A slope remote video monitoring stability analysis system and method

Abstract

This invention relates to a remote video monitoring stability analysis system and method for slopes. The method includes the following steps: S1: Constructing a three-dimensional unsteady viscoelastic-plastic shear creep model and acquiring slope stability data under rainfall; S2: Calculating the shear creep rate on the landslide body, defining the main slip line and the sliding section, and determining the blocking section; S3: Determining the nonlinear mapping relationship between the slope surface displacement or crack width and the shear creep rate at the corresponding blocking section; S4: Setting up sensors and targets based on the main slip line; S5: Acquiring the displacement or crack width of the targets; S6: Outputting to a remote computer; S7: Obtaining the shear creep rate at the blocking section corresponding to the targets; S8: Judging the slope stability based on the shear creep rate at the blocking section corresponding to the targets. Compared with the prior art, this invention has the advantages of realizing dynamic evaluation of soil landslide stability and improving the accuracy of instability prediction.

Description

Technical Field

[0001] This invention relates to the field of soil landslides and slope engineering, and in particular to a remote video monitoring and stability analysis system and method for slopes. Background Technology

[0002] Currently, some research has been conducted on the early warning system for landslide disaster risks caused by rainfall:

[0003] (1) Based on rainfall data, a landslide risk early warning model was established. A macro-risk early warning and forecasting model for regional landslide disasters was established based on an effective rainfall model. According to the risk zoning results at different scales and the three types of rainfall thresholds, early warning indicators and early warning levels were set respectively, and a regional loess landslide risk early warning calculation and analysis model was developed.

[0004] (2) Landslide risk early warning is conducted based on meteorological information data such as rainfall. A dynamic risk forecast for landslides is made using a landslide disaster risk prediction map based on WEBGIS and combined with real-time regional rainfall information. The probability of rainfall-induced landslides in a region is calculated based on the spatial probability of landslides and the temporal probability of rainfall-induced landslides, and risk warnings are issued in accordance with risk zones. Based on the matter-element theory in extension theory and referring to the geological disaster meteorological risk early warning classification, a landslide risk early warning level table is determined.

[0005] (3) Based on measured displacement data, landslide risk early warning is conducted. A landslide early warning model based on dynamic prediction of future velocity states is established using Markov chain theory. In-situ monitoring data is integrated into the engineering risk rate analysis method, and a real-time risk rate quantification model and early warning method for reservoir bank slope operation are proposed. A landslide trend-velocity ratio early warning criterion is established, and a fusion method of trend-velocity ratio and displacement-velocity ratio is used to determine changes in landslide risk.

[0006] (4) Employ artificial intelligence, machine learning, and deep learning algorithms for landslide disaster risk early warning. Establish a meteorological risk early warning method for landslide disasters based on BP neural networks. Develop a fuzzy neural network risk identification model to identify the risk level of soft rock slopes based on meteorological conditions, topographic factors, soil and rock properties, and measured horizontal displacement. Combine spatial convolutional neural networks, fuzzy neural networks, and other technologies to achieve geological disaster early warning.

[0007] Currently, domestic and international research has focused on intelligent early warning and forecasting technologies for slope hazards: Real-time monitoring has been conducted using the BeiDou high-precision displacement and deformation monitoring system, primarily monitoring rainfall, soil moisture content, and landslide deformation; a GPS multi-antenna monitoring system for highway slope deformation has been established, achieving high precision, automation, and all-weather monitoring of highway slope disasters; a slope disaster information processing system based on "3S" (Remote Sensing RS, Geographic Information System GIS, and Global Positioning System GPS) has achieved all-weather automatic monitoring; research has been conducted on the generation of slope disaster early warning information; based on slope characteristics, usage conditions, and the economic impact of damage, a reliability map method for slope monitoring systems has been constructed, evaluating the effectiveness and reliability of the monitoring system in predicting critical slope failure, based on the expected deformation failure and critical readout frequency of slopes; an early warning system for debris flows and slope damage has been established using radial basis function rainfall indices; and an evolutionary polynomial regression method for predicting the stability of soil and rock slopes has been proposed, combining genetic algorithms and the least squares method. Currently, in disaster early warning technology, based on the principle of fuzzy hierarchical analysis, a predictive model for the instability of highway cutting slopes has been established, and corresponding slope instability prediction and risk assessment software has been developed. Furthermore, the grey relational analysis principle is used to analyze the weights of factors affecting roadbed stability under rainfall conditions, and corresponding preventative measures are proposed. High-precision sensors are used to monitor key indicators such as slope surface displacement, rainfall, and soil moisture content in real time. Once the monitored data exceeds a preset safety threshold, the system automatically triggers an early warning mechanism. However, these methods require extensive expert experience and suffer from shortcomings such as insufficient basis for slope monitoring point placement, unreasonable stability assessment, immature instability prediction, low accuracy, and cumbersome implementation. Therefore, developing a remote video monitoring stability analysis system and method for slopes with high precision, accuracy, and reliability is particularly urgent.

[0008] The above method has drawbacks:

[0009] The indicators used in the above methods rely on expert experience and cannot reliably and accurately assess the time of slope instability, i.e., the stability of the slope. They also cannot accurately define the potential main sliding line, sliding surface or sliding zone, and the slope's anti-slip section. As a result, the slope monitoring points do not detect key data for predicting potential landslides, and the instability prediction accuracy is low. Summary of the Invention

[0010] The purpose of this invention is to provide a remote video monitoring stability analysis system and method for slopes to achieve dynamic evaluation of soil landslide stability and improve the accuracy of instability prediction. This system calculates the rate of change of shear creep rate of the slope soil, defines the potential main sliding line, sliding surface or sliding zone, and anti-slip section of the slope based on the rate of change of shear creep rate, detects key data for predicting potential landslides, and makes the obtained data more reflective of changes in slope stability. It also calculates the rate of change of shear creep rate at the corresponding anti-slip section, more accurately reflecting the actual dynamic changes in soil slope stability, and facilitating timely dynamic evaluation of slope stability under rainfall conditions.

[0011] The objective of this invention can be achieved through the following technical solutions:

[0012] A method for remote video monitoring and stability analysis of slopes, comprising the following steps:

[0013] S1: Construct a three-dimensional unsteady viscoelastic-plastic shear creep model to obtain slope stability data under rainfall.

[0014] S2: Calculate the shear creep rate on the landslide body based on the slope stability data under rainfall, define the potential main sliding line and sliding part based on the shear creep rate, and define the last segment that is connected in the sliding part as the blocking segment.

[0015] S3: Determine the nonlinear mapping relationship between the slope surface displacement or crack width at the projected location of the anti-slip segment in the sliding section and the shear creep rate at the corresponding anti-slip segment:

[0016] S4: Sensors and targets are set based on the main slide rail;

[0017] S5: Obtain the displacement of the target or the crack width;

[0018] S6: Output the displacement or crack width to a remote computer;

[0019] S7: The remote computer obtains the shear creep rate at the corresponding anti-slip section of the target based on the target displacement or crack width and nonlinear mapping relationship.

[0020] S8: Determine slope stability based on the shear creep rate at the anti-slip section corresponding to the target, and obtain the slope stability analysis results.

[0021] Furthermore, the specific steps of S1 are as follows:

[0022] S1-1: Construct a three-dimensional unsteady viscoelastic-plastic shear creep model;

[0023] S1-2: Obtain the data required for the model through triaxial creep tests of unsaturated / saturated soil;

[0024] S1-3: Based on the model and the parameters required by the model, a three-dimensional finite element numerical simulation of slope stability under rainfall is performed to obtain slope stability data under rainfall.

[0025] Furthermore, the data required for the model include permeability coefficient, suction, cohesion, internal friction angle, long-term strength, elastic modulus, viscosity coefficient, and soil yield stress σ. f Ultimate strength τ of soil in flow state T And creep calculation parameters.

[0026] Furthermore, the specific steps of S2 are as follows:

[0027] S2-1: Calculate the shear creep rate based on the lower slope stability data, and define the longitudinal line with the fastest change in shear creep rate on the landslide body as the potential main sliding line, which represents the main sliding direction of the overall slope sliding.

[0028] S2-2: A continuous surface or zone in a landslide body with a rate of change of shear creep rate greater than 0 is defined as a sliding surface or sliding zone.

[0029] S2-3: Define the last section that is connected in the sliding section as the slope anti-slip section;

[0030] S2-4: Determine the blocking segment in the sliding section and the projection location of the blocking segment on the ground surface, the potential main sliding line, and the range of possible slope instability and failure.

[0031] Furthermore, the shear creep rate is:

[0032]

[0033] Where, in the formula, Let σ be the shear creep rate, and σ be the current stress. f η is the yield stress of the soil, η2 is the viscosity coefficient of unsteady shear creep, A1, A2, m, and n are creep calculation parameters, and τ is the viscosity coefficient of unsteady shear creep. T It represents the ultimate strength of the soil in its flow state; t represents the creep time history.

[0034] Furthermore, the specific steps of S4 are as follows:

[0035] S4-1: Install non-contact intelligent sensing sensors and machine vision deformation meters within a certain range of the slope;

[0036] S4-2: Install targets on the ground surface at the projection location of the slope's anti-slip section within a certain range of the potential main sliding line and its vicinity.

[0037] Furthermore, the specific steps of S5 are as follows:

[0038] S5-1: Real-time preliminary frame images of slope deformation captured by non-contact intelligent sensing sensors;

[0039] S5-2: Establish an optimization model for a Bayesian-convolutional neural network by inputting the initial frame images into the optimization model;

[0040] S5-3: Optimize the model to obtain the best feature point sequence and output a high-resolution image;

[0041] S5-4: Based on a machine vision deformer, acquire the displacement or crack width of the target and video data from high-resolution images.

[0042] Furthermore, the specific steps of S6 are as follows:

[0043] S6-1: A data acquisition station powered by solar energy is set up at a certain location of the machine vision deformer, and the data acquisition station is connected to the machine vision deformer via a fiber optic cable.

[0044] S6-2: Transmit the displacement of the target or the crack width to a remote computer as a motion signal.

[0045] Furthermore, the specific steps of S8 are as follows:

[0046] S8-1: If the rate of change of shear creep rate at the anti-slip section corresponding to the target is less than 0, the slope is considered to be stabilizing.

[0047] S8-2: If the rate of change of shear creep rate at the anti-slip section corresponding to the target is equal to 0, it is considered that the slope will tend to fail.

[0048] S8-3: If the rate of change of shear creep rate at the anti-slip section corresponding to the target is greater than 0, it is considered that the slope is about to become unstable and fail.

[0049] In another aspect, the present invention provides a slope remote video monitoring stability analysis system, including a memory, a processor, and a program stored in the memory, wherein the processor executes the program to implement the method described above.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] (1) This invention establishes a three-dimensional unsteady viscoelastic-plastic shear creep model, calculates the rate of change of shear creep rate of slope soil, defines the potential main sliding line and sliding surface or sliding zone of slope and the slope blocking section according to the rate of change of shear creep rate, and conducts dynamic evaluation of soil landslide stability, making the evaluation results of this invention more accurate and effective, and providing strong technical support for predicting the occurrence of landslide disasters.

[0052] (2) Based on the slope potential main slip line and slip surface or slip zone and slope anti-slip section defined by the shear creep rate change rate, the present invention installs a machine vision deformer equipped with non-contact intelligent sensing sensors and integrates machine vision and video monitoring technology and sets up targets, so that the obtained data can better reflect the changes in slope stability.

[0053] (3) Based on the target displacement or crack width obtained by the machine vision deformation instrument equipped with non-contact intelligent sensing sensor and integrating machine vision and video monitoring technology, the present invention calculates the rate of change of shear creep rate at the corresponding anti-slip section, which more accurately reflects the actual situation of dynamic changes in soil slope stability and is conducive to timely dynamic evaluation of slope stability under rainfall conditions. Attached Figure Description

[0054] Figure 1 This is a schematic diagram of the process of the present invention;

[0055] Figure 2 This is a flowchart illustrating the internal calculation process of the three-dimensional unsteady viscoelastic-plastic shear-creep model of the present invention. Detailed Implementation

[0056] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0057] This invention proposes a remote video monitoring and stability analysis system and method for slopes. The flowchart of this invention is as follows: Figure 1 As shown, this invention includes: three-dimensional unsteady viscoelastic-plastic shear creep finite element numerical simulation; determining the resisting segments in the potential sliding surface and their projected positions on the ground surface, the potential main slip line, and the range of possible slope instability and failure; establishing a nonlinear mapping relationship between the slope surface displacement or crack width at the projected position of the resisting segment in the potential sliding surface and the shear creep rate at the corresponding resisting segment; installing a machine vision deformable device and a target equipped with non-contact intelligent sensing sensors and integrating machine vision and video monitoring technologies; acquiring the displacement or crack width of the target; automatically transmitting the acquired displacement or crack width and video data to a remote client computer in real time via IoT virtual IP technology; calculating the shear creep rate at the corresponding resisting segment of the target; and dynamically evaluating slope stability. Compared with existing technologies, this invention has advantages such as high precision, high accuracy, and high reliability.

[0058] The method of the present invention includes the following steps:

[0059] S1: Construct a three-dimensional unsteady viscoelastic-plastic shear creep model to obtain slope stability data under rainfall.

[0060] S2: Calculate the shear creep rate on the landslide body based on the slope stability data under rainfall, define the potential main sliding line and sliding part based on the shear creep rate, and define the last segment that is connected in the sliding part as the blocking segment.

[0061] S3: Determine the nonlinear mapping relationship between the slope surface displacement or crack width at the projected location of the anti-slip segment in the sliding section and the shear creep rate at the corresponding anti-slip segment:

[0062] S4: Sensors and targets are set based on the main slide rail;

[0063] S5: Obtain the displacement of the target or the crack width;

[0064] S6: Output the displacement or crack width to a remote computer;

[0065] S7: The remote computer obtains the shear creep rate at the corresponding anti-slip section of the target based on the target displacement or crack width and nonlinear mapping relationship.

[0066] S8: Determine slope stability based on the shear creep rate at the anti-slip section corresponding to the target, and obtain the slope stability analysis results.

[0067] In S1, a dedicated NGXJCRM program module can be developed (the program module performs calculations such as...). Figure 2 (as shown) and embedded in ABAQUS software for three-dimensional unsteady viscoelastic-plastic shear creep finite element numerical simulation;

[0068] S1-1: Develop a material subroutine NGXJCRM module for a three-dimensional unsteady viscoelastic-plastic shear creep model in ABAQUS software using creep element model theory, and embed the NGXJCRM module into ABAQUS software.

[0069] S1-2: Obtain permeability coefficient, suction, cohesion, internal friction angle, long-term strength, and elastic modulus, viscosity coefficient, and soil yield stress σ required by the NGXJCRM dedicated module through triaxial creep tests on unsaturated / saturated soil. f Ultimate strength τ of soil in flow state T Creep calculation parameters (A1, A2, m, and n);

[0070] S1-3: Using ABAQUS finite element software with the NGXJCRM dedicated module developed in S1-1 embedded, a three-dimensional finite element numerical simulation of slope stability under rainfall was performed.

[0071] In S2, the blocking section in the potential sliding surface and its projection position on the ground surface, the potential main sliding line, and the range of possible slope instability and failure are determined by S1.

[0072] S2-1: The longitudinal line with the fastest change in shear creep rate on the landslide body is defined as the potential principal sliding line, representing the main sliding direction of the overall slope:

[0073] Shear creep rate ε VP Calculation formula:

[0074]

[0075] In the formula, Let σ be the shear creep rate, and σ be the current stress. f η is the yield stress of the soil, η2 is the viscosity coefficient of unsteady shear creep, and A1, A2, m, and n are creep calculation parameters, which can be determined based on indoor creep tests; τ T It represents the ultimate strength of the soil in its flow state; t represents the creep time history.

[0076] S2-2: A continuous surface or zone in a landslide body with a rate of change of shear creep rate greater than 0 is defined as a sliding surface or sliding zone.

[0077] S2-3: The last segment that is connected in the sliding surface or sliding zone obtained in S2-2 is defined as the slope anti-slip segment;

[0078] S2-4: Determine the blocking segments in the potential sliding surface and their projected locations on the ground surface, as well as the potential main sliding line and the extent of possible slope instability and failure, through S1 and S2-3.

[0079] In S3, a nonlinear mapping relationship between the slope surface displacement or crack width at the projection location of the anti-slip segment on the ground surface in the potential sliding surface and the shear creep rate at the corresponding anti-slip segment is established through S1 and S2.

[0080] In S4, a machine vision deformer and target are installed in S2, which are equipped with non-contact intelligent sensing sensors and integrate machine vision and video surveillance technologies.

[0081] S4-1: A machine vision deformation meter equipped with non-contact intelligent sensing sensors and integrating machine vision and video monitoring technologies, installed at a distance of 10 to 500 meters from the monitored slope that can be clearly monitored.

[0082] S4-2: Install targets on the ground surface at the projection positions of the anti-slip sections in the potential sliding surface within a potential sliding surface range of 5-30m on the potential main sliding line of the slope determined by S2.

[0083] In S5, the displacement of the target or the crack width is obtained through S4;

[0084] S5-1: Real-time preliminary frame images of slope deformation are captured via S4.

[0085] S5-2: Optimize the Bayesian-convolutional neural network model based on S5-1 and perform mapping learning on the captured real-time preliminary frame images;

[0086] S5-3: Obtain the optimal feature point sequence through S5-2 to generate a higher resolution image;

[0087] S5-4: Using visual deformation monitoring technology in S5-3, calculate and obtain the displacement or crack width of the slope target and video data.

[0088] In S6, the displacement or crack width and video data obtained in S5 are automatically transmitted in real time to the computer of the remote client through the virtual IP technology of the Internet of Things.

[0089] S6-1: Set up a data acquisition station powered by solar energy at a distance of 1 to 10m from the machine vision deformer installed in S4-1. The data acquisition station is connected to the machine vision deformer via a fiber optic cable.

[0090] S6-2: Using mobile signals, the displacement or crack width of the slope target and video data acquired by S5 are automatically transmitted in real time to the computer of a remote client via the virtual IP technology of the Internet of Things.

[0091] In S7, the shear creep rate at the corresponding anti-slip section of the target is calculated using S2 and S6.

[0092] In S8, slope stability is dynamically evaluated through S7.

[0093] S8-1: If the rate of change of shear creep rate at the anti-slip section corresponding to the target obtained in S7 is less than 0, it is considered that the slope will tend to be stable.

[0094] S8-2: If the rate of change of shear creep rate at the anti-slip section corresponding to the target obtained in S7 is equal to 0, it is considered that the slope will tend to fail.

[0095] S8-3: If the rate of change of shear creep rate at the anti-slip section corresponding to the target obtained in S7 is greater than 0, it is considered that the slope is about to become unstable and fail.

[0096] A super-high embankment slope is supported by a reinforced soil retaining wall. The retaining wall is 150m long and 26m wide, with a height of 15m as an example. It is divided into two levels: the first level is 7m high, and the second level is 8m high. The step width between the upper and lower levels is 2m, the slope ratio is 1:0.5, and the walkway width is 2m. The geogrid reinforcement is 10m long with a vertical spacing of 1m. To monitor the stability of the slope and ensure its safety, the slope remote video monitoring stability analysis system and method of this invention are used. The physical and mechanical parameters of the slope soil are shown in Table 1.

[0097] Table 1 Physical and mechanical calculation parameters of slope soil

[0098]

[0099]

[0100] First, based on the calculation parameters shown in Table 1, a three-dimensional unsteady viscoelastic-plastic shear creep model is established. The rate of change of shear creep rate of the slope soil is calculated through finite element numerical simulation. Based on the rate of change of shear creep rate, the potential main slip line, slip surface or slip zone, and the slope resisting section are defined to conduct dynamic evaluation of the stability of the soil landslide.

[0101] Then, based on the slope potential main slip line and slip surface or slip zone defined by the rate of change of shear creep rate, as well as the slope anti-slip section, a machine vision deformable device equipped with non-contact intelligent sensing sensors and integrating machine vision and video monitoring technologies is installed and targets are deployed.

[0102] Finally, based on the target displacement or crack width obtained by a machine vision deformation analyzer equipped with non-contact intelligent sensing sensors and integrating machine vision and video monitoring technologies, the rate of change of shear creep rate at the corresponding anti-slip section is calculated. This more accurately reflects the actual dynamic changes in soil slope stability and is more conducive to timely dynamic evaluation of slope stability under rainfall conditions. Thus, the evaluation results are more accurate and effective, allowing the obtained data to more accurately reflect changes in slope stability.

[0103] The advantages of this invention are:

[0104] (1) This invention establishes a three-dimensional unsteady viscoelastic-plastic shear creep model, calculates the rate of change of shear creep rate of slope soil, defines the potential main sliding line and sliding surface or sliding zone of slope and the slope blocking section according to the rate of change of shear creep rate, and conducts dynamic evaluation of soil landslide stability, making the evaluation results of this invention more accurate and effective, and providing strong technical support for predicting the occurrence of landslide disasters.

[0105] (2) Based on the slope potential main slip line and slip surface or slip zone and slope anti-slip section defined by the shear creep rate change rate, the present invention installs a machine vision deformer equipped with non-contact intelligent sensing sensors and integrates machine vision and video monitoring technology and sets up targets, so that the obtained data can better reflect the changes in slope stability.

[0106] (3) Based on the target displacement or crack width obtained by the machine vision deformation instrument equipped with non-contact intelligent sensing sensor and integrating machine vision and video monitoring technology, the present invention calculates the rate of change of shear creep rate at the corresponding anti-slip section, which more accurately reflects the actual situation of dynamic changes in soil slope stability and is conducive to timely dynamic evaluation of slope stability under rainfall conditions.

[0107] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method for stability analysis of slopes via remote video monitoring, characterized in that, The method includes the following steps: S1: Construct a three-dimensional unsteady viscoelastic-plastic shear creep model to obtain slope stability data under rainfall. S2: Calculate the shear creep rate on the landslide body based on the slope stability data under rainfall, define the potential main sliding line and sliding part based on the shear creep rate, and define the last segment that is connected in the sliding part as the blocking segment. S3: Determine the nonlinear mapping relationship between the slope surface displacement or crack width at the projected location of the anti-slip segment in the sliding section and the shear creep rate at the corresponding anti-slip segment: S4: Sensors and targets are set based on the main slide rail; S5: Obtain the displacement of the target or the crack width; S6: Output the displacement or crack width to a remote computer; S7: The remote computer obtains the shear creep rate at the corresponding anti-slip section of the target based on the target displacement or crack width and nonlinear mapping relationship. S8: Determine slope stability based on the shear creep rate at the anti-slip section corresponding to the target, and obtain the slope stability analysis results; The specific steps of S2 are as follows: S2-1: Calculate the shear creep rate based on the lower slope stability data, and define the longitudinal line with the fastest change in shear creep rate on the landslide body as the potential main sliding line, which represents the main sliding direction of the overall slope sliding. S2-2: A continuous surface or zone in a landslide body with a rate of change of shear creep rate greater than 0 is defined as a sliding surface or sliding zone. S2-3: Define the last section that is connected in the sliding section as the slope anti-slip section; S2-4: Determine the blocking section in the sliding part and the projection position of the blocking section on the ground surface, the potential main sliding line, and the range of possible slope instability and failure; The shear creep rate is: Where, in the formula, This represents the shear creep rate. For the current stress, For soil yield stress, The viscosity coefficient for unsteady shear creep. , , and These are all parameters for creep calculation. It is the ultimate strength of the soil in its flow state; This represents the creep time history.

2. The method for remote video monitoring and stability analysis of slopes according to claim 1, characterized in that, The specific steps of S1 are as follows: S1-1: Construct a three-dimensional unsteady viscoelastic-plastic shear creep model; S1-2: Obtain the data required for the model through triaxial creep tests of unsaturated / saturated soil; S1-3: Based on the model and the parameters required by the model, a three-dimensional finite element numerical simulation of slope stability under rainfall is performed to obtain slope stability data under rainfall.

3. The method for remote video monitoring and stability analysis of slopes according to claim 2, characterized in that, The data required for the model include permeability coefficient, suction, cohesion, internal friction angle, long-term strength, elastic modulus, viscosity coefficient, and soil yield stress. Ultimate strength of soil in flow state And creep calculation parameters.

4. The method for remote video monitoring and stability analysis of slopes according to claim 1, characterized in that, The specific steps of S4 are as follows: S4-1: Install non-contact intelligent sensing sensors and machine vision deformation meters within a certain range of the slope; S4-2: Install targets on the ground surface at the projection location of the slope's anti-slip section within a certain range of the potential main sliding line and its vicinity.

5. The method for remote video monitoring and stability analysis of slopes according to claim 1, characterized in that, The specific steps of S5 are as follows: S5-1: Real-time preliminary frame images of slope deformation captured by non-contact intelligent sensing sensors; S5-2: Establish an optimization model for a Bayesian-convolutional neural network by inputting the initial frame images into the optimization model; S5-3: Optimize the model to obtain the best feature point sequence and output a high-resolution image; S5-4: Based on a machine vision deformer, acquire the displacement or crack width of the target and video data from high-resolution images.

6. The method for remote video monitoring and stability analysis of slopes according to claim 1, characterized in that, The specific steps of S6 are as follows: S6-1: A data acquisition station powered by solar energy is set up at a certain location of the machine vision deformer, and the data acquisition station is connected to the machine vision deformer via a fiber optic cable. S6-2: Transmit the displacement of the target or the crack width to a remote computer as a motion signal.

7. The method for remote video monitoring and stability analysis of slopes according to claim 1, characterized in that, The specific steps for S8 are as follows: S8-1: If the rate of change of shear creep rate at the anti-slip section corresponding to the target is less than 0, the slope is considered to be stabilizing. S8-2: If the rate of change of shear creep rate at the anti-slip section corresponding to the target is equal to 0, it is considered that the slope will tend to fail. S8-3: If the rate of change of shear creep rate at the anti-slip section corresponding to the target is greater than 0, it is considered that the slope is about to become unstable and fail.

8. A remote video monitoring and stability analysis system for slopes, characterized in that, The method includes a memory, a processor, and a program stored in the memory, characterized in that the processor executes the program to implement the method as described in any one of claims 1-6.

Images