Slope landslide geological disaster spot diagram intelligent identification method and system

Abstract

The invention relates to the field of image recognition, and discloses a slope landslide geological disaster spot diagram intelligent recognition method and system, and the method comprises the steps: carrying out the preprocessing of collected slope real-time data, carrying out the feature extraction of landslide-related feature parameters, generating a real-time spot diagram according to the feature parameters after feature extraction, constructing a spot diagram feature library, and carrying out the recognition of a slope landslide geological disaster spot diagram. Whether the current state of the slope is related to the landslide risk or not is judged by performing matching analysis on the extracted feature parameters and feature parameters in a spot diagram feature library, and the slope landslide risk is recognized by comparing and analyzing the difference between a real-time spot diagram and an original spot diagram obtained from a database. According to the method, the feature parameters of the slope landslide are acquired, then the slope landslide risk on the spot diagram is evaluated according to the feature parameters, the slope landslide risk area is subjected to risk grade division according to the risk evaluation result, and the risk division result is pre-warned, so that the slope landslide risk can be accurately predicted and quickly identified, and efficient risk pre-warning is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of image recognition, and more specifically to an intelligent recognition method and system for a spot image of a landslide geological disaster. Background Art

[0002] Slope landslide is a common geological disaster that poses a serious threat to people's lives and property. Therefore, it is crucial to accurately and timely identify and predict slope landslides. With the continuous development of science and technology, especially the rapid progress of Internet of Things technology, sensor technology, machine vision technology and big data analysis technology, the monitoring and early warning methods of slope landslide geological disasters are also constantly updated and improved. Therefore, the slope landslide geological disaster identification method came into being. Through data units, feature analysis units and identification and prediction units, the intelligent identification and prediction of slope landslide geological disasters can be realized, providing a scientific basis for disaster prevention and mitigation.

[0003] However, the above process still has the following disadvantages:

[0004] First, the existing methods for identifying landslide geological hazards lack real-time monitoring and the collection of multi-faceted data on the surrounding environment of the slope for feature analysis, which may lead to the prediction of landslide risks not being timely and accurate, and unable to achieve efficient risk warning;

[0005] Second, the existing method for identifying geological hazards of slope landslide lacks the ability to perform similarity analysis based on the currently extracted feature parameters and the generated spot map to determine whether the current state of the slope is related to the landslide risk, and then perform difference identification through the spot map, which leads to errors in the detection results of landslide risk and makes it impossible to judge the landslide risk more accurately. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a method and system for intelligent identification of spot diagrams of landslide geological disasters on slopes, so as to solve the problems existing in the above-mentioned background technology.

[0007] The present invention provides the following technical solution: a method for intelligently identifying a spot map of a landslide geological disaster, comprising:

[0008] S1: collects real-time slope data, including topographic and geological data, meteorological and environmental data, and human disturbance data, by using remote sensing monitoring methods, and enters the collected real-time slope data into the database, and transmits the collected real-time slope data to S2;

[0009] S2: used to pre-process the collected real-time slope data, and then extract the characteristic parameters related to the landslide from the pre-processed real-time slope data to obtain the slope displacement, slope, fault density, rainfall intensity, soil moisture and land development intensity, and transmit the characteristic parameters to S3;

[0010] S3: Generate a real-time spot map according to the feature parameters after feature extraction, collect spot map data under different slope conditions, and build a spot map feature library;

[0011] S4: by matching and analyzing the extracted characteristic parameters with the characteristic parameters in the spot map characteristic library, the characteristic similarity is obtained, and the characteristic similarity is used to determine whether the current state of the slope is related to the landslide risk. If it is determined that the current state of the slope is related to the landslide risk, the determination result is transmitted to S5;

[0012] S5: By comparing and analyzing the difference between the real-time spot map and the original spot map obtained from the database, a difference recognition coefficient is obtained, and the slope landslide risk is identified through the difference recognition coefficient, and the identification result is transmitted to S6;

[0013] S6: Perform risk assessment analysis on the spot map based on the extracted characteristic parameters, calculate the landslide risk score, conduct risk assessment on the landslide risk area of ​​the slope on the spot map through the landslide risk score, and transmit the risk assessment result to S7;

[0014] S7: classify the risk level of the slope landslide risk area according to the risk assessment result, and classify the slope landslide risk area into low risk, medium risk and high risk according to the landslide risk score, and transmit the risk classification result to S8;

[0015] S8: Issue risk level warning for the landslide risk area based on the risk classification results, and generate warning feedback prompt information and send it to the management terminal.

[0016] Preferably, the S1 collects real-time slope data by deploying meteorological sensors, including temperature sensors, humidity sensors, wind speed sensors and air pressure sensors, in the slope area or its periphery, and then combines with satellite-borne remote sensing sensors to monitor the surrounding environment of the slope in real time, and records and stores the collected real-time slope data in a database according to the collection time; the topographic geological data include the slope, aspect, height, soil type and rock type of the slope; the meteorological environment data include rainfall, rainfall intensity, rainfall duration, temperature, humidity, wind speed, solar radiation intensity and sunshine time; the human disturbance data include human flow, vegetation felling amount and land use change rate.

[0017] Preferably, S2 performs a preprocessing process on the real-time slope data including data cleaning, data conversion and data integration, and then extracts features from the preprocessed real-time slope data, and analyzes the slope displacement Δr, slope S, fault density D, rainfall intensity I, soil moisture P and land development intensity L.

[0018] Preferably, S3 sets multiple image attributes as a representation method for each characteristic parameter according to the slope stability, and then uses image processing technology to map the characteristic parameters into the image space, maps the displacement into the color depth of the spot, and maps the slope into the shape of the spot, to generate an initial spot map, and enhances the initial spot map, wherein the image attributes include color, shape, and size. Then, by combining the generated spot maps of different slopes together, a spot map feature library is constructed, and the generated real-time spot map is recorded and saved in a database in real time.

[0019] Preferably, the feature similarity calculation formula of S4 is: x i The i-th characteristic parameter representing the current slope state, y i represents the i-th characteristic parameter of the landslide risk state in the characteristic database, n represents the number of characteristic parameters, D max Represents the maximum value of the Euclidean distance among all feature parameter combinations;

[0020] By comparing the feature similarity F with the preset similarity threshold θ, it is determined whether the current state of the slope is related to the landslide risk. If the feature similarity F ≥ the preset similarity threshold θ, it means that the current state of the slope is similar to the landslide risk state, and the current state of the slope is determined to be related to the landslide risk, and the determination result is transmitted to the difference analysis module. If the feature similarity F < the preset similarity threshold θ, it means that the current state of the slope is not similar to the landslide risk state, and the current state of the slope is determined to be irrelevant to the landslide risk, and the slope condition continues to be monitored.

[0021] Preferably, the calculation formula of the difference recognition coefficient of S5 is: T j represents the jth characteristic parameter value in the real-time spot map, B j Represents the jth characteristic parameter value in the original spot map;

[0022] By comparing the difference recognition coefficient K with the difference threshold μ, if the difference recognition coefficient K>difference threshold μ, it is considered that the slope has a landslide risk, and the recognition result is transmitted to the landslide prediction module; if the difference recognition coefficient K≤difference threshold μ, it is considered that the slope is relatively stable.

[0023] Preferably, S6 calculates the landslide risk score of the identified slope landslide risk area based on the extracted characteristic parameters, and the calculated landslide risk score is R=α1×ln(Δr+1)+α2×ln(S+1)+α3×ln(D+1)+α4×ln(I+1)+α5×ln(P+1)+α6×ln(L+1)+1, Δr represents slope displacement, S represents slope, D represents fault density, I represents rainfall intensity, P represents soil moisture, L represents land development intensity, and α1, α2, α3, α4, α5, and α6 are weight coefficients.

[0024] Preferably, S7 sets a first risk score threshold η1 and a second risk score threshold η2, and η1<η2, compares the landslide risk score R with the first risk score threshold η1 and the second risk score threshold η2, and divides the slope landslide risk area into low risk, medium risk and high risk respectively; when the landslide risk score R<the first risk score threshold η1, it is determined that the slope landslide risk area is at low risk; when the first risk score threshold η1≤landslide risk score R<the second risk score threshold η2, it is determined that the slope landslide risk area is at medium risk; when the landslide risk score R≥the second risk score threshold η2, it is determined that the slope landslide risk area is at high risk.

[0025] Preferably, the S8 is used to receive the risk level information transmitted by the risk level classification module, generate corresponding warning information according to the risk level information, and when low-risk information is received, send blue warning information to prompt management personnel to pay attention to monitoring; when medium-risk information is received, send yellow warning information to prompt management personnel to conduct further investigation; when high-risk information is received, send red warning information and automatically alarm to remind management personnel to take landslide prevention measures immediately.

[0026] To achieve the above object, the present invention provides the following technical solution: a system for intelligent identification of spot diagrams of landslide geological disasters, which implements the above-mentioned method for intelligent identification of spot diagrams of landslide geological disasters, including:

[0027] Real-time data acquisition module: collects real-time slope data in real time by using remote sensing monitoring methods, including topographic and geological data, meteorological and environmental data, and human disturbance data, and enters the collected real-time slope data into the database, and transmits the collected real-time slope data to the real-time data feature extraction module;

[0028] Real-time data feature extraction module: used to pre-process the collected real-time slope data, and then extract the characteristic parameters related to the landslide from the pre-processed real-time slope data to obtain the slope displacement, slope, fault density, rainfall intensity, soil moisture and land development intensity, and transmit the characteristic parameters to the spot map generation module;

[0029] Spot map generation module: generates real-time spot maps based on feature parameters after feature extraction, collects spot map data under different slope conditions, and builds a spot map feature library;

[0030] Feature matching module: The feature parameters extracted are matched and analyzed with the feature parameters in the spot map feature library to obtain feature similarity. The feature similarity is used to determine whether the current state of the slope is related to the landslide risk. If the current state of the slope is determined to be related to the landslide risk, the determination result is transmitted to the difference analysis module.

[0031] Difference analysis module: by comparing and analyzing the difference between the real-time spot map and the original spot map obtained from the database, the difference recognition coefficient is obtained, and the landslide risk of the slope is identified through the difference recognition coefficient, and the identification result will be transmitted to the landslide risk assessment module;

[0032] Landslide risk assessment module: Based on the extracted characteristic parameters, the spot map is analyzed for risk assessment, and the landslide risk score is calculated. The landslide risk area on the slope on the spot map is assessed by the landslide risk score, and the risk assessment result is transmitted to the risk level classification module;

[0033] Risk level classification module: classify the risk level of the slope landslide risk area according to the risk assessment results, and classify the slope landslide risk area into low risk, medium risk and high risk according to the landslide risk score, and transmit the risk classification results to the early warning feedback module;

[0034] Early warning feedback module: According to the risk classification results, risk level early warning is issued for the slope landslide risk area, and early warning feedback prompt information is generated and sent to the management terminal.

[0035] Technical effects and advantages of the present invention:

[0036] The invention pre-processes the collected real-time data of the slope, and extracts characteristic parameters related to the landslide, generates a real-time spot map according to the characteristic parameters after the feature extraction, collects spot map data under different slope states, builds a spot map feature library, matches and analyzes the extracted characteristic parameters with the characteristic parameters in the spot map feature library to obtain characteristic similarity, thereby judging whether the current state of the slope is related to the landslide risk, identifying the slope landslide risk by comparing and analyzing the difference between the real-time spot map and the original spot map obtained from the database, and then performing risk assessment analysis on the slope landslide risk on the spot map according to the extracted characteristic parameters, dividing the risk level of the slope landslide risk area according to the risk assessment result, and providing early warning feedback for the risk division result, which is conducive to all-round real-time monitoring and analysis of the surrounding environment of the slope, thereby accurately predicting the slope landslide risk and realizing efficient risk early warning, and determining whether the current state of the slope is related to the landslide risk by performing similarity analysis between the currently extracted characteristic parameters and the generated spot map, making the landslide risk detection more accurate and reliable, and facilitating more accurate and rapid identification of the landslide risk. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a diagram of the method steps of the present invention.

[0038] Figure 2 It is a system structure block diagram of the present invention. DETAILED DESCRIPTION

[0039] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. In addition, the forms of the various structures recorded in the following embodiments are only examples. The method and system for intelligent identification of spot maps of slope landslide geological disasters involved in the present invention are not limited to the various structures recorded in the following embodiments. All other implementations obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0040] like Figure 1 The embodiment shown provides a method for intelligently identifying a spot map of a landslide geological disaster, including:

[0041] S1: Real-time slope data, including topographic and geological data, meteorological and environmental data, and human disturbance data, are collected by remote sensing monitoring methods, and the collected real-time slope data are entered into the database, and the collected real-time slope data are transmitted to S2.

[0042] In this embodiment, the S1 collects real-time slope data by deploying meteorological sensors, including temperature sensors, humidity sensors, wind speed sensors and air pressure sensors, in the slope area or its periphery, and then combines with satellite-borne remote sensing sensors to monitor the surrounding environment of the slope in real time, and records and stores the collected real-time slope data in a database according to the collection time; the topographic geological data includes the slope, slope direction, height, soil type and rock type of the slope; the meteorological environment data includes rainfall, rainfall intensity, rainfall duration, temperature, humidity, wind speed, solar radiation intensity and sunshine time; the human disturbance data includes human flow, vegetation felling amount and land use change rate.

[0043] S2: used to preprocess the collected real-time slope data, and then extract the characteristic parameters related to the landslide from the preprocessed real-time slope data to obtain the slope displacement, slope, fault density, rainfall intensity, soil moisture and land development intensity, and transmit the characteristic parameters to S3.

[0044] In this embodiment, S2 performs a preprocessing process on the real-time slope data, including data cleaning, data conversion and data integration, and then extracts features from the preprocessed real-time slope data, and analyzes the slope displacement Δr, slope S, fault density D, rainfall intensity I, soil moisture P and land development intensity L.

[0045] It should be noted that the slope displacement is calculated by using synthetic aperture radar interferometry to measure the slope displacement data. c represents the speed of light, Δφ represents the phase difference, λ represents the radar wavelength, and π represents a constant;

[0046] The slope calculation formula is: Δz represents the elevation difference, Δx represents the horizontal distance of the slope in the east-west direction, and Δy represents the horizontal distance of the slope in the north-south direction;

[0047] The calculation formula of fault density is: L v represents the length of the vth fault, A represents the area of ​​the studied fault, and V represents the total number of faults in the studied area;

[0048] The calculation formula for rainfall intensity is: p represents the current collected rainfall, and t represents the current rainfall time;

[0049] The calculation formula for soil moisture is: SWC represents the current soil volume water content. max Indicates the maximum soil water content, SWC min Indicates the minimum value of soil moisture content;

[0050] The calculation formula for land development intensity is: U 已 represents the area of ​​developed land, U 总 Represents the total area of ​​the land.

[0051] S3: Generate a real-time spot map based on the feature parameters after feature extraction, collect spot map data under different slope conditions, and build a spot map feature library.

[0052] In this embodiment, S3 sets multiple image attributes as a representation method for each characteristic parameter according to the slope stability, and then uses image processing technology to map the characteristic parameters to the image space, maps the displacement to the color depth of the spot, and maps the slope to the shape of the spot to generate an initial spot map, and enhances the initial spot map. The image attributes include color, shape, and size. Then, by combining the generated different slope spot maps together, a spot map feature library is constructed, and the generated real-time spot map is recorded and saved in a database in real time.

[0053] S4: The extracted feature parameters are matched and analyzed with the feature parameters in the spot map feature library to obtain feature similarity, and the feature similarity is used to determine whether the current state of the slope is related to the landslide risk. If it is determined that the current state of the slope is related to the landslide risk, the determination result is transmitted to S5.

[0054] In this embodiment, the feature similarity calculation formula of S4 is: x i The i-th characteristic parameter representing the current slope state, y i represents the i-th characteristic parameter of the landslide risk state in the characteristic database, n represents the number of characteristic parameters, D max Represents the maximum value of the Euclidean distance among all feature parameter combinations;

[0055] By comparing the feature similarity F with the preset similarity threshold θ, it is determined whether the current state of the slope is related to the landslide risk. If the feature similarity F ≥ the preset similarity threshold θ, it means that the current state of the slope is similar to the landslide risk state, and the current state of the slope is determined to be related to the landslide risk, and the determination result is transmitted to the difference analysis module. If the feature similarity F < the preset similarity threshold θ, it means that the current state of the slope is not similar to the landslide risk state, and the current state of the slope is determined to be irrelevant to the landslide risk, and the slope condition continues to be monitored.

[0056] S5: By comparing and analyzing the difference between the real-time spot map and the original spot map obtained from the database, a difference recognition coefficient is obtained, and the slope landslide risk is identified through the difference recognition coefficient, and the identification result is transmitted to S6.

[0057] In this embodiment, the calculation formula of the difference recognition coefficient of S5 is: T j represents the jth characteristic parameter value in the real-time spot map, B j Represents the jth characteristic parameter value in the original spot map;

[0058] By comparing the difference recognition coefficient K with the difference threshold μ, if the difference recognition coefficient K>difference threshold μ, it is considered that the slope has a landslide risk, and the recognition result is transmitted to the landslide prediction module; if the difference recognition coefficient K≤difference threshold μ, it is considered that the slope is relatively stable.

[0059] S6: Based on the extracted characteristic parameters, the spot map is subjected to risk assessment analysis, and a landslide risk score is calculated. The landslide risk area on the slope on the spot map is subjected to risk assessment through the landslide risk score, and the risk assessment result is transmitted to S7.

[0060] In this embodiment, S6 calculates the landslide risk score of the identified slope landslide risk area based on the extracted characteristic parameters, and the calculated landslide risk score is R=α1×ln(Δr+1)+α2×ln(S+1)+α3×ln(D+1)+α4×ln(I+1)+α5×ln(P+1)+α6×ln(L+1)+1, Δr represents slope displacement, S represents slope, D represents fault density, I represents rainfall intensity, P represents soil moisture, L represents land development intensity, and α1, α2, α3, α4, α5, and α6 are weight coefficients.

[0061] S7: Classify the risk level of the slope landslide risk area according to the risk assessment results, and divide the slope landslide risk area into low risk, medium risk and high risk according to the landslide risk score, and transmit the risk classification result to S8.

[0062] In this embodiment, S7 sets a first risk score threshold η1 and a second risk score threshold η2, and η1<η2, compares the landslide risk score R with the first risk score threshold η1 and the second risk score threshold η2, and divides the slope landslide risk area into low risk, medium risk and high risk respectively; when the landslide risk score R<the first risk score threshold η1, it is determined that the slope landslide risk area is at low risk; when the first risk score threshold η1≤landslide risk score R<the second risk score threshold η2, it is determined that the slope landslide risk area is at medium risk; when the landslide risk score R≥the second risk score threshold η2, it is determined that the slope landslide risk area is at high risk.

[0063] S8: Issue risk level warning for the landslide risk area based on the risk classification results, and generate warning feedback prompt information and send it to the management terminal.

[0064] In this embodiment, the S8 is used to receive the risk level information transmitted by the risk level classification module, and generate corresponding warning information according to the risk level information. When low-risk information is received, a blue warning message is sent to prompt management personnel to pay attention to monitoring. When medium-risk information is received, a yellow warning message is sent to prompt management personnel to conduct further investigation. When high-risk information is received, a red warning message is sent, and an automatic alarm is sounded to remind management personnel to take immediate landslide prevention measures.

[0065] like Figure 2 The embodiment shown provides an implementation system corresponding to the intelligent identification method of spot map of slope landslide geological disasters, including a real-time data acquisition module, a real-time data feature extraction module, a spot map generation module, a feature matching module, a difference analysis module, a landslide risk assessment module, a risk level classification module and an early warning feedback module, the real-time data acquisition module is connected to the real-time data feature extraction module, the real-time data feature extraction module is connected to the spot map generation module, the spot map generation module is connected to the feature matching module, the feature matching module is connected to the difference analysis module, the difference analysis module is connected to the landslide risk assessment module, the real-time data feature extraction module is connected to the landslide risk assessment module, the landslide risk assessment module is connected to the risk level classification module, and the risk level classification module is connected to the early warning feedback module.

[0066] The real-time data acquisition module collects real-time slope data in real time by using remote sensing monitoring methods, including topographic and geological data, meteorological and environmental data, and human disturbance data, and enters the collected real-time slope data into a database, and transmits the collected real-time slope data to the real-time data feature extraction module;

[0067] The real-time data feature extraction module is used to pre-process the collected real-time slope data, and then extract the characteristic parameters related to the landslide from the pre-processed real-time slope data to obtain the slope displacement, slope, fault density, rainfall intensity, soil moisture and land development intensity, and transmit the characteristic parameters to the spot map generation module;

[0068] The spot map generation module generates a real-time spot map according to the feature parameters after feature extraction, collects spot map data under different slope conditions, and constructs a spot map feature library;

[0069] The feature matching module matches and analyzes the extracted feature parameters with the feature parameters in the spot map feature library to obtain feature similarity, and determines whether the current state of the slope is related to the landslide risk through the feature similarity. If it is determined that the current state of the slope is related to the landslide risk, the determination result is transmitted to the difference analysis module;

[0070] The difference analysis module obtains a difference recognition coefficient by comparing and analyzing the difference between the real-time spot map and the original spot map obtained from the database, identifies the slope landslide risk through the difference recognition coefficient, and transmits the identification result to the landslide risk assessment module;

[0071] The landslide risk assessment module performs risk assessment analysis on the spot map based on the extracted characteristic parameters, calculates the landslide risk score, performs risk assessment on the slope landslide risk area on the spot map according to the landslide risk score, and transmits the risk assessment result to the risk level classification module;

[0072] The risk level classification module classifies the risk level of the slope landslide risk area according to the risk assessment result, and classifies the slope landslide risk area into low risk, medium risk and high risk according to the landslide risk score, and transmits the risk classification result to the early warning feedback module;

[0073] The early warning feedback module issues a risk level early warning to the slope landslide risk area according to the risk classification result, and generates early warning feedback prompt information and sends it to the management personnel terminal.

[0074] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0075] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for intelligent identification of spot diagrams of landslide geological hazards, characterized in that: include: S1: collects real-time slope data, including topographic and geological data, meteorological and environmental data, and human disturbance data, by using remote sensing monitoring methods, and enters the collected real-time slope data into the database, and transmits the collected real-time slope data to S2; S2: used to pre-process the collected real-time slope data, and then extract the characteristic parameters related to the landslide from the pre-processed real-time slope data to obtain the slope displacement, slope, fault density, rainfall intensity, soil moisture and land development intensity, and transmit the characteristic parameters to S3; S3: Generate a real-time spot map according to the feature parameters after feature extraction, collect spot map data under different slope conditions, and build a spot map feature library; S4: by matching and analyzing the extracted characteristic parameters with the characteristic parameters in the spot map characteristic library, the characteristic similarity is obtained, and the characteristic similarity is used to determine whether the current state of the slope is related to the landslide risk. If it is determined that the current state of the slope is related to the landslide risk, the determination result is transmitted to S5; S5: By comparing and analyzing the difference between the real-time spot map and the original spot map obtained from the database, a difference recognition coefficient is obtained, and the slope landslide risk is identified through the difference recognition coefficient, and the identification result is transmitted to S6; S6: Perform risk assessment analysis on the spot map based on the extracted characteristic parameters, calculate the landslide risk score, conduct risk assessment on the landslide risk area of ​​the slope on the spot map through the landslide risk score, and transmit the risk assessment result to S7; S7: classify the risk level of the slope landslide risk area according to the risk assessment result, and classify the slope landslide risk area into low risk, medium risk and high risk according to the landslide risk score, and transmit the risk classification result to S8; S8: Issue risk level warning for the landslide risk area based on the risk classification results, and generate warning feedback prompt information and send it to the management terminal.

2. The method for intelligent identification of spot diagrams of landslide geological hazards according to claim 1 is characterized in that: The S1 collects real-time slope data by deploying meteorological sensors, including temperature sensors, humidity sensors, wind speed sensors and air pressure sensors, in the slope area or its periphery, and then combines with satellite-borne remote sensing sensors to monitor the surrounding environment of the slope in real time, and records and stores the collected real-time slope data in a database according to the collection time; the topographic geological data include the slope, slope direction, height, soil type and rock type of the slope; the meteorological environment data include rainfall, rainfall intensity, rainfall duration, temperature, humidity, wind speed, solar radiation intensity and sunshine time; the human disturbance data include human flow, vegetation felling amount and land use change rate.

3. The method for intelligent identification of spot diagrams of landslide geological hazards according to claim 1 is characterized in that: The S2 performs a preprocessing process on the real-time slope data, including data cleaning, data conversion and data integration, and then extracts features from the preprocessed real-time slope data, and analyzes the slope displacement Δr, slope S, fault density D, rainfall intensity I, soil moisture P and land development intensity L.

4. The method for intelligent identification of spot diagrams of landslide geological hazards according to claim 1 is characterized in that: The S3 sets a plurality of image attributes as a representation method for each characteristic parameter according to the slope stability, and then uses image processing technology to map the characteristic parameters to the image space, maps the displacement to the color depth of the spot, and maps the slope to the shape of the spot, generates an initial spot map, and enhances the initial spot map. The image attributes include color, shape, and size. Then, by combining the generated different slope spot maps together, a spot map feature library is constructed, and the generated real-time spot map is recorded and saved in a database in real time.

5. The method for intelligent identification of spot diagrams of landslide geological disasters according to claim 1 is characterized in that: The feature similarity calculation formula of S4 is: x i The i-th characteristic parameter representing the current slope state, y i represents the i-th characteristic parameter of the landslide risk state in the characteristic database, n represents the number of characteristic parameters, D max Represents the maximum value of the Euclidean distance among all feature parameter combinations; By comparing the feature similarity F with the preset similarity threshold θ, it is determined whether the current state of the slope is related to the landslide risk. If the feature similarity F ≥ the preset similarity threshold θ, it means that the current state of the slope is similar to the landslide risk state, and the current state of the slope is determined to be related to the landslide risk, and the determination result is transmitted to the difference analysis module. If the feature similarity F < the preset similarity threshold θ, it means that the current state of the slope is not similar to the landslide risk state, and the current state of the slope is determined to be irrelevant to the landslide risk, and the slope condition continues to be monitored.

6. The method for intelligent identification of spot diagrams of landslide geological disasters according to claim 1 is characterized in that: The calculation formula of the difference recognition coefficient of S5 is: T j represents the jth characteristic parameter value in the real-time spot map, B j Represents the jth characteristic parameter value in the original spot map; By comparing the difference recognition coefficient K with the difference threshold μ, if the difference recognition coefficient K>difference threshold μ, it is considered that the slope has a landslide risk, and the recognition result is transmitted to the landslide prediction module; if the difference recognition coefficient K≤difference threshold μ, it is considered that the slope is relatively stable.

7. The method for intelligent identification of spot diagrams of landslide geological disasters according to claim 1 is characterized in that: The S6 calculates the landslide risk score of the identified slope landslide risk area based on the extracted characteristic parameters, and the calculated landslide risk score is R=α1×ln(Δr+1)+α2×ln(S+1)+α3×ln(D+1)+α4×ln(I+1)+α5×ln(P+1)+α6×ln(L+1)+1, Δr represents slope displacement, S represents slope, D represents fault density, I represents rainfall intensity, P represents soil moisture, L represents land development intensity, and α1, α2, α3, α4, α5, and α6 are weight coefficients.

8. The method for intelligent identification of spot diagrams of landslide geological hazards according to claim 1 is characterized in that: S7 sets a first risk score threshold η1 and a second risk score threshold η2, and η1<η2, compares the landslide risk score R with the first risk score threshold η1 and the second risk score threshold η2, and divides the slope landslide risk area into low risk, medium risk and high risk respectively; When the landslide risk score R is less than the first risk score threshold η1, the slope landslide risk area is judged to be at low risk; when the first risk score threshold η1≤landslide risk score R<second risk score threshold η2, the slope landslide risk area is judged to be at medium risk; when the landslide risk score R≥second risk score threshold η2, the slope landslide risk area is judged to be at high risk.

9. The method for intelligent identification of spot diagrams of landslide geological disasters according to claim 1 is characterized in that: The S8 is used to receive the risk level information transmitted by the risk level classification module, and generate corresponding warning information according to the risk level information. When low-risk information is received, a blue warning message is sent to prompt management personnel to pay attention to monitoring. When medium-risk information is received, a yellow warning message is sent to prompt management personnel to conduct further investigation. When high-risk information is received, a red warning message is sent and an automatic alarm is sounded to remind management personnel to take landslide prevention measures immediately.

10. A system for intelligent identification of spot patterns of landslide geological hazards, implementing a method for intelligent identification of spot patterns of landslide geological hazards as claimed in any one of claims 1 to 9, characterized in that: include: Real-time data acquisition module: collects real-time slope data in real time by using remote sensing monitoring methods, including topographic and geological data, meteorological and environmental data, and human disturbance data, and enters the collected real-time slope data into the database, and transmits the collected real-time slope data to the real-time data feature extraction module; Real-time data feature extraction module: used to pre-process the collected real-time slope data, and then extract the characteristic parameters related to the landslide from the pre-processed real-time slope data to obtain the slope displacement, slope, fault density, rainfall intensity, soil moisture and land development intensity, and transmit the characteristic parameters to the spot map generation module; Spot map generation module: generates real-time spot maps based on feature parameters after feature extraction, collects spot map data under different slope conditions, and builds a spot map feature library; Feature matching module: The feature parameters extracted are matched and analyzed with the feature parameters in the spot map feature library to obtain feature similarity. The feature similarity is used to determine whether the current state of the slope is related to the landslide risk. If the current state of the slope is determined to be related to the landslide risk, the determination result is transmitted to the difference analysis module. Difference analysis module: by comparing and analyzing the difference between the real-time spot map and the original spot map obtained from the database, the difference recognition coefficient is obtained, and the landslide risk of the slope is identified through the difference recognition coefficient, and the identification result will be transmitted to the landslide risk assessment module; Landslide risk assessment module: Based on the extracted characteristic parameters, the spot map is analyzed for risk assessment, and the landslide risk score is calculated. The landslide risk area on the slope on the spot map is assessed by the landslide risk score, and the risk assessment result is transmitted to the risk level classification module; Risk level classification module: classify the risk level of the slope landslide risk area according to the risk assessment results, and classify the slope landslide risk area into low risk, medium risk and high risk according to the landslide risk score, and transmit the risk classification results to the early warning feedback module; Early warning feedback module: According to the risk classification results, risk level early warning is issued for the slope landslide risk area, and early warning feedback prompt information is generated and sent to the management terminal.

Images