Hydraulic engineering video monitoring intelligent stationing optimization design method based on three-dimensional model

Abstract

A hydraulic engineering video monitoring intelligent stationing optimization design method based on a three-dimensional model realizes video monitoring intelligent stationing planning design based on a mode of combined development of a three-dimensional modeling platform and a BIM lightweight engine. The method comprises five stages of camera adaptive arrangement, monitoring area three-dimensional visualization, monitoring coverage check, design optimization and design result display output. According to the method, the overall planning of the video monitoring intelligent stationing of the water conservancy and hydropower engineering is enhanced, the advantages and disadvantages of the video monitoring stationing are scientifically and quantitatively evaluated, and a quantifiable video monitoring intelligent stationing planning scheme is provided. According to the method, the requirements of the current video monitoring distribution point and the monitoring coverage rate are combined, a monitoring model optimization algorithm is utilized, the monitoring coverage rate or the camera configuration is taken as an optimization target, the monitoring coverage rate is improved, the monitoring blind area range is reduced, monitoring redundancy setting is avoided, and the monitoring camera configuration is optimized.

Description

Technical Field

[0001] The invention belongs to the field of water conservancy project video monitoring systems, and in particular relates to a water conservancy project video monitoring intelligent point layout optimization design method based on a three-dimensional model. Background Art

[0002] The amount of calculation required to achieve intelligent video surveillance point planning and coverage optimization is large, and the original empirical design calculation obviously cannot meet the requirements of intelligent design. At present, some researchers in the industry use simulation tools such as Matlab to plan video surveillance points and optimize coverage. However, this method is only applicable to simple model scenarios and is difficult to apply to complex three-dimensional spatial models in the water conservancy engineering industry and application scenarios with many visual obstacles. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a method for optimizing the intelligent layout of video surveillance points for water conservancy projects based on a three-dimensional model. In combination with the current needs of video surveillance layout and surveillance coverage, a monitoring model optimization algorithm is used to optimize the surveillance coverage or camera configuration, thereby improving the surveillance coverage, reducing the scope of surveillance blind spots, avoiding redundant surveillance settings, and optimizing the surveillance camera configuration.

[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is: A method for optimizing the intelligent layout of water conservancy project video monitoring based on a three-dimensional model, comprising the following steps: Perform adaptive camera layout: realize adaptive layout of video cameras based on 3D BIM model, including camera position layout, camera angle layout, camera translation and rotation, and camera addition, copy and deletion items; Conduct 3D visualization of the monitoring area: including one-click generation of visualization area, classification of monitoring area types, parametric modeling of monitoring area, and setting adjustable drawing effects; Perform monitoring coverage verification: including the generation of monitoring coverage verification results, coverage calculation, repetition rate calculation, area highlighting and setting text box prompts; Perform design optimization: including the surveillance camera angle optimization layout algorithm based on virtual force, iterative optimization of given cameras, and iterative optimization of given coverage; Display design results: including 3D visualization output of monitoring area, camera layout design output and monitoring coverage analysis and calculation output.

[0005] Preferably, the method for performing adaptive camera arrangement is: Step 1.1: Generate a camera pre-deployment plan, including the following steps: ① Model import: The import of the 3D model of the water conservancy project monitored by the video surveillance system mainly includes: importing the model generated by 3D modeling software, importing the 3D model synthesized from the 2D drawings of the traditional built water conservancy project, and importing the 3D model obtained by laser scanning; ② Establish a knowledge base: By specifying the camera deployment priority, the implementation priority range, and the number and type of cameras in each monitoring area, it provides a basis for automatically generating camera installation plans; ③ Generate camera pre-deployment plan: Based on the existing rule knowledge base definition, generate a monitoring plan for the specified monitoring equipment for the currently imported water conservancy and hydropower engineering model scene; Step 1.2: Arrange the camera positions. There are two ways to arrange the camera positions: ① Pick up and locate, automatically capture the model positioning coordinates; ② Input coordinates in the dialog box for precise positioning; Step 1.3: Arrange the camera angle. There are two ways to arrange the camera angle: ①After the camera position is determined, the angle between the camera and other surfaces is constrained; ②In the dialog box, enter the angle relationship between the camera axis and the installation plane, and the camera angle layout will be automatically generated when the camera is arranged; Step 1.4: Pan and rotate the camera: After the camera is arranged, you can quickly select it in three dimensions and set the pan distance and rotation angle; Step 1.5: Set the camera deletion function. Camera deletion includes two forms: ①Support fast 3D point selection and deletion; ②In the dialog box, enter the camera number to be deleted; Step 1.6: Retrieve camera information. Camera information retrieval includes two forms: ①Supports the display of camera information list of video surveillance points, which is used to realize the linkage between camera information and 3D. Click any camera information in the video surveillance point list to automatically locate and display the 3D information of the camera in 3D; ②Supports camera quantity statistics and displays information on the number of various cameras in each area of ​​the project.

[0006] Preferably, the three-dimensional visualization stage of the monitoring area includes the steps of: Step 2.1: One-click generation of visualization area: Generate a visualization model of the monitoring area with one click. According to the relevant parameters of the camera, including the sensor size, focal length range, resolution, rotation angle and illumination, simulate the monitoring range of the camera in the three-dimensional scene, and realize the 3D field of view simulation of video monitoring according to the 3D field of view angle and 3D visible field length parameters of the camera; analyze and present the three-dimensional visualization monitoring area model of the virtual camera illumination range and the camera monitoring range through optical principles and camera parameters; Step 2.2: The monitoring area classification includes: ①Support the division of monitoring areas including confidential areas, restricted areas, protection areas, and surveillance areas, and set different coverage requirements for different areas; ②Support the setting of obstacle areas; Step 2.3: Parametric modeling of camera monitoring area: input the camera field of view monitoring angle range and effective recognition distance parameters, and automatically generate the camera monitoring model based on this; Step 2.4: Adjustable drawing effect: Set the color and transparency of the monitoring area model to be adjustable.

[0007] Preferably, the monitoring coverage verification stage includes the steps of: Step 3.1: Generate monitoring coverage verification results with one click: Generate monitoring coverage verification results with one click, including grid division, automatic calculation, area display, text prompts and result warning functions; Step 3.2: Calculation of three-dimensional coverage rate R. Supports three-dimensional coverage rate calculation. The specific implementation method is to divide the model of the area to be monitored into several grid points (the grid size is adjustable), determine whether each unit grid point is in the monitoring area, and calculate the coverage rate based on the ratio of the grid points in the monitoring area to the total monitoring grid points. The calculation formula of the three-dimensional coverage rate R is: ; Where: a, b, c are the total number of grid points in the x, y, and z coordinate directions after the total area is discretely gridded; M(i, j, k) is the number of grid points P i (x i ,y i ,z i ) is a judgment value of whether the node is covered. If the point is covered, M takes the value of 1, otherwise, if it is not covered, M takes the value of 0.

[0008] Step 3.3: Calculate the repetition rate: Calculate the repetition rate of the repeated coverage area in the same way as the coverage rate, except that it needs to be multiplied by a weighting factor, that is, the number of single camera monitoring areas where the repeated coverage area is located; Step 3.4: Area highlighting: Set the highlighting of the blind areas that are not covered by monitoring, the areas with repeated coverage, and the areas where the coverage rate does not meet the standard; Step 3.5: Text box prompts mainly include the following 4 types: ① Set text box prompts for the numbers of blind areas that are not covered by monitoring, areas with repeated coverage, and areas where coverage does not meet the standards; ② For the repeated coverage area, the repeated coverage camera number should be displayed; ③ For areas with coverage that meet the standard and areas that do not meet the standard, and areas with repetition rate that meet the standard and areas that do not meet the standard, text boxes should be used to indicate the area name and specific coverage and repetition rate values; ④ Set text box prompts for video surveillance camera applause not processed, applause processed, abnormal warning, and replacement prompts. Preferably, the design optimization stage comprises the steps of: Step 4.1: Iterative optimization algorithm for video camera placement based on virtual force; Step 4.2: Given the number of cameras, iterate to meet the coverage requirements; Step 4.3: Given the coverage requirement, iterate to find the optimal camera layout.

[0009] Preferably, the design results display output stage includes the following steps: Step 5.1: 3D visualization output of monitoring area: Set up 360° display of monitoring area without blind spots, and the results are intuitive; Step 5.2: One-click output of camera layout design results: Set one-click output of camera layout design results, which include parameters such as quantity, coordinates, angle, camera monitoring angle range, and effective recognition distance; Step 5.3: The output of monitoring coverage analysis calculation results mainly includes: ① Set the output to focus on the geometric parameters within the monitoring area and the analysis and calculation results of coverage, repetition rate, blind area ratio, and number of cameras; ②The area of ​​focus is selected before the results are generated.

[0010] Preferably, in step 3.2, the three-dimensional coverage rate calculation is performed by establishing a three-dimensional coverage rate calculation model for multi-viewpoint video surveillance. The three-dimensional coverage rate calculation method of the multi-viewpoint video surveillance system is used to evaluate the coverage of a group of monitoring nodes on the monitored object. Both the camera and the covered object are three-dimensional models, and the input of the coverage rate calculation method is the given covered object and the relevant parameters of the camera, which includes the following steps: ① Establish a three-dimensional coordinate system, transform the monitored object into a rectangular model, and establish a three-dimensional camera model based on the camera parameters and installation position; ② Determine whether the camera can effectively cover the object of the rectangular model based on the monitoring radius; ③ When effective coverage is possible, project the camera and the covered object of the rectangular model on the horizontal plane, and calculate the three-dimensional coverage angle formed by the camera on the covered object of the rectangular model; ④ By taking the union of the coverage angles of multiple cameras, the three-dimensional coverage rate of a group of cameras on the covered object is solved.

[0011] Preferably, the iterative optimization algorithm for video camera arrangement based on virtual force in step 4.1, on the basis of the traditional virtual force method, gives a definition of the gravitational force of the uncovered area on the geometric center of the camera detection fan-shaped area; then, the total coverage rate of the video camera network is used as the objective function, and a discriminant mathematical expression for detecting blind spot nodes in the presence of obstacles is established; since the iterative optimization process is directly related to the total coverage rate, it is used to improve the detection efficiency of the video camera network.

[0012] Preferably, when there are multiple obstacles in the application scenario, a traversal order problem is involved. In this case, the iterative optimization process is as follows: Step a: Area division: Divide the three-dimensional space into camera deployment area, obstacle area, monitoring area and non-monitoring area. When designing the coverage algorithm, only the effective area of ​​the monitoring area is considered; Step b: Region rasterization: The divided regions are divided into adjacent fine-grained cubic blocks, and the center of each small block is used as the block representative point, that is, the virtual force action point; Step c: Equipment deployment: According to the actual situation, the deployable positions of camera equipment are relatively fixed. During the simulation, some deployable points are randomly set, and some equipment with random monitoring perception directions are placed; Step d: Calculate the virtual force F: for the first coordinate S(x 0 ,y 0 ,z 0 ) camera sensor, searching for all grid points P within the gravitational range of its center i (x i ,y i ,z i ), determine whether it is an uncovered node. If it is covered, determine the next grid point. Otherwise, the grid point exerts a gravitational force on the sensor center and takes a certain adjustment angle ω. The calculation formulas of the adjustment angle ω and the virtual resultant force F are: ; ; Where: α is the adjustment step size; F αis the adjusted virtual force; e is the natural base; G is the gravitational constant.

[0013] Step e: Calculate the three-dimensional coverage: Calculate the coverage of the first sensor after adjusting the impact of all grid points within its detection range, and continue to make similar adjustments to the next sensor, traversing each sensor in turn and calculating the coverage each time.

[0014] Step f: Repeat steps d and e until the three-dimensional coverage reaches the required standard; or perform a finite number of iterations to output the camera sensor detection direction when the three-dimensional coverage takes the maximum value.

[0015] A water conservancy project video monitoring intelligent point distribution optimization system adopts a program instruction for a water conservancy project video monitoring intelligent point distribution optimization design method based on a three-dimensional model.

[0016] The present invention can achieve the following beneficial effects: 1. This method realizes the planning and design of intelligent video surveillance point layout based on the combination of 3D modeling platform and BIM lightweight engine. It combines the current needs of video surveillance point layout and surveillance coverage, uses the surveillance model optimization algorithm, and takes surveillance coverage or camera configuration as the optimization target to improve surveillance coverage, reduce the scope of surveillance blind spots, avoid redundant surveillance settings, and optimize surveillance camera configuration.

[0017] 2. This method includes five steps: adaptive camera layout, three-dimensional visualization of monitoring area, monitoring coverage verification, design optimization, and display and output of design results. It makes the optimal design of video monitoring points for water conservancy projects more scientific, accurate, efficient, and intelligent, and has important guiding significance for the design of video monitoring systems for water conservancy projects.

[0018] 3. Assist the whole process of video surveillance point design for water conservancy projects, making the optimization design of video surveillance point layout for water conservancy projects more scientific, accurate, efficient and intelligent; 4. The present invention realizes the optimization design of the camera point distribution of the video surveillance system based on the three-dimensional surveillance coverage, studies the surveillance model optimization algorithm, improves the surveillance coverage, reduces the surveillance blind area, avoids the redundant surveillance settings, and optimizes the surveillance camera configuration; 5. The present invention realizes the three-dimensional visualization display and delivery of video surveillance, improves the visualization level of video surveillance system design, and realizes the intelligent design, auxiliary construction, and visualization delivery of results of the video surveillance system; 6. The present invention realizes automation and intelligence in the planning and design process of video surveillance intelligent deployment, greatly improves the design efficiency and quality of the video surveillance system, and reduces the number of rework times in the design of the video surveillance system. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The present invention will be further described below in conjunction with the accompanying drawings and embodiments: Figure 1 It is a schematic diagram of the whole process of a water conservancy project video monitoring intelligent point layout optimization design method based on a three-dimensional model of the present invention; Figure 2 It is a schematic diagram of information retrieval of a video surveillance camera of the present invention; Figure 3 A schematic diagram of a three-dimensional visualization model of the monitoring range of a video surveillance camera of the present invention; Figure 4 It is a schematic diagram of the three-dimensional coverage calculation model of the video surveillance camera of the present invention. DETAILED DESCRIPTION

[0020] The preferred solution is Figures 1 to 4 As shown in the figure, a water conservancy project video surveillance intelligent layout optimization design method based on a three-dimensional model is developed by combining a three-dimensional modeling platform and a BIM lightweight engine to realize the full process planning and design of video surveillance intelligent layout, including five stages: camera adaptive layout, three-dimensional visualization of monitoring area, monitoring coverage verification, design optimization, and design results display and output. It is characterized by: The camera adaptive layout stage realizes the adaptive layout of video cameras based on the three-dimensional BIM model in the intelligent layout design of the video surveillance system, including the following steps: Step 1.1: Generate camera pre-location plan. It mainly includes the following steps: ① Model import. The import of the three-dimensional model of the water conservancy project monitored by the video surveillance system mainly includes: importing models generated by three-dimensional modeling software, importing three-dimensional models synthesized from two-dimensional drawings of traditional built water conservancy projects, and importing three-dimensional models obtained by laser scanning.

[0021] ② Establish a knowledge base. By specifying the camera deployment priority, the implementation priority range, and the number and type of cameras in each monitoring area, a basis is provided for automatically generating camera installation plans.

[0022] ③ Generate camera pre-deployment plan. According to the existing rule knowledge base definition, for the current imported water conservancy and hydropower engineering model scene, generate a monitoring plan for the specified monitoring equipment.

[0023] Step 1.2: Arrange the camera positions. There are two ways to arrange the camera positions: ① Pick up and locate, automatically capture the model positioning coordinates; ② Enter the coordinates in the dialog box for precise positioning.

[0024] Step 1.3: Arrange the camera angle. There are two ways to arrange the camera angle: ①After the camera position is determined, the angle between the camera and other surfaces is constrained; ②In the dialog box, enter the angle between the camera axis and the installation plane, and the camera angle layout will be automatically generated when the camera is arranged.

[0025] Step 1.4: Camera translation and rotation. After the camera is arranged, it supports quick 3D point selection and setting of translation distance and rotation angle.

[0026] Step 1.5: Camera deletion. Camera deletion supports two forms: ①Support fast 3D point selection and deletion; ②In the dialog box, enter the camera number to be deleted.

[0027] Step 1.6: Retrieve camera information, such as Figure 2 As shown. Camera information retrieval supports two forms: ①Supports the display of the camera information list of video surveillance points, and can realize the linkage between camera information and 3D. Click any camera information in the video surveillance point list to automatically locate and display the 3D information of the camera in 3D; ②Supports camera quantity statistics and displays information on the number of various cameras in each area of ​​the project.

[0028] The three-dimensional visualization stage of the monitoring area is an important stage for realizing the visualization of intelligent video surveillance point layout, and is an important link for realizing intelligent design, auxiliary construction, and visualized delivery, including the following steps: Step 2.1: The visualization area is generated with one click, such as Figure 3 As shown. It supports one-click generation of a visual model of the monitoring area. According to the relevant parameters of the camera (photoreceptor size, focal range, resolution, rotation angle, illumination, etc.), the monitoring range of the camera is simulated in a three-dimensional scene, and the 3D field of view simulation of video monitoring is realized according to the parameters such as the camera's 3D field of view angle and 3D visible field of view length. Through optical principles and camera parameters, the three-dimensional visual monitoring area model of the virtual camera's illumination range and the camera's monitoring range is analyzed and presented.

[0029] Step 2.2: Classification of monitoring areas. Including: ①Support the division of monitoring areas such as confidential areas, restricted areas, protection areas, and surveillance areas, and set different coverage requirements for different areas; ②Support the setting of obstacle area.

[0030] Step 2.3: Parametric modeling of camera monitoring area. You can input parameters such as the camera field of view monitoring angle range, effective recognition distance, etc., and automatically generate a camera monitoring model based on this.

[0031] Step 2.4: The drawing effect can be adjusted. The color and transparency of the monitoring area model can be adjusted.

[0032] The monitoring coverage verification stage is a key stage to ensure the accuracy of video surveillance intelligent deployment, ensuring a high quality of video surveillance intelligent deployment results, including the following steps: Step 3.1: One-click generation of monitoring coverage verification results. Supports one-click generation of monitoring coverage verification results, including grid division, automatic calculation, area display, text prompts, result warnings and other functions.

[0033] Step 3.2: Calculation of three-dimensional coverage rate R. Supports three-dimensional coverage rate calculation. The specific implementation method is to divide the model of the area to be monitored into several grid points (the grid size is adjustable), determine whether each unit grid point is in the monitoring area, and calculate the coverage rate based on the ratio of the grid points in the monitoring area to the total monitoring grid points. The calculation formula of the three-dimensional coverage rate R is: ; Where: a, b, c are the total number of grid points in the x, y, and z coordinate directions after the total area is discretely gridded; M(i, j, k) is the number of grid points P i (x i ,y i ,z i ) is a judgment value of whether the node is covered. If the point is covered, M takes the value of 1, otherwise, if it is not covered, M takes the value of 0.

[0034] Step 3.3: Calculation of repetition rate. The calculation of repetition rate of repeated coverage area is supported. The calculation method is similar to that of coverage rate, but it needs to be multiplied by the weighting coefficient, that is, the number of single camera monitoring areas where the repeated coverage area is located.

[0035] Step 3.4: Area highlighting: Supports highlighting of blind areas that are not covered by monitoring, areas with repeated coverage, and areas where coverage does not meet the standard.

[0036] Step 3.5: Text box prompts. There are mainly four types: ①Support text box prompts for the numbers of blind areas that are not covered by monitoring, areas with repeated coverage, and areas where coverage does not meet the standards; ② For the repeated coverage area, the repeated coverage camera number should be displayed; ③ For areas with coverage that meet the standard and areas that do not meet the standard, and areas with repetition rate that meet the standard and areas that do not meet the standard, text boxes should be used to indicate the area name and specific coverage and repetition rate values; ④Supports text box prompts for video surveillance camera applause that has not been processed, applause that has been processed, abnormal warnings, and replacement prompts.

[0037] The design optimization stage is a necessary condition for the optimization design of video surveillance coverage, and has a direct impact on the cost optimization of the monitoring system. According to the iterative optimization algorithm of video camera layout based on virtual force and the video surveillance requirements of water conservancy projects, the iterative optimization of video surveillance design is realized, and finally the three requirements for video surveillance are met. The first is that video surveillance is required to cover the entire monitoring area, that is, to increase the coverage rate of the monitoring area as much as possible; the second is to capture as many key monitoring targets as possible without increasing the number of front-end cameras; finally, the monitoring system is required to optimize the number of cameras to avoid duplicate construction while ensuring a certain coverage rate. Including steps: Step 4.1: Iterative optimization algorithm for video camera placement based on virtual force.

[0038] Step 4.2: Given the number of cameras, iterate to meet the coverage requirement.

[0039] Step 4.3: Given the coverage requirement, iterate to find the optimal camera layout.

[0040] The design results display output stage directly affects the final delivery effect of the intelligent monitoring point layout of the video surveillance system, including the following steps: Step 5.1: 3D visualization output of the monitoring area. Support 360° display of the monitoring area without blind spots, and the results are intuitive.

[0041] Step 5.2: One-click output of camera layout design results. Support one-click output of camera layout design results, including parameters such as quantity, coordinates, angle, camera monitoring angle range, effective recognition distance, etc.

[0042] Step 5.3: Monitor the output of coverage analysis calculation results. Mainly includes: ①Supports the output of the geometric parameters in the monitoring area, as well as the analysis and calculation results such as coverage, repetition rate, blind area ratio, and number of cameras; ②The key focus area can be selected before the results are generated.

[0043] Among them, the three-dimensional coverage calculation in step 3.2 establishes a multi-viewpoint video surveillance three-dimensional coverage calculation model, such as Figure 4 The three-dimensional coverage calculation method of the multi-viewpoint video surveillance system is used to evaluate the coverage of a group of monitoring nodes on the monitored object. The camera and the covered object are both three-dimensional models, which conforms to the actual situation. The input of the coverage calculation method is the given covered object and the relevant parameters of the camera, which includes the following steps: ① Establish a three-dimensional coordinate system, transform the monitored object into a rectangular model, and establish a three-dimensional camera model based on the camera parameters and installation position; ② Determine whether the camera can effectively cover the object of the rectangular model based on the monitoring radius; ③ When effective coverage is possible, project the camera and the covered object of the rectangular model on the horizontal plane, and calculate the three-dimensional coverage angle formed by the camera on the covered object of the rectangular model; ④ By taking the union of the coverage angles of multiple cameras, the three-dimensional coverage rate of a group of cameras on the covered object is solved.

[0044] Among them, the virtual force-based iterative optimization algorithm for video camera placement in step 4.1, based on the traditional virtual force method, gives a definition of the gravity of the uncovered area on the geometric center of the camera detection fan-shaped area; then, the total coverage rate of the video camera network is used as the objective function, and a discriminant mathematical expression for detecting blind spot nodes in the presence of obstacles is established. Since the iterative optimization process is directly related to the total coverage rate, the problem of the iterative optimization process not being easy to converge is overcome, and the detection efficiency of the video camera network is effectively improved.

[0045] When there are multiple obstacles in the application scenario, a traversal order problem is involved. The iterative optimization process is as follows.

[0046] Step a: Area division. The three-dimensional space area is divided into camera deployment area, obstacle area, monitoring area, non-monitoring area, etc. When designing the coverage algorithm, only the effective area of ​​the monitoring area is considered.

[0047] Step b: Region rasterization. The divided regions are divided into adjacent fine-grained cubic blocks, and the center of each small block is used as the block representative point, that is, the virtual force application point.

[0048] Step c: Equipment deployment. According to the actual situation, the deployable positions of camera equipment are relatively fixed. During the simulation, some deployable points are randomly set, and some equipment with random monitoring perception directions are placed.

[0049] Step d: Calculate the virtual force F: for the first coordinate S(x 0 ,y 0 ,z 0 ) camera sensor, searching for all grid points P within the gravitational range of its center i (x i ,y i ,z i ), determine whether it is an uncovered node. If it is covered, determine the next grid point. Otherwise, the grid point exerts a gravitational force on the sensor center and takes a certain adjustment angle ω. The calculation formulas of the adjustment angle ω and the virtual resultant force F are: ; ; Where: α is the adjustment step size; F α is the adjusted virtual force; e is the natural base; G is the gravitational constant.

[0050] Step e: Calculate the three-dimensional coverage: Calculate the coverage of the first sensor after adjusting the impact of all grid points within its detection range, and continue to make similar adjustments to the next sensor, traversing each sensor in turn and calculating the coverage each time.

[0051] Step f: Repeat steps d and e until the three-dimensional coverage reaches the required standard; or perform a finite number of iterations to output the camera sensor detection direction when the three-dimensional coverage takes the maximum value.

[0052] The above embodiments are only preferred technical solutions of the present invention and should not be regarded as limiting the present invention. The protection scope of the present invention shall be the technical solutions recorded in the claims, including equivalent replacement solutions of the technical features in the technical solutions recorded in the claims. That is, equivalent replacement improvements within this scope are also within the protection scope of the present invention.

Claims

1. A method for optimizing the intelligent layout of water conservancy project video monitoring based on a three-dimensional model, characterized in that: Perform adaptive camera layout: realize adaptive layout of video cameras based on 3D BIM model, including camera position layout, camera angle layout, camera translation and rotation, and camera addition, copy and deletion items; Conduct 3D visualization of the monitoring area: including one-click generation of visualization area, classification of monitoring area types, parametric modeling of monitoring area, and setting adjustable drawing effects; Perform monitoring coverage verification: including the generation of monitoring coverage verification results, coverage calculation, repetition rate calculation, area highlighting and setting text box prompts; Perform design optimization: including the surveillance camera angle optimization layout algorithm based on virtual force, iterative optimization of given cameras, and iterative optimization of given coverage; Display design results: including 3D visualization output of monitoring area, camera layout design output and monitoring coverage analysis and calculation output.

2. According to the three-dimensional model-based water conservancy project video monitoring intelligent point optimization design method of claim 1, it is characterized by: The method for performing adaptive camera arrangement is as follows: Step 1.1: Generate a camera pre-deployment plan, including the following steps: ① Model import: The import of the 3D model of the water conservancy project monitored by the video surveillance system mainly includes: importing the model generated by 3D modeling software, importing the 3D model synthesized from the 2D drawings of the traditional built water conservancy project, and importing the 3D model obtained by laser scanning; ② Establish a knowledge base: By specifying the camera deployment priority, the implementation priority range, and the number and type of cameras in each monitoring area, it provides a basis for automatically generating camera installation plans; ③ Generate camera pre-deployment plan: Based on the existing rule knowledge base definition, generate a monitoring plan for the specified monitoring equipment for the currently imported water conservancy and hydropower engineering model scene; Step 1.2: Arrange the camera positions. There are two ways to arrange the camera positions: ① Pick up and locate, automatically capture the model positioning coordinates; ② Input coordinates in the dialog box for precise positioning; Step 1.3: Arrange the camera angle. There are two ways to arrange the camera angle: ①After the camera position is determined, the angle between the camera and other surfaces is constrained; ②In the dialog box, enter the angle relationship between the camera axis and the installation plane, and the camera angle layout will be automatically generated when the camera is arranged; Step 1.4: Pan and rotate the camera: After the camera is arranged, you can quickly select it in three dimensions and set the pan distance and rotation angle; Step 1.5: Set the camera deletion function. Camera deletion includes two forms: ①Support fast 3D point selection and deletion; ②In the dialog box, enter the camera number to be deleted; Step 1.6: Retrieve camera information. Camera information retrieval includes two forms: ①Supports the display of camera information list of video surveillance points, which is used to realize the linkage between camera information and 3D. Click any camera information in the video surveillance point list to automatically locate and display the 3D information of the camera in 3D; ②Supports camera quantity statistics and displays information on the number of various cameras in each area of ​​the project.

3. The method for optimizing the intelligent layout of water conservancy project video monitoring based on a three-dimensional model according to claim 1 is characterized in that: The three-dimensional visualization stage of the monitoring area includes the following steps: Step 2.1: One-click generation of visualization area: Generate a visualization model of the monitoring area with one click. According to the relevant parameters of the camera, including the sensor size, focal length range, resolution, rotation angle and illumination, simulate the monitoring range of the camera in a three-dimensional scene, and realize the 3D field of view simulation of video monitoring according to the 3D field of view angle and 3D visible field of view length parameters of the camera; Analyze and present the 3D visual monitoring area model of the virtual camera illumination range and camera monitoring range through optical principles and camera parameters; Step 2.2: The monitoring area classification includes: ①Support the division of monitoring areas including confidential areas, restricted areas, protection areas, and surveillance areas, and set different coverage requirements for different areas; ②Support the setting of obstacle areas; Step 2.3: Parametric modeling of camera monitoring area: input the camera field of view monitoring angle range and effective recognition distance parameters, and automatically generate the camera monitoring model based on this; Step 2.4: Adjustable drawing effect: Set the color and transparency of the monitoring area model to be adjustable.

4. The method for optimizing the intelligent layout of water conservancy project video monitoring based on a three-dimensional model according to claim 1 is characterized in that: The monitoring coverage verification stage includes the following steps: Step 3.1: Generate monitoring coverage verification results with one click: Generate monitoring coverage verification results with one click, including grid division, automatic calculation, area display, text prompts and result warning functions; Step 3.2: Calculation of three-dimensional coverage rate R: Supports three-dimensional coverage rate calculation, which is implemented by dividing the model of the area to be monitored into several grid points, determining whether each unit grid point is within the monitoring area, and obtaining the coverage rate based on the ratio of the grid points in the monitoring area to the total monitoring grid points; the calculation formula of the three-dimensional coverage rate R is: ； Where: a, b, c are the total number of grid points in the x, y, and z coordinate directions after the total area is discretely gridded; M(i, j, k) is the number of grid points P i (x i ,y i ,z i ) is the judgment value of whether the node is covered. If the node is covered, M takes the value of 1, otherwise, if it is not covered, M takes the value of 0; Step 3.3: Calculate the repetition rate: Calculate the repetition rate of the repeated coverage area in the same way as the coverage rate, except that it needs to be multiplied by a weighting factor, that is, the number of single camera monitoring areas where the repeated coverage area is located; Step 3.4: Area highlighting: Set the highlighting of the blind areas that are not covered by monitoring, the areas with repeated coverage, and the areas where the coverage rate does not meet the standard; Step 3.5: Text box prompts mainly include the following 4 types: ① Set text box prompts for the numbers of blind areas that are not covered by monitoring, areas with repeated coverage, and areas where coverage does not meet the standards; ② For the repeated coverage area, the repeated coverage camera number should be displayed; ③ For areas with coverage that meet the standard and areas that do not meet the standard, and areas with repetition rate that meet the standard and areas that do not meet the standard, text boxes should be used to indicate the area name and specific coverage and repetition rate values; ④ Set text box prompts for video surveillance camera applause not processed, applause processed, abnormal warning, and replacement prompts.

5. The method for optimizing the intelligent layout of water conservancy project video monitoring based on a three-dimensional model according to claim 1 is characterized by: The design optimization phase includes the steps of: Step 4.1: Iterative optimization algorithm for video camera placement based on virtual force; Step 4.2: Given the number of cameras, iterate to meet the coverage requirements; Step 4.3: Given the coverage requirement, iterate to find the optimal camera layout.

6. The method for optimizing the intelligent layout of water conservancy project video monitoring based on a three-dimensional model according to claim 1 is characterized by: The design results display output stage includes the following steps: Step 5.1: 3D visualization output of monitoring area: Set up 360° display of monitoring area without blind spots, and the results are intuitive; Step 5.2: One-click output of camera layout design results: Set one-click output of camera layout design results, which include parameters such as quantity, coordinates, angle, camera monitoring angle range, and effective recognition distance; Step 5.3: The output of monitoring coverage analysis calculation results mainly includes: ① Set the output to focus on the geometric parameters within the monitoring area and the analysis and calculation results of coverage, repetition rate, blind area ratio, and number of cameras; ②The area of ​​focus is selected before the results are generated.

7. The method for optimizing the intelligent layout of water conservancy project video monitoring based on a three-dimensional model according to claim 3 is characterized by: In step 3.2, the three-dimensional coverage calculation is performed to establish a three-dimensional coverage calculation model for multi-view video surveillance. The three-dimensional coverage calculation method of the multi-view video surveillance system is used to evaluate the coverage of a group of monitoring nodes on the monitored object. Both the camera and the covered object are three-dimensional models, and the input of the coverage calculation method is the given covered object and the relevant parameters of the camera, which includes the following steps: ① Establish a three-dimensional coordinate system, transform the monitored object into a rectangular model, and establish a three-dimensional camera model based on the camera parameters and installation position; ② Determine whether the camera can effectively cover the object of the rectangular model based on the monitoring radius; ③ When effective coverage is possible, project the camera and the covered object of the rectangular model on the horizontal plane, and calculate the three-dimensional coverage angle formed by the camera on the covered object of the rectangular model; ④ By taking the union of the coverage angles of multiple cameras, the three-dimensional coverage rate of a group of cameras on the covered object is solved.

8. The method for optimizing the intelligent layout of water conservancy project video monitoring based on a three-dimensional model according to claim 5 is characterized by: The iterative optimization algorithm for video camera layout based on virtual force in step 4.1, on the basis of the traditional virtual force method, gives the definition of the gravitational force of the uncovered area on the geometric center of the camera detection fan-shaped area; then, the total coverage rate of the video camera network is used as the objective function, and a discriminant mathematical expression for detecting blind spot nodes in the presence of obstacles is established; since the iterative optimization process is directly related to the total coverage rate, it is used to improve the detection efficiency of the video camera network.

9. The method for optimizing the intelligent layout of water conservancy project video monitoring based on a three-dimensional model according to claim 8 is characterized by: When there are multiple obstacles in the application scenario, it involves a traversal order problem. At this time, the iterative optimization process is as follows: Step a: Area division: Divide the three-dimensional space into camera deployment area, obstacle area, monitoring area and non-monitoring area. When designing the coverage algorithm, only the effective area of ​​the monitoring area is considered; Step b: Region rasterization: The divided regions are divided into adjacent fine-grained cubic blocks, and the center of each small block is used as the block representative point, that is, the virtual force action point; Step c: Equipment deployment: According to the actual situation, the deployable positions of camera equipment are relatively fixed. During the simulation, some deployable points are randomly set, and some equipment with random monitoring perception directions are placed; Step d: Calculate the virtual force F: For the camera sensor at the first coordinate S(x0,y0,z0), search for all grid points P within the gravitational range of its center i (x i ,y i ,z i ), determine whether it is an uncovered node. If it is covered, determine the next grid point. Otherwise, the grid point exerts a gravitational force on the sensor center and takes a certain adjustment angle ω. The calculation formulas of the adjustment angle ω and the virtual resultant force F are: ； ； Where: α is the adjustment step size; F α is the adjusted virtual force; e is the natural base; G is the gravitational constant; Step e: Calculate the three-dimensional coverage: Calculate the coverage of all grid points within the detection range of the first sensor after adjusting its influence, and continue to make similar adjustments to the next sensor, traversing each sensor in turn, and calculating the coverage each time the adjustment is made; Step f: Repeat steps d and e until the three-dimensional coverage reaches the required standard; or perform a finite number of iterations to output the camera sensor detection direction when the three-dimensional coverage takes the maximum value.

10. A water conservancy project video monitoring intelligent point optimization system, characterized by: The program instructions used are a three-dimensional model-based water conservancy project video monitoring intelligent layout optimization design method according to any one of claims 1-9.

Images