finely adapt to the monitoring needs of different device types.

[0083] The Normalized Difference Vegetation Index (NDVI) is an index that characterizes the surface vegetation condition calculated from the reflectance of the red band (Red) and the near-infrared band (Near Infrared, NIR) in remote sensing images. The calculation formula is (NIR - Red) / (NIR + Red), where the value range is usually between -1 and +1. For example, an NDVI value close to 1 indicates high-density healthy vegetation, close to 0 indicates sparse vegetation or bare land, and close to -1 indicates water bodies or snow cover. Of course, in order to adapt to the characteristics of different sensors, the band selection can be adjusted according to the image source. For example, the red-edge band can be used to enhance vegetation sensitivity. In this exemplary embodiment, there is no special limitation on the NDVI calculation band combination. In an alternative implementation, the remote sensing images can be sourced from medium- and high-resolution satellites such as Sentinel-2, Landsat-8, GF-1, etc. When higher accuracy is required, near-surface images collected by an unmanned aerial vehicle equipped with a multispectral camera can also be used to calculate NDVI.

[0084] After obtaining multi-temporal remote sensing images, radiometric calibration, atmospheric correction, and topographic correction are performed on each image to ensure comparability between images. Subsequently, NDVI is calculated pixel by pixel, and multi-temporal NDVI data fusion is performed according to a set time window (such as the four seasons or the growing season) to generate a vegetation coverage distribution map. The vegetation coverage distribution map refers to a layer that maps continuous NDVI values to the surface spatial distribution, which is used to quantitatively describe the vegetation coverage at each location. For example, it can be set that NDVI > 0.6 is a high-density vegetation area, 0.3 < NDVI ≤ 0.6 is a medium-density vegetation area, and NDVI ≤ 0.3 is a low-density vegetation area or bare land. Of course, the classification threshold can also be adjusted according to the actual vegetation type and seasonal changes. In this exemplary embodiment, there is no special limitation on the vegetation density division standard. Optionally, the vegetation coverage can also be generated by the mean or maximum value synthesis of time-series NDVI (such as Maximum Value Composite, MVC) to highlight the vegetation coverage characteristics with higher stability.

[0085] Based on the generated vegetation coverage distribution map, the vegetation coverage density value corresponding to each position coordinate is extracted, and the vegetation density index of this position can be determined, which is used for subsequent disturbance parameter scoring, monitoring equipment deployment feasibility assessment, and InSAR image coherence prediction.

[0086] Historical geological disaster data refers to the collection of records of the time, location, scale, causes, and losses of geological disasters such as landslides, collapses, and debris flows that have occurred in the dam slope area and its surrounding areas. For example, historical geological disaster data can be sourced from the results of geological disaster surveys by government departments, academic literature, engineering monitoring records, or local archival materials. Of course, historical event data can also be supplemented through ground surveys and aerial survey analysis to improve data integrity and accuracy. This exemplary embodiment does not particularly limit the sources of disaster data. In an optional implementation, disaster frequency, scale level, and other indicators can be sorted separately according to different disaster types to construct a disaster spatial distribution database to support subsequent risk quantification analysis.

[0087] Geological exploration results refer to the basic data on the geological structure, lithology, fault development, hydrogeological conditions, etc. of the dam slope area obtained through means such as drilling, trenching, geological mapping, geophysical exploration, and remote sensing interpretation. For example, geological exploration can reveal key information such as the distribution of potential weak interlayers, slip surfaces, and the mechanical properties of rock and soil masses. Of course, comprehensive geophysical exploration (such as seismic wave reflection method, resistivity imaging method) can also be used in combination with direct exploration means to improve the accuracy of underground structure interpretation. This exemplary embodiment does not particularly limit the technical means of geological exploration. In an optional implementation, exploration data can be combined with three-dimensional geological modeling technology to establish a detailed three-dimensional geological model of the slope for more accurately determining the location and scale of potential unstable structural units.

[0088] After obtaining the historical geological disaster data and geological exploration results, spatial overlay analysis is carried out in combination with a geographic information system (GIS) to evaluate the probability of disaster occurrence and the possible impact degree at each location, and to determine the potential risk level of geological disasters at each location. The risk level can be divided into low risk, medium risk, and high risk according to preset criteria. For example, the high-risk area is the area with historical disaster records and the presence of poor geological bodies, the medium-risk area is the area with certain geological hidden dangers but no disasters, and the low-risk area is the area with stable geological conditions and no disaster records. Of course, risk classification can also be refined using semi-quantitative or quantitative risk assessment methods (such as a comprehensive model based on hazard, vulnerability, and exposure). This exemplary embodiment does not particularly limit the risk level classification method.

[0089] By obtaining multi-source basic environmental data such as terrain elevation, signal occlusion, vegetation cover, and geological disaster risks in the dam slope area and performing standardized quantification processing, it is possible to provide accurate and comprehensive environmental feature support for subsequent slope monitoring area selection and data fusion, significantly improving the environmental adaptability and monitoring accuracy of the slope deformation monitoring system.

[0090] In an exemplary embodiment of the present disclosure, the data fusion weights corresponding to each data fusion observation area can be determined through the steps in Figure 4 with reference toFigure 4 As shown, it may specifically include: Step S410: Obtain the environmental impact factors of each monitoring device in the multi-source monitoring system under the conditions of the horizontal distance of the dam, slope gradient, satellite signal occlusion degree, vegetation coverage density, and potential risk level of geological disasters. Step S420: Calculate the environmental adaptability scores of each monitoring device in the current data fusion observation area through the slope deformation disturbance parameters and the environmental impact factors in the data fusion observation area. Step S430: Normalize the environmental adaptability scores of each monitoring device to obtain the data fusion weights corresponding to the data fusion observation area.

[0091] Among them, the environmental impact factor refers to a quantitative index of the performance of each monitoring device under different environmental characteristic parameter conditions, which is used to reflect the adaptability and data reliability of the device in a specific environment. For example, the environmental impact factor can be obtained through statistical analysis of historical monitoring data. For example, if the GNSS reception accuracy decreases by 20% in an area with a large slope, the corresponding slope impact factor is 0.8; the environmental impact factor can also be derived based on a theoretical performance model. For example, according to the electromagnetic wave propagation theory, the decreasing trend of the interference coherence of InSAR in a vegetation-dense area can be estimated. Of course, the environmental impact factor can be established through offline simulation or on-site sampling tests. For example, by carrying a GNSS receiver and an InSAR device to conduct a terrain occlusion experiment and statistically analyzing the variation law of signal intensity and monitoring error; of course, it can also be combined with an artificial experience scoring system, and qualitative or semi-quantitative scores are given to the adaptability of the device under different environmental conditions according to expert experience. In this exemplary embodiment, the acquisition method of the environmental impact factor is not particularly limited.

[0092] In specific implementation, for the horizontal distance parameter of the dam, the interference degree of the multi-path effect of GNSS in the near-dam strong reflection area can be analyzed to set the environmental impact factor; for the slope gradient parameter, the stability and measurement error of the total station in a high-slope area can be evaluated to set the corresponding factor; for the satellite signal occlusion degree, the change of GNSS positioning quality can be measured to determine the impact factor; for the vegetation coverage density, the impact factor can be set through the change rate of the coherence coefficient of the InSAR image; for the potential risk level of geological disasters, the potential threat of surface disturbance to the stability and observation continuity of the monitoring device can be considered to set the corresponding impact factor.

[0093] The environmental adaptability score refers to the quantitative result of comprehensively evaluating the monitoring performance of a device in the current area based on the specific environmental characteristics of the data fusion observation area and the environmental impact factors corresponding to each monitoring device. For example, the environmental adaptability score can combine the impact factors corresponding to each environmental parameter into a single score value through weighted product or weighted summation; it can also adopt the method of fuzzy comprehensive evaluation or machine learning regression model to generate the adaptability score from multi-parameter features. Of course, a linear model can be used. For example, adaptability score = ∑(perturbation parameter weight × impact factor value); of course, a non-linear fusion method can also be used, such as setting a non-linear amplification weight for important perturbation parameters to enhance the sensitivity of the score to extreme environmental conditions. The calculation method of the environmental adaptability score in this exemplary embodiment is not particularly limited.

[0094] In the specific calculation process, the actual values of each perturbation parameter in the current observation area can be extracted, such as the average dam horizontal distance, average slope, average satellite signal occlusion degree, average vegetation coverage density, and potential geological disaster risk level. Subsequently, according to the actual values of each perturbation parameter, the corresponding factor values can be found or interpolated and calculated in the environmental impact factor table of the monitoring device. Finally, the factor values can be weighted and combined according to the preset weight rule to obtain the environmental adaptability score of the monitoring device in the observation area.

[0095] Normalization refers to standardizing the adaptability scores with different numerical dimensions and large numerical range differences between different monitoring devices to a unified scale interval, usually standardized to between 0 and 1 for easy direct comparison and fusion. For example, normalization can adopt the maximum normalization method to standardize all device adaptability scores according to the maximum value; it can also adopt the Z-score normalization method to adjust the score distribution to a standard normal distribution with a mean of 0 and a standard deviation of 1. Of course, a preset normalization formula can also be used, such as normalized score = original score / maximum score of all devices. A score threshold setting mechanism can also be introduced. When the device adaptability score is lower than a certain level, it is directly set to zero to eliminate the interference of extremely low-reliability data sources on the fusion weight. The normalization method in this exemplary embodiment is not particularly limited.

[0096] After the normalization process, the data fusion weight corresponding to each type of monitoring device in each data fusion observation area is obtained. The fusion weight directly determines the contribution ratio of each data source to the final fusion result in the subsequent multi-source data fusion process. Data sources with high scores account for a large proportion in the fusion, while data sources with low scores account for a small proportion or are excluded, so that the fusion result is more in line with the monitoring reliability and adaptability of each data source in the actual environment.

[0097] By obtaining the environmental impact factors of each monitoring device under different environmental characteristics, combining the slope deformation disturbance parameters of the data fusion observation area to calculate the environmental adaptability score, and then performing normalization processing to determine the data fusion weight, it is possible to dynamically adapt the contribution degree of each data source during the multi-source monitoring data fusion process to the actual environmental conditions, significantly improving the data fusion rationality, and the accuracy and stability of the monitoring results of the slope deformation monitoring system in complex and variable environments.

[0098] In an optional implementation manner, before performing normalization processing on the environmental adaptability scores of each monitoring device to obtain the data fusion weights corresponding to the data fusion observation area, if it is determined that the environmental adaptability score of the current monitoring device is less than or equal to a preset score threshold, the environmental adaptability score of the current monitoring device can be set to zero.

[0099] Among them, the environmental adaptability score refers to a quantitative index comprehensively calculated based on the matching degree between the environmental disturbance characteristics and the device performance within a specific data fusion observation area. For example, the environmental adaptability score can comprehensively consider environmental impact factors such as the horizontal distance of the dam, the slope of the slope, the degree of satellite signal occlusion, the vegetation coverage density, and the potential risk level of geological disasters. Of course, to meet different monitoring requirements, the environmental adaptability score can also be calculated by setting a multi-dimensional feature weighted combination model according to the actual monitoring accuracy requirements. The specific calculation method of the environmental adaptability score in this exemplary embodiment is not particularly limited.

[0100] The score threshold refers to a preset boundary value used to screen the rationality and reliability of the device adaptability score. For example, the score threshold can be set to 0.3, that is, if the adaptability score of the device in the current area is lower than or equal to 0.3, it is determined that the monitoring reliability of the device under this environmental condition is insufficient. For example, the score threshold can also be set to 0.2. Specifically, the score threshold can be dynamically adjusted according to factors such as monitoring accuracy requirements, device types, and environmental condition complexity. Of course, the score threshold can be uniformly set to a fixed value to simplify the system design, or it can be adaptively adjusted according to the environmental complexity of each data fusion observation area. For example, the score threshold requirement is increased in high-risk areas and the threshold is appropriately relaxed in low-risk areas. The specific setting method of the score threshold in this exemplary embodiment is not particularly limited.

[0101] In specific implementation, first, the environmental adaptability score of each monitoring device in the current observation area is judged. If the score value is less than or equal to the preset score threshold, the environmental adaptability score of the monitoring device can be directly set to zero, that is, in the subsequent data fusion weight normalization process, the data contribution of the monitoring device is completely eliminated. Through the zero-setting process, it can effectively prevent monitoring devices with significantly deteriorated performance under specific environmental conditions from participating in data fusion, avoiding the introduction of noise, errors or distortion due to low-quality data sources, thereby improving the overall accuracy and stability of the fused data.

[0102] By introducing a score threshold screening mechanism before the normalization process of the environmental adaptability score and setting the scores of devices that do not meet the threshold requirements to zero, it can effectively enhance the rigor and adaptive ability of device screening in the multi-source data fusion process, ensuring that the finally generated data fusion weights can more truly reflect the effectiveness and reliability of each monitoring device in the current environment, and further improving the adaptability of the slope deformation monitoring system to complex and changeable environments, the robustness of the data fusion results, and the sensitivity of abnormal change recognition.

[0103] In an exemplary embodiment of the present disclosure, the periodic acquisition of multi-source monitoring data in each data fusion observation area by a multi-source monitoring system can be implemented through the following steps, which may specifically include: The layout density and acquisition frequency corresponding to the data fusion observation area can be determined according to the environmental adaptability scores of the global satellite signal receiver, total station, and interferometric synthetic aperture radar in the multi-source monitoring system in the data fusion observation area; the multi-source monitoring data in each data fusion observation area is acquired through the multi-source monitoring system at the layout density and the acquisition frequency.

[0104] Among them, the environmental adaptability score refers to the performance quantization value calculated by evaluating the monitoring applicability of each monitoring device under disturbance parameters such as the horizontal distance of the dam, slope gradient, satellite signal occlusion degree, vegetation coverage density, and potential geological hazard risk level corresponding to the data fusion observation area. For example, a device with a high adaptability score has good monitoring effects under the current environmental conditions and is suitable for a higher layout density or an increased sampling frequency; of course, a device with a lower adaptability score has poor application effects in this area, and the layout density or sampling frequency can be appropriately reduced. The specific calculation logic of the adaptability score in this exemplary embodiment is not particularly limited.

[0105] The deployment density refers to the quantity index of monitoring devices arranged within a unit area in the data fusion observation area. For example, the deployment density can be expressed by the number of GNSS receivers set per square kilometer; it can also be described based on the number of sampling points on each monitoring profile line. Of course, it can also be designed according to the rule that the environmental adaptability score is positively correlated with the deployment density, that is, the higher the adaptability score, the higher the deployment density, to enhance the monitoring sensitivity; it can also adopt a hierarchical deployment strategy according to resource constraints, only increasing the density in areas where the adaptability score is higher than a certain threshold, and adopting sparse deployment in other areas. The present exemplary embodiment does not make a special limitation on the representation method of the deployment density.

[0106] The acquisition frequency refers to the time interval or cycle index for data update acquisition of the same observation area. For example, the acquisition frequency can be expressed as once a day, twice a week, etc.; of course, it can also be described by a continuous sampling interval, such as sampling once every 10 minutes. The present exemplary embodiment does not make a special limitation on the setting method of the acquisition frequency. Optionally, the acquisition frequency can also be dynamically adjusted according to the environmental adaptability score. For example, a high sampling frequency is set in areas with a high adaptability score to improve the spatio-temporal continuity of monitoring data; of course, the acquisition frequency can be appropriately reduced in areas with a low adaptability score to reduce resource consumption and improve the overall operation efficiency of the system.

[0107] During the actual acquisition process, data acquisition control is carried out according to the set deployment density and acquisition frequency. For example, GNSS receivers can be deployed according to the set density in high-adaptability areas and continuously acquire position information in high-frequency mode; the surveying robot regularly inspects key monitoring points in high-adaptability areas according to the set survey plan; InSAR data preferentially updates high-frequency data for observation areas with a high adaptability score according to the orbit revisit period. In an optional implementation manner, an automated data acquisition system can be combined to uniformly schedule the acquisition plans and data transmission tasks of each device in the central control system to ensure that the data acquisition in each observation area meets the established density and frequency standards; of course, an abnormal trigger mechanism can also be introduced. When displacement acceleration or abnormal changes are detected in a certain area, the data acquisition frequency and density in that area are automatically increased to form a dynamic response acquisition system.

[0108] By dynamically determining the deployment density and acquisition frequency based on the environmental adaptability scores of each monitoring device in the data fusion observation area and acquiring multi-source monitoring data accordingly, it is possible to adaptively optimize the monitoring resource allocation according to the actual environmental conditions, so that while ensuring the observation accuracy and timeliness of key areas, the monitoring system can effectively reduce resource consumption and redundant burden, and significantly improve the environmental adaptability, monitoring sensitivity and overall operation efficiency of the dam slope deformation monitoring system.

[0109] In an exemplary embodiment of the present disclosure, the multi-source monitoring data includes monitoring point change monitoring data, surface deformation monitoring data, and local high-precision monitoring data. The multi-source monitoring data within the data fusion observation area can be collected through the following steps, which specifically may include: The monitoring point change monitoring data in the Earth-centered coordinate system can be collected by the global satellite signal receivers set at each monitoring point in the data fusion observation area; the surface deformation monitoring data in the image coordinate system can be collected by the interferometric synthetic aperture radar set in the data fusion observation area; and the local high-precision monitoring data in the local coordinate system can be collected by the total station set in the data fusion observation area.

[0110] Among them, the monitoring point change monitoring data is monitored and collected by the global satellite signal receivers set at each monitoring point in the data fusion observation area. A GNSS receiver is a device capable of receiving and processing GNSS signals from the Global Navigation Satellite System, and is used to measure the three-dimensional position change of the monitoring point in the Earth-centered, Earth-fixed (ECEF) coordinate system in real time or periodically. For example, a GNSS receiver can support multi-system observations such as GPS, GLONASS, Galileo, and BeiDou (Compass), achieving positioning accuracies at the sub-meter, decimeter, or even centimeter level. Of course, for deformation monitoring requirements, dual-band or multi-band receivers can also be selected to improve the anti-multipath interference ability and observation accuracy. In this exemplary embodiment, the specific model and configuration method of the GNSS receiver are not particularly limited. The continuously operating reference station (CORS) mode can be adopted to operate on fixed monitoring points for a long time to obtain high-precision time series displacement data. Of course, the mobile monitoring mode can also be adopted, where GNSS receivers are periodically set at different monitoring points for short-term measurements to achieve wide-area monitoring under limited resources.

[0111] In actual deployment, the GNSS receivers can be installed at key locations such as representative geological units, slip zones, crack concentration areas, and slope deformation warning areas in the dam slope monitoring sub-area to collect the monitoring point change monitoring data. The collected monitoring point change monitoring data can be represented by the three-dimensional coordinates XYZ in the Earth-centered coordinate system for subsequent point position change analysis, rate calculation, and deformation anomaly identification.

[0112] The surface deformation monitoring data is collected by an interferometric synthetic aperture radar (InSAR) set up in the data fusion observation area. InSAR is a remote sensing monitoring technology that can obtain surface images by transmitting and receiving microwave radar signals, and perform interferometric processing based on the phase differences between multi-temporal images to calculate surface deformation information. For example, InSAR can be used for regional large-scale surface monitoring based on spaceborne platforms (such as Sentinel-1, TerraSAR-X), or for high-precision continuous observation in a local area based on ground-based synthetic aperture radar (GBSAR). Of course, to meet different monitoring requirements, time series InSAR (TS-InSAR) or persistent scatterer InSAR (PS-InSAR) technology can be adopted to improve the spatio-temporal resolution and accuracy of the monitoring data. In this exemplary embodiment, the specific configuration and interferometric processing method of the InSAR system are not particularly limited.

[0113] During the acquisition process, the surface deformation data output by the InSAR system is represented in the image coordinate system (i.e., the radar perspective projection coordinate system), and subsequent coordinate transformation is required to unify it with other data sources. The surface deformation monitoring data can reflect the large-scale surface settlement, uplift, or horizontal displacement changes in the monitoring area, and is particularly suitable for continuous monitoring of slow cumulative deformation processes and identification of abnormal change trends.

[0114] The local high-precision monitoring data refers to the monitoring data collected by a total station set up in the data fusion observation area under the local coordinate system. A total station is a measuring device that can achieve high-precision angle measurement and distance measurement based on an automated control system, and then obtain spatial three-dimensional coordinate data. For example, a total station can be equipped with an automatic target recognition (ATR) module to achieve fast locking and continuous tracking measurement by recognizing a reflecting prism or natural feature targets. Of course, a total station can also be integrated with a three-dimensional laser scanning unit to collect high-density point cloud data in a local area to enhance the ability to capture fine-grained terrain changes. In this exemplary embodiment, the internal configuration and measurement method of the total station are not particularly limited.

[0115] During the data acquisition process, the total station measures the relative three-dimensional displacement of each monitoring prism or feature point in real time according to its set local coordinate system (usually with a fixed control point as the origin, and the horizontal and vertical axis directions are set). The data under the local coordinate system has high measurement accuracy and timeliness, and is particularly suitable for high-frequency and high-precision dynamic monitoring of key areas with severe local deformation, fast change rate, or subtle spatial changes.

[0116] Specifically, the unification of the monitoring coordinate systems for multi-source monitoring data can be achieved through the following steps, which may specifically include: A rotation matrix and a translation matrix can be constructed based on the position coordinates of the selected reference monitoring point, and the monitoring data of the monitoring point changes can be transformed into the local coordinate system through the rotation matrix and the translation matrix to obtain the unified monitoring data of the point position changes; the projection matrix can be solved based on the position coordinates and image coordinates of the reference monitoring point, and the surface deformation monitoring data can be transformed into the local coordinate system through the projection matrix to obtain the unified surface deformation monitoring data; the local high-precision monitoring data can be corrected for deviation and aligned based on the position coordinates of the reference monitoring point to obtain the unified local high-precision monitoring data; the unified monitoring data of the monitoring point changes, the unified surface deformation monitoring data, and the unified local high-precision monitoring data are subjected to consistency verification to obtain the multi-source monitoring data within the data fusion observation area.

[0117] Among them, the reference monitoring point refers to a specific monitoring point selected in the slope monitoring network, which has a stable position, excellent observation conditions and is representative, and is used as a reference benchmark for coordinate transformation and data unification. For example, the reference monitoring point can be selected as a GNSS continuous observation point located in a geologically stable area and with no significant displacement changes for a long time; of course, the best reference point can also be selected according to the long-term stability indicators of multiple candidate points. This exemplary embodiment does not particularly limit the selection criteria for the reference monitoring point. In some alternative embodiments, a main reference point and a backup reference point can be set, and automatic switching can be performed when the main reference point fails to ensure the consistency and continuity of the monitoring data.

[0118] The rotation matrix refers to a three-dimensional orthogonal matrix that describes the attitude transformation relationship between the local coordinate system and the geocentric coordinate system and is used to align the directions of the coordinate systems. For example, the rotation matrix can be calculated through the known position relationships of the reference monitoring point and its surrounding auxiliary points by using the three-point orientation in space or the least squares fitting method; of course, it can also be directly deduced based on the conversion relationship between the known geographic coordinate system (such as WGS84) and the local engineering coordinate system. This exemplary embodiment does not particularly limit the calculation method of the rotation matrix.

[0119] The translation matrix refers to a three-dimensional vector that describes the origin offset relationship between the local coordinate system and the geocentric coordinate system and is used to unify the position of the coordinate origin. For example, the translation matrix can be set as the position vector of the reference monitoring point in the geocentric coordinate system as the translation amount of the origin of the local coordinate system; of course, more accurate translation parameters can also be obtained by combining the overall network adjustment calculation. This exemplary embodiment does not particularly limit the calculation method of the translation matrix.

[0120] The image coordinates refer to the pixel spatial positions of the surface target points in the radar images acquired by the InSAR system, usually represented by row and column numbers or in-image coordinates (range, azimuth). For example, each surface point corresponds to a specific two-dimensional pixel position in the radar image, reflecting its spatial projection relationship within the radar field of view. Of course, the image coordinates can be further associated with the geographic coordinate system through image geometric correction (such as orthorectification). In this exemplary embodiment, there are no special limitations on the acquisition and expression methods of the image coordinates.

[0121] The projection matrix refers to the matrix mapping relationship that describes the conversion relationship between the image coordinate system and the local spatial rectangular coordinate system. For example, the projection matrix can be established based on the positions of known reference monitoring points in the local coordinate system and their corresponding image coordinates through methods such as least squares fitting, Homography Matrix solution, or bundle adjustment vector inversion. Of course, it can also be solved by multi-point joint calculation based on the Ground Control Points (GCP) system to improve the accuracy. In this exemplary embodiment, there are no special limitations on the projection matrix solution method.

[0122] In the actual conversion process, the projection relationship model between the local three-dimensional coordinates and the image two-dimensional coordinates can be established first using the known position information of the reference points. Subsequently, the projection matrix obtained by solving is applied to map and convert the surface deformation data output by the InSAR system from the image coordinate system to the local three-dimensional coordinate system, so that the surface deformation monitoring data and other data sources maintain a unified spatial reference benchmark.

[0123] The local high-precision monitoring data refers to the data set obtained by collecting with total stations set in the data fusion observation area, based on the local coordinate system, describing the three-dimensional displacement changes of monitoring prisms or feature points. For example, this type of data directly reflects the minute displacement changes in the local space and has the characteristics of high precision and high time resolution. Of course, the local high-precision monitoring data may also be affected by equipment initial orientation errors, station drift, or local measurement deviations, and requires unified reference correction. In this exemplary embodiment, there are no special limitations on the acquisition method of the local high-precision monitoring data.

[0124] Deviation correction refers to identifying and correcting systematic errors in local data in terms of translation, rotation, or scale by comparing the actual measurement results of control points in local monitoring data with the standard positions of reference monitoring points. For example, the overall translation vector and rotation matrix can be estimated based on the offsets of multiple common control points through the least squares adjustment method. Of course, a rigid transformation model can also be used to perform only translation and rotation corrections while ignoring scale changes. The deviation correction model is not particularly limited in this exemplary embodiment. In an alternative implementation, rigid translation correction can be performed based on a single reference point, or multi-parameter joint fitting correction can be performed based on multiple reference points to improve the overall correction accuracy and local consistency.

[0125] In specific implementation, through the corresponding analysis of the spatial positions of reference monitoring points in local high-precision monitoring data and their positions in the unified local coordinate system, correction parameters are calculated and applied to align the local high-precision monitoring data to the unified spatial reference, so that all monitoring data can be directly comparable under the same reference framework.

[0126] Consistency verification refers to comparing and analyzing the displacement observation results of different data sources at the same or similar spatial positions and time nodes in the unified coordinate system to verify the spatial correspondence, time synchronization, and deformation trend consistency among the data. For example, consistency verification can be performed by calculating the deviation between the displacement amounts of the same target point from different data sources and evaluating whether the deviation is within the preset tolerance range. Of course, the overall deformation trend consistency of each data source can also be analyzed statistically, such as linearly regressing and fitting the deformation curve and calculating the correlation coefficient. The specific method of consistency verification is not particularly limited in this exemplary embodiment. In an alternative implementation, a strict consistency threshold can be set for the monitoring requirements in high-security-level areas, or a loose consistency standard can be adopted for preliminary screening of potential abnormal points to improve the processing efficiency.

[0127] In specific implementation, the point position change data collected by the GNSS receiver can be used as the preliminary reference, and the InSAR surface deformation monitoring data and the local high-precision monitoring data of the total station are compared point by point. On the basis of time alignment, the displacement direction and amplitude are checked one by one to identify and eliminate abnormal data points that exceed the reasonable deviation range, and at the same time, the data sets with good consistency are marked as the input for subsequent data fusion processing.

[0128] By uniformly converting the monitoring data of the changes in monitoring points, the surface deformation monitoring data, and the local high-precision monitoring data in different coordinate systems to the local coordinate system and performing consistency verification processing, it is possible to effectively eliminate the spatial deviation and systematic error introduced by the difference in the observation reference of multi-source monitoring data, realize the consistency of the data spatial reference and the consistency of the deformation trend, significantly improve the accuracy of data fusion processing and the reliability of comprehensive analysis of multi-source deformation information, thereby providing a data basis with high precision, high stability, and high credibility for the slope deformation monitoring system.

[0129] Figure 5 Schematically shows a scene diagram of dividing a data fusion observation area in a dam slope area according to some embodiments of the present disclosure.

[0130] Reference Figure 5 As shown, the dam slope area can be divided into multiple monitoring sub-areas 510 through a preset division grid. The monitoring sub-areas 510 are formed by initially dividing the entire dam slope area to be monitored according to the preset division grid, and each monitoring sub-area is distributed in a regular grid manner in space, and is used to support subsequent further area classification processing based on disturbance parameters.

[0131] Subsequently, through the slope deformation disturbance parameters collected in each monitoring sub-area 510, such as the dam horizontal distance, slope gradient, satellite signal occlusion degree, vegetation coverage density, and potential geological disaster risk level included in the slope deformation disturbance parameters, combined with the area classification and merging method provided in the embodiments of the present specification, each monitoring sub-area 510 is classified and fused, thereby forming multiple data fusion observation areas. As Figure 5 shown, the first data fusion observation area 520, the second data fusion observation area 530, and the third data fusion observation area 540 are finally generated.

[0132] For example, for the first data fusion observation area 520, it can schematically represent an area with a relatively high potential geological disaster risk level, a relatively large slope gradient, and relatively less GNSS signal occlusion. In this area, both GNSS receivers and surveying robots have relatively high environmental adaptability scores. Therefore, in the data fusion process, higher fusion weights are assigned to these two types of data, and higher layout densities and acquisition frequencies are set to enhance the timeliness response to sudden slips and rapid deformations; at the same time, since InSAR may be affected by multipath interference in the near-dam high-reflection area, its data adaptability score is relatively low, so its weight is appropriately reduced in this area.

[0133] For the second data fusion observation area 530, it can schematically represent an area where the characteristics of the disturbance parameters are between high risk and low risk, and the overall environmental adaptability score is at a medium level. In this area, the adaptability scores of the three types of monitoring devices are relatively close. Therefore, a weight balancing strategy is adopted in the data fusion process, that is, the data collected by GNSS, InSAR, and total station are weighted and fused according to the normalized score ratio to ensure that various types of data play complementary advantages in the fusion, thereby constructing a stable and moderately redundant deformation perception ability.

[0134] For the third data fusion observation area 540, it can schematically represent an area with gentle terrain, weak geological disturbance, serious signal occlusion, and dense vegetation coverage, resulting in low adaptability scores of GNSS receivers and total stations. Since InSAR has the ability of wide-area coverage and can achieve large-scale and low-cost monitoring in such areas, the weight of InSAR data is increased in the fusion process, while the participation of data from other devices is appropriately reduced, and the acquisition frequency and layout density are reduced to control resource consumption and improve monitoring efficiency.

[0135] Through the above differential data fusion strategy, a customized monitoring strategy configuration for different data fusion observation areas can be realized, thereby effectively improving the overall performance of the multi-source monitoring system in terms of spatial resource allocation, data fusion accuracy, and disaster perception ability, and enhancing the robustness and accuracy self-adaptability of the monitoring system under multiple environmental conditions.

[0136] It should be noted that although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that these steps must be executed in that specific order, or that all the steps shown must be executed to achieve the desired result. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution, etc.

[0137] In addition, in the present exemplary embodiment, a slope deformation monitoring device based on multi-source data fusion is also provided. Referring to Figure 6 As shown, the slope deformation monitoring device 600 based on multi-source data fusion includes: a multi-source monitoring module 610, a monitoring selection area division module 620, a fusion weight module 630, a multi-source monitoring data acquisition module 640, and a slope deformation analysis module 650. Among them: The multi-source monitoring module 610 is used to deploy a monitoring coverage network composed of a global satellite signal receiver, a total station, and an interferometric synthetic aperture radar in the dam slope area to be monitored, forming a multi-source monitoring system; The monitoring selection area division module 620 is used to divide the monitoring selection areas of the dam slope area according to the collected slope deformation disturbance parameters to obtain multiple data fusion observation areas; A fusion weight module 630, configured to determine data fusion weights corresponding to each of the data fusion observation regions; A multi-source monitoring data acquisition module 640, configured to periodically acquire multi-source monitoring data within each of the data fusion observation regions through the multi-source monitoring system; A slope deformation analysis module 650, configured to perform weighted fusion processing on the acquired multi-source monitoring data based on the data fusion weights of each of the data fusion observation regions, and generate slope deformation monitoring results corresponding to each of the data fusion observation regions.

[0138] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the monitoring selection area division module 620 is configured to: preliminarily divide the dam slope area into multiple monitoring sub-areas according to a preset division grid; classify the monitoring sub-areas based on the slope deformation disturbance parameters in each of the monitoring sub-areas, and merge and fuse the monitoring sub-areas belonging to the same classification to obtain multiple data fusion observation regions.

[0139] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the slope deformation disturbance parameters include the dam horizontal distance, slope gradient, satellite signal occlusion degree, vegetation coverage density, and potential geological disaster risk level at each position coordinate in the dam slope area; the monitoring selection area division module 620 is configured to: determine the quantitative scores of the dam horizontal distance, the slope gradient, the satellite signal occlusion degree, the vegetation coverage density, and the potential geological disaster risk level in each of the monitoring sub-areas; fuse the quantitative scores according to the weight ratios of the slope deformation disturbance parameters to obtain the area classification scores of each of the monitoring sub-areas; classify the monitoring sub-areas through the area classification scores, and merge and fuse the monitoring sub-areas belonging to the same classification to obtain multiple data fusion observation regions.

[0140] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the slope deformation monitoring device 600 based on multi-source data fusion further includes a slope deformation disturbance parameter acquisition module, which is configured to: obtain the topographic elevation data of the dam slope area, and determine the horizontal distance of the dam from each position coordinate in the dam slope area according to the topographic elevation data; calculate the slope of each dam slope area through the topographic elevation data to determine the slope of the slope at each position coordinate; obtain the signal occlusion test result based on the unmanned aerial vehicle, and determine the satellite signal occlusion degree at each position coordinate according to the signal strength in the signal occlusion test result; calculate the normalized difference vegetation index through multi-temporal remote sensing images of the dam slope area, generate a vegetation coverage distribution map based on the normalized difference vegetation index, and determine the vegetation coverage density at each position coordinate through the vegetation coverage distribution map; determine the potential risk level of geological disasters at each position coordinate according to the historical geological disaster data and geological exploration results corresponding to the dam slope area.

[0141] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the fusion weight module 630 is configured to: obtain the environmental impact factors of each monitoring device in the multi-source monitoring system under the horizontal distance of the dam, the slope of the slope, the satellite signal occlusion degree, the vegetation coverage density, and the potential risk level of geological disasters; calculate the environmental adaptability scores of each monitoring device in the current data fusion observation area through the slope deformation disturbance parameters and the environmental impact factors in the data fusion observation area; perform normalization processing on the environmental adaptability scores of each monitoring device to obtain the data fusion weight corresponding to the data fusion observation area.

[0142] In some exemplary embodiments of the present disclosure, based on the foregoing solution, before performing normalization processing on the environmental adaptability scores of each monitoring device to obtain the data fusion weight corresponding to the data fusion observation area, the slope deformation monitoring device 600 based on multi-source data fusion further includes an environmental adaptability score adjustment module, which is configured to: if it is determined that the environmental adaptability score of the current monitoring device is less than or equal to a preset score threshold, set the environmental adaptability score of the current monitoring device to zero.

[0143] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the multi-source monitoring data acquisition module 640 is configured to: determine the layout density and acquisition frequency in the data fusion observation area according to the environmental adaptability scores of the global satellite signal receiver, the total station, and the interferometric synthetic aperture radar in the multi-source monitoring system in the data fusion observation area; collect multi-source monitoring data in each data fusion observation area through the multi-source monitoring system at the layout density and acquisition frequency.

[0144] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the multi-source monitoring data includes monitoring point change monitoring data, surface deformation monitoring data, and local high-precision monitoring data. The multi-source monitoring data acquisition module 640 is configured to: collect the monitoring point change monitoring data in the geocentric coordinate system through the global satellite signal receivers arranged at each monitoring point in the data fusion observation area; collect the surface deformation monitoring data in the image coordinate system through the interferometric synthetic aperture radar arranged in the data fusion observation area; and collect the local high-precision monitoring data in the local coordinate system through the total stations arranged in the data fusion observation area.

[0145] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the slope deformation monitoring device 600 based on multi-source data fusion further includes a coordinate unification module, which is configured to: construct a rotation matrix and a translation matrix based on the position coordinates of the selected reference monitoring point, and convert the monitoring point change monitoring data to the local coordinate system through the rotation matrix and the translation matrix to obtain the unified point position change monitoring data; solve the projection matrix based on the position coordinates and image coordinates of the reference monitoring point, and convert the surface deformation monitoring data to the local coordinate system through the projection matrix to obtain the unified surface deformation monitoring data; perform deviation correction and alignment on the local high-precision monitoring data based on the position coordinates of the reference monitoring point to obtain the unified local high-precision monitoring data; and perform consistency verification on the unified monitoring point change monitoring data, the unified surface deformation monitoring data, and the unified local high-precision monitoring data to obtain the multi-source monitoring data within the data fusion observation area.

[0146] The specific details of each module of the slope deformation monitoring device based on multi-source data fusion have been described in detail in the corresponding slope deformation monitoring method based on multi-source data fusion, and thus will not be elaborated herein.

[0147] It should be noted that although several modules or units of the slope deformation monitoring device based on multi-source data fusion are mentioned in the foregoing detailed description, such a division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of the two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0148] In addition, in the exemplary embodiments of the present disclosure, an electronic device capable of implementing the above-mentioned slope deformation monitoring method based on multi-source data fusion is also provided.

[0149] Those skilled in the art can understand that various aspects of the present disclosure can be implemented as a system, a method, or a program product. Therefore, various aspects of the present disclosure can be specifically implemented in the following forms, namely: a complete hardware embodiment, a complete software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software aspects, which can be collectively referred to as "circuitry", "module", or "system" here.

[0150] Reference is made below Figure 7 to describe the electronic device 700 according to such an embodiment of the present disclosure. Figure 7 The illustrated electronic device 700 is merely an example and should not impose any limitation on the functions and usage scope of the embodiments of the present disclosure.

[0151] As Figure 7 shown, the electronic device 700 is presented in the form of a general-purpose computing device. The components of the electronic device 700 may include, but are not limited to: at least one of the above-mentioned processing units 710, at least one of the above-mentioned storage units 720, a bus 730 connecting different system components (including the storage unit 720 and the processing unit 710), and a display unit 740.

[0152] Among them, the storage unit stores program code, and the program code can be executed by the processing unit 710, so that the processing unit 710 executes the steps according to various exemplary embodiments of the present disclosure described in the above "Exemplary Method" section of this specification. For example, the processing unit 710 can execute steps such as Figure 1 shown in S110, deploying a monitoring coverage network composed of a global satellite signal receiver, a total station robot, and an interferometric synthetic aperture radar in the dam slope area to be monitored to form a multi-source monitoring system; step S120, dividing the monitoring selection area of the dam slope area according to the collected slope deformation disturbance parameters to obtain a plurality of data fusion observation areas; step S130, determining the data fusion weights corresponding to each of the data fusion observation areas; step S140, periodically collecting multi-source monitoring data in each of the data fusion observation areas through the multi-source monitoring system; step S150, respectively performing weighted fusion processing on the collected multi-source monitoring data based on the data fusion weights of each of the data fusion observation areas to generate slope deformation monitoring results corresponding to each of the data fusion observation areas.

[0153] The storage unit 720 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 721 and / or a cache storage unit 722, and may further include a read-only storage unit (ROM) 723.

[0154] The storage unit 720 may also include a program / utility 724 having a set (at least one) of program modules 725. Such program modules 725 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment.

[0155] The bus 730 may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus structures.

[0156] The electronic device 700 may also communicate with one or more external devices 770 (such as a keyboard, a pointing device, a Bluetooth device, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 700, and / or may communicate with any device that enables the electronic device 700 to communicate with one or more other computing devices (such as a router, a modem, etc.). Such communication may be through the input / output (I / O) interface 750. And, the electronic device 700 may also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through the network adapter 760. As shown in the figure, the network adapter 760 communicates with other modules of the electronic device 700 through the bus 730. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 700, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0157] Through the description of the above embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software, or can be implemented by a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which may be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which may be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.

[0158] In an exemplary embodiment of the present disclosure, there is also provided a computer-readable storage medium, on which a program product capable of implementing the above methods in this specification is stored. In some possible embodiments, various aspects of the present disclosure may also be implemented in the form of a program product, which includes program code. When the program product runs on a terminal device, the program code is used to cause the terminal device to execute the steps according to various exemplary embodiments of the present disclosure described in the "Exemplary Method" section above in this specification.

[0159] Reference Figure 8 As shown, a program product 800 for implementing the above slope deformation monitoring method based on multi-source data fusion according to an embodiment of the present disclosure is described. It may be a portable compact disc read-only memory (CD-ROM) and includes program code, and can run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited thereto. In this document, a readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0160] The program product may adopt any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may, for example, but not be limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0161] The computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries the readable program code. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable signal medium may also be any readable medium other than the readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0162] The program code contained on the readable medium may be transmitted by any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination of the above.

[0163] Program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and also including conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user's computing device, partly on the user's device, executed as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., by connecting through the Internet using an Internet service provider).

[0164] In addition, the above-mentioned drawings are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, rather than for limiting purposes. It is easy to understand that the processes shown in the above-mentioned drawings do not indicate or limit the chronological order of these processes. Additionally, it is also easy to understand that these processes may be executed synchronously or asynchronously in, for example, multiple modules.

[0165] Through the description of the above embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software or in a manner of software combined with necessary hardware. Therefore, the technical solution according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which may be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which may be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.

[0166] Those skilled in the art will readily think of other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure, which follow the general principles of the present disclosure and include known common general knowledge or conventional technical means in the technical field not disclosed in the present disclosure. The specification and embodiments are only regarded as exemplary, and the true scope and spirit of the present disclosure are pointed out by the claims.

[0167] It should be understood that the present disclosure is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present disclosure is only limited by the appended claims.

Claims

1. A slope deformation monitoring method based on multi-source data fusion, characterized in that Including: Deploy a monitoring coverage network consisting of a global satellite signal receiver, a surveying robot, and an interferometric synthetic aperture radar within the dam slope area to be monitored, forming a multi-source monitoring system; According to the collected slope deformation disturbance parameters, conduct monitoring area selection and division for the dam slope area to obtain multiple data fusion observation areas; Determine the data fusion weights corresponding to each of the data fusion observation areas; Periodically collect multi-source monitoring data within each of the data fusion observation areas through the multi-source monitoring system; Based on the data fusion weights of each of the data fusion observation areas, perform weighted fusion processing on the collected multi-source monitoring data respectively to generate slope deformation monitoring results corresponding to each of the data fusion observation areas.

2. The slope deformation monitoring method based on multi-source data fusion according to claim 1, characterized in that The step of conducting monitoring area selection and division for the dam slope area according to the collected slope deformation disturbance parameters to obtain multiple data fusion observation areas includes: Conduct preliminary area division on the dam slope area according to a preset division grid to obtain multiple monitoring sub-areas; Based on the slope deformation disturbance parameters in each of the monitoring sub-areas, classify the monitoring sub-areas, and merge and fuse the monitoring sub-areas belonging to the same classification to obtain multiple data fusion observation areas.

3. The slope deformation monitoring method based on multi-source data fusion according to claim 2, characterized in that The slope deformation disturbance parameters include the dam horizontal distance, slope gradient, satellite signal occlusion degree, vegetation coverage density, and potential geological hazard risk level at each position coordinate in the dam slope area; The step of classifying the monitoring sub-areas based on the slope deformation disturbance parameters in each of the monitoring sub-areas, and merging and fusing the monitoring sub-areas belonging to the same classification to obtain multiple data fusion observation areas includes: Determine the quantitative scores of the dam horizontal distance, slope gradient, satellite signal occlusion degree, vegetation coverage density, and potential geological hazard risk level in each of the monitoring sub-areas; Fuse the quantitative scores according to the weight ratios of each of the slope deformation disturbance parameters to obtain the area classification scores of each of the monitoring sub-areas; Classify the monitoring sub-areas through the area classification scores, and merge and fuse the monitoring sub-areas belonging to the same classification to obtain multiple data fusion observation areas.

4. The slope deformation monitoring method based on multi-source data fusion according to claim 3, characterized in that The method further includes: Obtain the terrain elevation data of the dam slope area, and determine the dam horizontal distance from each position coordinate in the dam slope area to the dam main body according to the terrain elevation data; Calculate the slope gradient of each dam slope area through the terrain elevation data to determine the slope gradient of each position coordinate; Obtain the signal occlusion test results based on an unmanned aerial vehicle, and determine the satellite signal occlusion degree at each position coordinate according to the signal intensity in the signal occlusion test results; Calculate the normalized difference vegetation index through multi-period remote sensing images of the dam slope area, generate a vegetation coverage distribution map based on the normalized difference vegetation index, and determine the vegetation coverage density at each position coordinate through the vegetation coverage distribution map; Determine the potential geological hazard risk level of each position coordinate according to the historical geological hazard data and geological exploration results corresponding to the dam slope area.

5. The slope deformation monitoring method based on multi-source data fusion according to claim 1, characterized in that Determining the data fusion weights corresponding to each of the data fusion observation regions includes: Obtaining the environmental impact factors of each monitoring device in the multi-source monitoring system under the horizontal distance of the dam, slope gradient, satellite signal occlusion degree, vegetation coverage density, and potential geological disaster risk level; Calculating the environmental adaptability scores of each monitoring device in the current data fusion observation region based on the slope deformation disturbance parameters and the environmental impact factors in the data fusion observation region; Normalizing the environmental adaptability scores of each monitoring device to obtain the data fusion weights corresponding to the data fusion observation region.

6. The slope deformation monitoring method based on multi-source data fusion according to claim 5, characterized in that Before normalizing the environmental adaptability scores of each monitoring device to obtain the data fusion weights corresponding to the data fusion observation region, the method further includes: If it is determined that the environmental adaptability score of the current monitoring device is less than or equal to a preset score threshold, the environmental adaptability score of the current monitoring device is set to zero.

7. The slope deformation monitoring method based on multi-source data fusion according to claim 5, characterized in that The periodically collecting multi-source monitoring data in each of the data fusion observation regions by the multi-source monitoring system includes: Determining the deployment density and collection frequency in the data fusion observation region according to the environmental adaptability scores of the global satellite signal receiver, total station, and interferometric synthetic aperture radar in the multi-source monitoring system in the data fusion observation region; Collecting multi-source monitoring data in each of the data fusion observation regions through the multi-source monitoring system at the deployment density and collection frequency.

8. The slope deformation monitoring method based on multi-source data fusion according to claim 1, characterized in that The multi-source monitoring data includes monitoring point change monitoring data, surface deformation monitoring data, and local high-precision monitoring data. The collecting multi-source monitoring data in each of the data fusion observation regions includes: Collecting monitoring point change monitoring data in the geocentric coordinate system through the global satellite signal receivers set at each monitoring point in the data fusion observation region; Collecting surface deformation monitoring data in the image coordinate system through the interferometric synthetic aperture radar set in the data fusion observation region; Collecting local high-precision monitoring data in the local coordinate system through the total stations set in the data fusion observation region.

9. The slope deformation monitoring method based on multi-source data fusion according to claim 8, characterized in that The method further includes: Constructing a rotation matrix and a translation matrix based on the position coordinates of the selected reference monitoring point, and converting the monitoring point change monitoring data to the local coordinate system through the rotation matrix and the translation matrix to obtain the unified point position change monitoring data; Solving a projection matrix based on the position coordinates and image coordinates of the reference monitoring point, and converting the surface deformation monitoring data to the local coordinate system through the projection matrix to obtain the unified surface deformation monitoring data; Performing deviation correction and alignment on the local high-precision monitoring data based on the position coordinates of the reference monitoring point to obtain the unified local high-precision monitoring data; Performing consistency verification on the unified monitoring point change monitoring data, the unified surface deformation monitoring data, and the unified local high-precision monitoring data to obtain the multi-source monitoring data in the data fusion observation region.

10. A slope deformation monitoring device based on multi-source data fusion, characterized in that, Including: A multi-source monitoring module, which is used to deploy a monitoring coverage network composed of a global satellite signal receiver, a surveying robot, and an interferometric synthetic aperture radar in the dam slope area to be monitored, so as to form a multi-source monitoring system; A monitoring selection area division module, which is used to divide the monitoring selection areas of the dam slope area according to the collected slope deformation disturbance parameters to obtain multiple data fusion observation areas; A fusion weight module, which is used to determine the data fusion weights corresponding to the respective data fusion observation areas; A multi-source monitoring data acquisition module, which is used to periodically acquire multi-source monitoring data in the respective data fusion observation areas through the multi-source monitoring system; A slope deformation analysis module, which is used to perform weighted fusion processing on the acquired multi-source monitoring data based on the data fusion weights of the respective data fusion observation areas, and generate slope deformation monitoring results corresponding to the respective data fusion observation areas.

Images