Foundation deformation monitoring radar site selection method, medium and equipment

Abstract

The invention relates to the technical field of wireless positioning, in particular to a foundation deformation monitoring radar site selection method, which comprises the following steps: shooting to obtain an image of a to-be-monitored area, and obtaining a three-dimensional point cloud model of the to-be-monitored area; preprocessing the image, extracting and classifying ground feature features, training a deep learning model to identify different ground feature types, and assigning the different ground feature types to corresponding point cloud attributes; selecting a radar layout area according to the geological environment data, and dividing the radar layout area into a plurality of grids; selecting a grid at the center of the radar layout area as a radar monitoring point; constructing a fitness function according to intervisibility from a radar monitoring point to each point cloud, signal intensity data of radar scanning and point cloud attributes; and searching to obtain a grid with the maximum fitness function value, namely an output radar site selection result. According to the invention, the intervisibility and the radar signal intensity are comprehensively considered, the global optimal radar arrangement position is found by using the particle swarm optimization algorithm, and the quality and reliability of monitoring data are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless positioning technology, and in particular to a site selection method for a ground deformation monitoring radar. Background Art

[0002] With the continuous development of remote sensing technology and radar technology, arc synthetic aperture ground deformation monitoring radar has been widely used in geological monitoring, disaster warning and other fields. The radar transmits and receives electromagnetic waves to perform high-resolution imaging of the surface, thereby realizing real-time monitoring of surface deformation. However, the monitoring effect of the radar depends to a large extent on its deployment location. An unreasonable deployment location may lead to problems such as monitoring blind spots and signal interference, which seriously affect the accuracy and reliability of the monitoring results. Traditional radar deployment mainly relies on manual experience for site selection, and usually requires manual selection of the radar location based on factors such as terrain features and line of sight conditions. This method is not only time-consuming and labor-intensive, but also prone to problems such as incomplete consideration and misjudgment, resulting in unsatisfactory monitoring results. Summary of the invention

[0003] The present invention aims to provide a method for selecting a site for a ground deformation monitoring radar for efficiently selecting a radar deployment location. The specific technical solution is as follows:

[0004] The present invention provides a method for selecting a site for a ground deformation monitoring radar, comprising the following steps:

[0005] S1: Use UAV aerial photogrammetry technology to capture images of the area to be monitored and obtain a three-dimensional point cloud model of the area to be monitored; preprocess the image to obtain a preprocessed image;

[0006] S2: Use deep learning algorithms to extract and classify the features of objects in the preprocessed images, train deep learning models to identify different object types, and assign object types to corresponding point cloud attributes;

[0007] S3: Select a radar deployment area based on geological environment data, and divide the radar deployment area into multiple grids; select the grid at the center of the radar deployment area as the radar monitoring point;

[0008] S4: Construct the fitness function of the radar deployment location based on the visibility from the radar monitoring point to each point cloud, the signal strength data of the radar scan, and the point cloud attributes;

[0009] S5: Based on the fitness function, the particle swarm optimization algorithm is used to search for the grid with the largest fitness function value, which is the output radar site selection result.

[0010] Optionally, in S2, the point cloud attribute corresponding to the ground object type with strong electromagnetic wave reflection ability is assigned to 1, and the others are assigned to 0;

[0011] The S4 includes:

[0012] S4.1. Thinning the three-dimensional point cloud model of the monitored area to obtain a thinned point cloud model, and calculating the number of point clouds in the thinned point cloud model;

[0013] S4.2. Deploy the radar at the radar monitoring point, calculate the visibility of the radar to each point cloud, and count the percentage of the unobstructed point cloud to the total number of point clouds. pt , where: the total number of point clouds does not include the occluded point clouds whose point cloud attributes are 0;

[0014] S4.3. Use radar to scan the monitored area to obtain a radar monitoring image consisting of multiple cells, and obtain the signal strength data of each cell. The statistical signal strength is greater than the set threshold. The percentage of cells in the total number of cells P st ;

[0015] S4.4. Construct the fitness function of the radar deployment location. The specific expression is as follows:

[0016] Score=P pt ×σ+(1-σ)×P st ;

[0017] Where: Score is the status score of the current radar monitoring point, that is, the fitness function of the radar deployment position; σ is the weight. Optionally, the specific calculation formula for the number of point clouds N of the thinned point cloud model in S4.1 is as follows:

[0018] N=S m / P all ×(Δr×Δr)×P tn ;

[0019] Where: S m is the area of ​​the area to be monitored, P all is the number of point clouds in the 3D point cloud model of the area to be monitored, and Δr is the length of the grid edge.

[0020] Optionally, the S4.1 includes:

[0021] Divide the monitoring area to be tested into key monitoring areas and non-key monitoring areas based on geological environment data;

[0022] Calculate the number of thinned point clouds in the cells in the key monitoring area and the non-key monitoring area respectively, where: the P corresponding to the cells in the key monitoring area tn The P value is greater than the P value of the cell in the non-key monitoring area. tn Get value;

[0023] The number of point clouds in the thinned point cloud model is obtained by adding the number of thinned point clouds in the cells in the key monitoring area and the non-key monitoring area.

[0024] Optionally, the S5 includes:

[0025] S5.1. Generate a uniformly distributed particle group near the radar monitoring point, with one particle corresponding to one grid;

[0026] S5.2, particle swarm initialization: Initialize the particle positions in the particle swarm, that is, randomly select a grid position as the initial deployment point of the radar, and randomly initialize the particle velocity, where: the position of each particle is a two-dimensional coordinate, that is, the coordinate of the particle corresponding to the grid in the radar deployment area;

[0027] S5.3. Calculate the fitness function value of each particle in the particle swarm in t iterations, obtain the historical optimal position pBesti(t) and the global optimal position gBest(t) in the t iteration, and the particle position corresponding to the maximum fitness function value is the historical optimal position pBest(t) of the particle in t iterations, where: t is the number of iterations, which is a natural number greater than 0; if pBesti(t)≥pBesti(t-1) and pBesti(t)≥gBest(t-1), then the global optimal position gBest(t)=pBesti(t) in t iterations, otherwise gBest(t)=gBest(t-1); when t=1, gBest(1)=pBesti(1);

[0028] S5.4. If t reaches the maximum iteration number threshold, or the fitness function value of the global optimal position changes less than the preset threshold compared with the t-1th time, the iteration stops, and the output global optimal position gBest(t) is the final site selection result of the radar; otherwise, let t = t + 1, and call the update formula to update the particle swarm position and speed, and return to S5.3.

[0029] Optionally, in S4.3, In S4.4, 0.3≤σ≤0.5.

[0030] Optionally, the features of the land objects include texture features, shape and structure features, and spatial relationship features; the types of land objects include rocks, trees, water flows, soil surfaces, grass, and other land objects.

[0031] The present invention also provides a readable storage medium having computer program instructions stored thereon, and when the computer program instructions are executed by a processor, the above-mentioned method for selecting a site for a ground deformation monitoring radar is implemented.

[0032] The present invention also provides an electronic device, comprising: at least one processor, at least one memory and computer program instructions stored in the memory, when the computer program instructions are executed by the processor, the above-mentioned method for selecting a site for a ground deformation monitoring radar is implemented.

[0033] By applying the technical solution of the present invention, by comprehensively considering multiple factors such as terrain visibility and radar signal strength, the particle swarm optimization algorithm can be used to find the global optimal radar deployment location, thereby improving the quality and reliability of monitoring data. The particle swarm optimization algorithm has efficient search capabilities and can find the optimal solution in a relatively short period of time, greatly improving the efficiency of site selection compared to traditional manual point selection methods. The present invention uses drone photogrammetry technology and deep learning algorithms to further improve the accuracy and scientificity of radar site selection. At the same time, it can be adjusted and optimized according to different monitoring needs and scenarios, and has strong flexibility and scalability.

[0034] In addition to the above-described purposes, features and advantages, the present invention has other purposes, features and advantages. The present invention will be further described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The drawings constituting a part of this application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0036] Figure 1 is a three-dimensional point cloud model of the area to be monitored obtained by S1 in the embodiment of the present invention;

[0037] Figure 2 is a schematic diagram of grid division of the radar deployment area in S3 in an embodiment of the present invention;

[0038] Figure 3 is a three-dimensional point cloud model of the key monitoring area in an embodiment of the present invention;

[0039] Figure 4 is a three-dimensional point cloud model of a non-key monitoring area in an embodiment of the present invention;

[0040] Figure 5 is a schematic diagram of a point cloud before thinning of the key monitoring area in an embodiment of the present invention;

[0041] Figure 6 Schematic diagram of the point cloud after thinning of the key monitoring area in the embodiment of the present invention. DETAILED DESCRIPTION

[0042] The embodiments of the present invention are described in detail below with reference to the accompanying drawings, but the present invention can be implemented in many different ways as defined and covered by the claims.

[0043] In one embodiment, a method for selecting a site for a ground deformation monitoring radar includes the following steps:

[0044] S1: Use UAV aerial photogrammetry technology to capture images of the area to be monitored and obtain a three-dimensional point cloud model of the area to be monitored; preprocess the image to obtain a preprocessed image; see Figure 1 .

[0045] S2: Use deep learning algorithms to extract and classify the features of objects in the preprocessed images, train deep learning models to identify different object types, and assign object types to corresponding point cloud attributes;

[0046] Land feature characteristics include texture features, shape and structure features, and spatial relationship features; land feature types include rocks, trees, water flows, soil surfaces, grass, and other land features.

[0047] For example, with respect to texture features, rock and soil surfaces may exhibit different texture patterns, which can be captured and learned through multi-layer convolutions of CNN; with respect to shape and structural features, trees and buildings are significantly different in shape, and these features can be extracted through the convolutional layers of deep learning models; with respect to spatial relationship features, water flows are usually located in valleys, while trees may be distributed along rivers, and these spatial relationships can be captured through contextual information in deep learning models.

[0048] Convolutional Neural Network (CNN): CNN is a powerful tool for image recognition and classification in deep learning. In remote sensing image analysis, CNN can automatically learn features extracted from images and use them for classification tasks. CNN extracts the above-mentioned ground features through multiple layers of convolution and pooling operations, and classifies them through fully connected layers.

[0049] Among them, preprocessing is a processing method that comes with existing software, generally including:

[0050] Distortion correction: Perform geometric distortion correction on the image to ensure the geometric accuracy of the image;

[0051] Radiation correction: including radiation correction, reflectivity correction, bidirectional reflectance distribution function and shadow correction of the optical measurement system to ensure the accuracy of the radiation characteristics of the image data;

[0052] Geometric correction: Perform geometric distortion correction on the original image and geometric registration of multispectral images to ensure that images from different perspectives can be accurately aligned.

[0053] S3: Select the radar deployment area according to the geological environment data, and divide the radar deployment area into multiple grids; select the grid at the center of the radar deployment area as the radar monitoring point; see Figure 2.

[0054] The radar deployment area is preliminarily selected based on the visibility between the radar and the monitored area, the stability of the bedrock, whether there are high-voltage lines nearby, the convenience of power supply and other conditions obtained from the existing geological environment data.

[0055] S4: Construct the fitness function of the radar deployment location based on the visibility from the radar monitoring point to each point cloud, the signal strength data of the radar scan, and the point cloud attributes;

[0056] S4 includes:

[0057] S4.1. Thinning the three-dimensional point cloud model of the monitored area to obtain a thinned point cloud model, and calculating the number of point clouds in the thinned point cloud model;

[0058] The specific calculation formula for the number of point clouds N of the thinned point cloud model in S4.1 is as follows:

[0059] N=S m / P all ×(Δr×Δr)×P tn ;

[0060] Where: S m is the area of ​​the area to be monitored, P all is the number of point clouds in the 3D point cloud model of the area to be monitored, and Δr is the length of the grid edge.

[0061] In S2, the point cloud attributes corresponding to the ground object types with strong electromagnetic wave reflection ability are assigned 1, and the others are 0.

[0062] S4.1 includes:

[0063] According to the geological environment data, the monitoring area to be tested is divided into key monitoring areas and non-key monitoring areas; see Figure 3 and Figure 4 .

[0064] Calculate the number of thinned point clouds in the cells in the key monitoring area and the non-key monitoring area respectively, where: the P corresponding to the cells in the key monitoring area tn The P value is greater than the P value of the cell in the non-key monitoring area. tn Get value;

[0065] The number of point clouds in the thinned point cloud model is obtained by adding the number of thinned point clouds in the cells in the key monitoring area and the non-key monitoring area. Figure 5 and Figure 6 .

[0066] S4.2. Deploy the radar at the radar monitoring point, calculate the visibility of the radar to each point cloud, and count the percentage of the unobstructed point cloud to the total number of point clouds. pt, where: the total number of point clouds does not include the occluded point clouds whose point cloud attributes are 0;

[0067] When calculating the intervisibility of the radar to each point cloud, ray tracing, elevation diagram 3 method, point cloud common view estimation and other methods can be used for calculation. The specific process of calculation using ray tracing method is as follows:

[0068] ①、Define the ray KP starting from the radar monitoring point K i , where: P i is the i-th point cloud in the thinned point cloud model, with coordinates (x i ,y i ,z i );

[0069] ②For each point P i , calculate from K to P i Is there any point in the terrain point cloud blocking the line of sight on the line path? This can be done by checking whether all other points lie on this line (ray).

[0070] ③. Use geometric operations such as dot product or cross product to determine whether a point is on a line. If the dot product is greater than 0, it means the point is on one side of the line; if it is equal to 0, it is on the line; if it is less than 0, it is on the other side of the line.

[0071] If there is no point between K and P i If the straight line between point K and point P i There is visibility between them.

[0072] Conversely, if any point is on the straight line, then point K is considered to be i There is no line of sight between them.

[0073] For a point K(x k ,y k ,z k ) and a point P in the point cloud i (x i ,y i ,z i ), calculate the ray direction vector as:

[0074] Direction=(x i -x k ,y i -y k ,z i -z k )Direction=(x i -x k ,y i -y k ,z i-z k );

[0075] S4.2.5. For any two points A(x a ,y a ,z a ) and B(x b ,y b ,z b ), determine the third point C(x c, y c ,z c ) is on the AB line, the formula can be simplified by calculating the modulus of the cross product:

[0076] AB×AC=(y a -y b )(z c -z c )-(z a -z b )(y c -x c ),(z a -z b )(x b -x a )-(x a -x b )(z c -z c ),(x a -x b )(y c -y a )-(y a -y b )(x c

[0077] -x a ))AB×AC=(y a -y b )(z c -z c )-(z a -z b )(y c -x c ),(z a -z b )(x b -x a )-(x a -x b )(z c -z c ),(x a -x b )(y c -y a )-(ya -y b )(x c -x a ));

[0078] If the modulus of the above cross product result is 0, then point C is on straight line AB.

[0079] S4.3. Use radar to scan the monitored area to obtain a radar monitoring image consisting of multiple cells, and obtain the signal strength data of each cell. The statistical signal strength is greater than the set threshold. The percentage of cells in the total number of cells P st ;in:

[0080] The scene illuminated by the radar covers a certain terrain area, which is presented in the form of a fan-shaped grid. Each monitoring unit is called a fan-shaped cell. There are m×n monitoring units in the entire monitoring range, where m represents the number of units in the distance direction and n represents the number of units in the angle direction. The specific values ​​of m and n depend on the angular resolution, distance resolution and total monitoring range.

[0081] The results of radar monitoring mainly include two parts of data: deformation data and signal strength data. The deformation data reflects the cumulative deformation value of each monitoring unit relative to the initial monitoring time. Each frame of radar monitoring data contains m×n cumulative deformation monitoring values. The signal strength data indicates the signal strength reflected by the radar, with a value range of -50.0dB to 0.0dB. Each frame of radar monitoring data also contains m×n signal strength values.

[0082] S4.4. Construct the fitness function of the radar deployment location. The specific expression is as follows:

[0083] Score=P pt ×σ+(1-σ)×P st ;

[0084] Where: Score is the status score of the current radar monitoring point, that is, the fitness function of the radar deployment location; σ is the weight, 0.3≤σ≤0.5.

[0085] S5: Based on the fitness function, the particle swarm optimization algorithm is used to search for the grid with the largest fitness function value, which is the output radar site selection result.

[0086] S5 includes:

[0087] S5.1. Generate a uniformly distributed particle group near the radar monitoring point, with one particle corresponding to one grid;

[0088] S5.2, particle swarm initialization: Initialize the particle positions in the particle swarm, that is, randomly select a grid position as the initial deployment point of the radar, and randomly initialize the particle velocity, where: the position of each particle is a two-dimensional coordinate, that is, the coordinate of the particle corresponding to the grid in the radar deployment area;

[0089] S5.3. Calculate the fitness function value of each particle in the particle swarm in t iterations, obtain the historical optimal position pBesti(t) and the global optimal position gBest(t) in the t iteration, and the particle position corresponding to the maximum fitness function value is the historical optimal position pBest(t) of the particle in t iterations, where: t is the number of iterations, which is a natural number greater than 0; if pBesti(t)≥pBesti(t-1) and pBesti(t)≥gBest(t-1), then the global optimal position gBest(t)=pBesti(t) in t iterations, otherwise gBest(t)=gBest(t-1); when t=1, gBest(1)=pBesti(1);

[0090] S5.4. If t reaches the maximum iteration number threshold, or the fitness function value of the global optimal position changes less than the preset threshold compared with the t-1th time, the iteration stops, and the output global optimal position gBest(t) is the final site selection result of the radar; otherwise, let t = t + 1, and call the update formula to update the particle swarm position and speed, and return to S5.3.

[0091] In this embodiment, the cells of the monitored area are 10×10 cells, and the radar is initially deployed at the position (5,5). Through the particle swarm optimization algorithm, the radar gradually adjusts the deployment position, and each particle in the particle swarm corresponds to a potential radar deployment point. After 50 iterations, the radar is finally deployed at the position (8,7), so that the visibility percentage P pt Reaching 90%, the proportion of cells with signal intensity greater than the threshold P st Reaching 85%, the final state score was 0.875.

[0092] By comprehensively considering multiple factors such as terrain visibility and radar signal strength, the particle swarm optimization algorithm can be used to find the global optimal radar deployment location and improve the quality and reliability of monitoring data. The particle swarm optimization algorithm has efficient search capabilities and can find the optimal solution in a relatively short time, which greatly improves the efficiency of site selection compared to traditional manual point selection methods. The present invention uses drone photogrammetry technology and deep learning algorithms to further improve the accuracy and scientificity of radar site selection. At the same time, it can be adjusted and optimized according to different monitoring needs and scenarios, and has strong flexibility and scalability.

[0093] This embodiment also includes a readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the above-mentioned method for selecting a site for a ground deformation monitoring radar is implemented.

[0094] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0095] This embodiment also includes an electronic device, including: at least one processor, at least one memory, and computer program instructions stored in the memory. When the computer program instructions are executed by the processor, the mining slope stability analysis method as described above is performed.

[0096] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the electronic device.

[0097] The electronic device may be a computing device such as a mobile phone, a desktop computer, a notebook, a PDA, a cloud server, etc. The electronic device may include, but is not limited to, a processor and a memory. For example, the electronic device may also include an input / output device, a network access device, a bus, etc.

[0098] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, and uses various interfaces and lines to connect various parts of the entire electronic device.

[0099] The memory can be used to store the computer program and / or module, and the processor implements the computer program by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0100] Wherein, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0101] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for selecting a site for a ground deformation monitoring radar, characterized in that: The steps include: S1: Use UAV aerial photogrammetry technology to capture images of the area to be monitored and obtain a three-dimensional point cloud model of the area to be monitored; preprocess the image to obtain a preprocessed image; S2: Use deep learning algorithms to extract and classify the features of objects in the preprocessed images, train deep learning models to identify different object types, and assign object types to corresponding point cloud attributes; S3: Select a radar deployment area based on geological environment data, and divide the radar deployment area into multiple grids; select the grid at the center of the radar deployment area as the radar monitoring point; S4: Construct the fitness function of the radar deployment location based on the visibility from the radar monitoring point to each point cloud, the signal strength data of the radar scan, and the point cloud attributes; S5: Based on the fitness function, the particle swarm optimization algorithm is used to search for the grid with the largest fitness function value, which is the output radar site selection result.

2. The method for selecting a site for ground deformation monitoring radar according to claim 1, characterized in that: In S2, the point cloud attributes corresponding to the ground object type with strong electromagnetic wave reflection ability are assigned 1, and the others are assigned 0; The S4 includes: S4.

1. Thinning the three-dimensional point cloud model of the monitored area to obtain a thinned point cloud model, and calculating the number of point clouds in the thinned point cloud model; S4.

2. Deploy the radar at the radar monitoring point, calculate the visibility of the radar to each point cloud, and count the percentage of the unobstructed point cloud to the total number of point clouds. pt , where: the total number of point clouds does not include the occluded point clouds whose point cloud attributes are 0; S4.

3. Use radar to scan the monitored area to obtain a radar monitoring image consisting of multiple cells, and obtain the signal strength data of each cell. The statistical signal strength is greater than the set threshold. The percentage of cells in the total number of cells P st ; S4.

4. Construct the fitness function of the radar deployment location. The specific expression is as follows: Score=P pt ×σ+(1-σ)×P st ; Among them: Score is the status score of the current radar monitoring point, that is, the fitness function of the radar deployment location; σ is the weight.

3. The method for selecting a site for ground deformation monitoring radar according to claim 2, characterized in that: The specific calculation formula for the number of point clouds N of the thinned point cloud model in S4.1 is as follows: N=S m / P all ×(Δr×Δr)×P tn ; Where: S m is the area of ​​the area to be monitored, P all is the number of point clouds in the 3D point cloud model of the area to be monitored, and Δr is the length of the grid edge.

4. The method for selecting a site for ground deformation monitoring radar according to claim 3, characterized in that: The S4.1 includes: Divide the monitoring area to be tested into key monitoring areas and non-key monitoring areas based on geological environment data; Calculate the number of thinned point clouds in the cells in the key monitoring area and the non-key monitoring area respectively, where: the P corresponding to the cells in the key monitoring area tn The P value is greater than the P value of the cell in the non-key monitoring area. tn Get value; The number of point clouds in the thinned point cloud model is obtained by adding the number of thinned point clouds in the cells in the key monitoring area and the non-key monitoring area.

5. The method for selecting a site for ground deformation monitoring radar according to claim 3, characterized in that: The S5 includes: S5.

1. Generate a uniformly distributed particle group near the radar monitoring point, with one particle corresponding to one grid; S5.2, particle swarm initialization: Initialize the particle positions in the particle swarm, that is, randomly select a grid position as the initial deployment point of the radar, and randomly initialize the particle velocity, where: the position of each particle is a two-dimensional coordinate, that is, the coordinate of the particle corresponding to the grid in the radar deployment area; S5.

3. Calculate the fitness function value of each particle in the particle swarm in t iterations, obtain the historical optimal position pBesti(t) and the global optimal position gBest(t) in the t iteration, and the particle position corresponding to the maximum fitness function value is the historical optimal position pBest(t) of the particle in t iterations, where: t is the number of iterations, which is a natural number greater than 0; if pBesti(t)≥pBesti(t-1) and pBesti(t)≥gBest(t-1), then the global optimal position gBest(t)=pBesti(t) in t iterations, otherwise gBest(t)=gBest(t-1); when t=1, gBest(1)=pBesti(1); S5.

4. If t reaches the maximum iteration number threshold, or the fitness function value of the global optimal position changes less than the preset threshold compared with the t-1th time, the iteration stops, and the output global optimal position gBest(t) is the final site selection result of the radar; otherwise, let t = t + 1, and call the update formula to update the particle swarm position and speed, and return to S5.

3.

6. The method for selecting a site for ground deformation monitoring radar according to claim 5, characterized in that: In S4.3, In S4.4, 0.3≤σ≤0.

5.

7. The method for selecting a site for a ground deformation monitoring radar according to any one of claims 1 to 6, characterized in that: The ground feature features include texture features, shape and structure features, and spatial relationship features; The types of land features include rocks, trees, water flows, soil surfaces, grass and other land features.

8. A readable storage medium, characterized in that: Computer program instructions are stored thereon, and when the computer program instructions are executed by a processor, the method for selecting a site for a ground deformation monitoring radar as described in any one of claims 1 to 6 is implemented.

9. An electronic device, characterized in that: include: At least one processor, at least one memory and computer program instructions stored in the memory, when the computer program instructions are executed by the processor, the method for selecting a site for a ground deformation monitoring radar as described in any one of claims 1 to 6.

Images