A video acquisition equipment layout optimization method and system for line monitoring

Abstract

The present invention discloses a method and system for optimizing the layout of video acquisition equipment for line monitoring, belonging to the technical field of equipment layout optimization. The method comprises: generating initial installation positions for scene cameras and local cameras, dividing the layout area into multiple line sections, and obtaining impact data for each line section; integrating the line sections into multiple types of sub-sections based on the impact data, and evaluating each type of sub-section to obtain an evaluation grade for the line section; screening complex sections based on the evaluation grades, scanning the complex sections to generate three-dimensional point cloud data, and constructing a three-dimensional model based on the three-dimensional point cloud data; analyzing the three-dimensional model to obtain alternative installation positions for multiple cameras, and generating an optimization scheme for optimizing the positions of the scene cameras and local cameras based on the initial installation positions and the alternative installation positions. The present invention optimizes the installation positions to address problems existing in the monitoring area, thereby maximizing the monitoring effect of the scene cameras and local cameras.

Description

Technical Field

[0001] The present invention belongs to the technical field of video surveillance, and in particular relates to a method and system for optimizing the layout of video acquisition equipment for line monitoring. Background Art

[0002] For monitoring and maintenance of high-speed rail cable lines, existing camera deployment solutions typically rely on manual experience and fixed deployment principles. This results in the camera installation locations being untargeted and unable to effectively cover key areas of the line.

[0003] In order to solve the blind spot problem in large-scale monitoring scenarios, the existing technology has proposed the following technical solutions. For example, the Chinese patent document with publication number CN116227110A discloses a multi-objective layout optimization method for road traffic facilities considering quantity constraints. Under the condition of considering quantity constraints, this method analyzes the indicators to be optimized for three alternative points: traffic congestion, traffic accidents, and traffic violations, and improves the layout density of monitoring points in the road network to achieve the task of regional blind spot control. For example, a Chinese patent document with publication number CN120105012A discloses a method and system for optimizing the layout of video observation points, which performs correlation analysis on traffic flow parameters of existing observation points, and preliminarily determines the observation points that can be transferred and the locations of new observation points based on the correlation; performs regression analysis on traffic flow parameters, fits the flow and speed of the road section, and determines the optimal fitting function; determines the extreme value positions of the flow and speed based on the optimal fitting function, and determines the road section position according to the extreme value position as the specific location of the new observation points; considers the cost and expected benefits of the new observation points, and determines the optimal layout plan for the new observation points.

[0004] However, the above technical solution is applied in urban and highway scenarios. For the maintenance of high-speed rail lines, the influence of line complexity, terrain variability, etc. needs to be considered. Therefore, the above existing technology cannot be used to solve the monitoring optimization problem in high-speed rail cable lines. Summary of the Invention

[0005] To solve the above problems, the present invention provides a method, system and storage medium for optimizing the layout of video acquisition equipment for line monitoring, so as to solve the problems existing in the above background technology.

[0006] In order to achieve the above-mentioned object of the invention, the present invention proposes a method for optimizing the layout of video acquisition equipment for line monitoring, comprising:

[0007] Determine the layout area of ​​the line and generate an initial installation plan for the cameras in the layout area. The initial installation plan includes the initial installation positions of the scene cameras and local cameras.

[0008] Install cameras based on the initial installation plan, divide the deployment area into multiple line sections based on line length and terrain data, and obtain impact data for each line section after a predetermined time interval, including environmental data and line alarm data;

[0009] Based on the impact data, a clustering algorithm is used to integrate the line sections into multiple types of sub-sections. Each type of sub-section is evaluated to obtain the evaluation level of each line section under the sub-section.

[0010] Based on the evaluation level, some line sections are selected as complex sections, the complex sections are scanned to generate 3D point cloud data, and a 3D model is constructed based on the 3D point cloud data;

[0011] The three-dimensional model is analyzed to obtain alternative installation positions of multiple cameras, and an optimization scheme for optimizing the positions of scene cameras and local cameras is generated based on the initial installation position and the alternative installation positions.

[0012] Furthermore, analyzing the three-dimensional model to obtain candidate installation positions of multiple cameras includes the following steps:

[0013] A plurality of monitoring targets are marked in a three-dimensional model, and the three-dimensional model is gridded to obtain a plurality of spatial grids. The spatial grids are divided into line grids and non-line grids based on whether the spatial grids are components of the monitoring targets. A plurality of first values ​​are set based on parameters of a scene camera. A generation point is set in the three-dimensional model, and a plurality of spherical areas are generated with the generation point as the center and the first value as the radius. A plurality of monitoring points are set at the boundaries of the spherical areas. The center point of each line grid is located, and the monitoring points are connected to the center point to obtain a visible line. If the visible line passes through at least one line grid, the visible line is deleted.

[0014] Count the total number of visible lines at each monitoring point, delete the monitoring points whose total number is less than the second value, and obtain the occlusion rate of the camera at the initial installation position. If the occlusion rate is greater than the first threshold, the initial installation position is eliminated, and the retained monitoring points and the initial installation position that are not eliminated are used as alternative installation positions.

[0015] Furthermore, generating an optimization solution for the scene camera based on the alternative installation positions includes the following steps:

[0016] Set the starting value, which is the default number of scene cameras to be installed. Based on the starting value, traverse and combine the alternative installation positions to obtain multiple basic installation solutions.

[0017] A basic installation scheme is selected, and the selected basic installation scheme is simulated based on the 3D model. Based on the simulation results, the effective shooting value of each scene camera under the basic installation scheme is obtained. The higher the effective shooting value, the more necessary the scene camera is to be installed. If the distribution of the effective shooting values ​​of the basic installation scheme meets the first condition, the number of scene cameras is reduced. If the distribution of the effective shooting values ​​meets the second condition, the number of scene cameras is increased. This step is repeated until the distribution of the effective shooting values ​​no longer meets the first and second conditions. The basic installation scheme at this time is defined as an advanced scheme.

[0018] Based on each basic installation solution, a corresponding advanced solution is obtained, and based on the installation cost and area coverage of each advanced solution, an evaluation value is obtained. Based on the evaluation value, an advanced solution is selected as the optimization solution for the scene camera.

[0019] Furthermore, obtaining the effective shooting values ​​of the cameras in each scene under the basic installation scheme based on the simulation results includes the following steps:

[0020] Based on the degree of importance, a different monitoring level is set for the surface of each line grid. Based on the size of the monitoring level, the surface of the corresponding line grid is divided into a corresponding number of monitoring areas. The shooting space of each scene camera is determined based on the camera parameters of the scene camera. The first number of monitoring areas included in the shooting space is located. The same monitoring areas in the shooting space of each scene camera are merged, and the second number of redundant monitoring areas included in each scene camera after the merger is recorded. The difference between the first number and the second number is used as the effective shooting value of the corresponding scene camera.

[0021] Furthermore, reducing the number of scene cameras includes the following steps:

[0022] If in the basic installation plan, the number of scene cameras with effective shooting values ​​lower than the second threshold is greater than the first value, the first condition is met, and the scene cameras with the smallest effective shooting value are eliminated in sequence until there are greater than or equal to the third number of monitoring areas that are not included in the shooting space after elimination.

[0023] Furthermore, increasing the number of scene cameras includes the following steps:

[0024] If in the basic installation plan, the effective shooting value is greater than the third threshold and the scene camera is less than the second value, and there are more than the third number of monitoring areas included in the shooting space, then the second condition is met, and an empty alternative installation position is selected in the three-dimensional model, and a scene camera is added in the selected alternative installation position. If the effective shooting value of the added scene camera is less than the second threshold, other alternative installation positions are selected to add scene cameras, and the alternative installation position is no longer used as the position for adding scene cameras in this basic installation plan.

[0025] Furthermore, generating an optimization solution for a local camera based on the alternative installation positions includes the following steps:

[0026] Obtain the occlusion rate of the current local camera. If the occlusion rate is greater than a first threshold, select an alternative installation position as an alternative position, set the local camera at the alternative position, and determine its shooting angle so that the local camera continues to shoot the original area.

[0027] The present invention also provides a video acquisition equipment layout optimization system for line monitoring, which is used to implement the above-mentioned method, and includes:

[0028] An input module, which determines a layout area of ​​the line, and is used to generate an initial installation plan for cameras in the layout area, wherein the initial installation plan includes initial installation positions of scene cameras and local cameras;

[0029] a division module, which installs cameras based on the initial installation plan, the division module dividing the deployment area into a plurality of line segments based on line length and terrain data, and acquiring impact data for each of the line segments after a predetermined time interval, the impact data including environmental data and line alarm data;

[0030] An evaluation module, based on the impact data, uses a clustering algorithm to integrate the route section into multiple types of sub-sections, evaluates each type of the sub-section, and obtains an evaluation grade for each of the route sections under the sub-section;

[0031] a construction module, screening out some of the line sections as complex sections based on the evaluation levels, scanning the complex sections to generate three-dimensional point cloud data, and constructing a three-dimensional model based on the three-dimensional point cloud data;

[0032] A generation module is used to analyze the three-dimensional model to obtain alternative installation positions of multiple cameras, and generate an optimization solution for optimizing the positions of the scene camera and the local camera based on the initial installation position and the alternative installation positions.

[0033] The present application also provides a computer-readable storage medium having instructions stored thereon. When the instructions are executed by a processor, the above-mentioned method for optimizing the layout of video acquisition equipment for line monitoring is implemented.

[0034] Beneficial Effects: Compared with existing technologies, this invention divides the deployment area into multiple line sections and collects environmental data and alarm data. Cluster analysis is then performed based on this data, thereby classifying line sections by similar characteristics. This facilitates subsequent hierarchical division of line sections, effectively screening out complex sections. Subsequent optimization is performed only on these complex sections, reducing unnecessary monitoring overhead and optimizing resource allocation. By constructing a 3D model based on 3D point cloud data for complex sections, a clear understanding of the terrain of different areas can be achieved. Based on this, the installation location can be optimized to address existing issues in the monitored area, thereby maximizing the monitoring effectiveness of scene cameras and local cameras and improving the safety monitoring capabilities of the entire line. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a flowchart of the steps of a method for optimizing the layout of video acquisition equipment for line monitoring according to the present invention;

[0036] Figure 2 This is a schematic diagram of the principle of determining monitoring points in the present invention. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0038] It is understood that the terms "first," "second," etc., used herein may be used to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script without departing from the scope of this application.

[0039] like Figure 1 As shown, a method for optimizing the layout of video acquisition equipment for line monitoring includes:

[0040] S1: Determine the layout area of ​​the line and generate an initial installation plan for the cameras in the layout area. The initial installation plan includes the initial installation positions of the scene cameras and local cameras.

[0041] Specifically, the deployment area is the high-speed rail cable laying area. Scene cameras are used to capture a wide range of scene video, while local cameras focus on smaller areas to monitor key components, such as cable connections. Scene cameras are typically installed at fixed intervals along the cable line. For example, based on the scene camera parameters, a group of scene cameras is placed every 100 meters. A telephoto camera is deployed at each cable connection for monitoring. The initial installation positions of the scene cameras and local cameras are determined using this method.

[0042] S2: Determine the layout area of ​​the line and generate an initial installation plan for the cameras in the layout area. The initial installation plan includes initial installation positions of scene cameras and local cameras.

[0043] After determining the initial installation plan, cameras are first installed based on the initial installation plan. The deployment area is then divided into multiple line sections based on the total length of the line and terrain data. For example, a line section is divided every 1 kilometer. After division, if a tunnel is divided into two line sections, the two line sections are merged. Alternatively, if a line section contains two types of terrain information, such as one section containing a tunnel and one section containing mountains, the line section is split. The ultimate goal of merging and splitting is to ensure that a line section only contains one type of terrain. After a predetermined period of time, such as one month, impact data for each line section is obtained. Environmental data includes temperature, wind speed, tree cover, altitude, and rainfall. A higher tree cover indicates a greater likelihood of camera obstruction. Line alarm data includes cable load anomalies and line fault alarms.

[0044] S3: Based on the impact data, a clustering algorithm is used to integrate the line sections into multiple types of sub-sections. Each type of sub-section is evaluated to obtain the evaluation level of each line section under the sub-section.

[0045] This embodiment uses the K-means clustering algorithm to cluster the impact data. Before clustering, the data is first standardized, and then the number of clusters is set. The number of clusters can be determined using the elbow rule. For example, after clustering is completed, the line section is divided into 10 types of sub-segments. For example, a sub-segment of a certain type includes 20 line sections. Since the impact data of the 20 line sections is similar, rating one of the line sections can obtain the evaluation grade of the remaining 19 line sections. The evaluation grade can be manually performed, for example, by setting a scoring table for each data type, scoring the temperature, wind speed, and load according to the scoring table, and finally obtaining a total score. The corresponding evaluation grade is determined based on the numerical range of the total score.

[0046] S4: Based on the evaluation level, some line sections are selected as complex sections, the complex sections are scanned to generate three-dimensional point cloud data, and a three-dimensional model is constructed based on the three-dimensional point cloud data.

[0047] A higher evaluation level indicates a more complex and variable environment within a route section. Therefore, route sections with higher evaluation levels are selected as complex sections. For example, on a scale of 1 to 10, route sections with a level of 9 or higher can be considered complex. This approach allows a small number of route sections to be identified as complex, reducing subsequent optimization costs. Point cloud data can be generated using drones or 3D laser scanners, and then converted into a 3D model using CloudCompare software.

[0048] S5: Analyze the three-dimensional model to obtain alternative installation positions of multiple cameras, and generate an optimization solution for optimizing the positions of scene cameras and local cameras based on the initial installation position and the alternative installation positions.

[0049] After establishing the 3D model, first determine whether the current camera can meet the monitoring requirements at the initial installation position. For example, whether there are blind spots in the monitoring area of ​​the scene camera, whether only one scene camera can meet the monitoring needs of the area, and whether the local camera is easily blocked during monitoring. If the above problems do not exist, there is no need to optimize the camera position.

[0050] If there are monitoring issues, multiple alternative installation locations are identified based on the 3D model. Cameras are then placed in these locations for testing to determine if the camera's coverage meets the requirements, whether blind spots are covered, or whether local cameras can capture the intended key areas. After testing, the locations and monitoring angles that meet the requirements are selected as the optimal placement, completing the camera placement optimization process.

[0051] This invention divides the deployment area into multiple line segments, collects environmental and alarm data, and then performs cluster analysis based on this data. This allows line segments to be classified by similar characteristics, facilitating subsequent hierarchical segmentation, effectively screening out complex segments. Subsequent optimization targets only these complex segments, reducing unnecessary monitoring overhead and optimizing resource allocation. By constructing a 3D model based on 3D point cloud data for complex segments, a clear understanding of the topography of different areas can be achieved. Based on this, installation locations can be optimized to address existing issues within the monitored area, maximizing the monitoring effectiveness of scene cameras and local cameras and improving the safety monitoring capabilities of the entire line.

[0052] In this embodiment, analyzing the three-dimensional model to obtain multiple candidate installation positions for cameras includes the following steps:

[0053] A plurality of monitoring targets are marked in a three-dimensional model, and the three-dimensional model is gridded to obtain a plurality of spatial grids. The spatial grids are divided into line grids and non-line grids according to whether the spatial grids are components of the monitoring targets. A plurality of first values ​​are set based on parameters of a scene camera. A generation point is set in the three-dimensional model. A plurality of spherical areas are generated with the generation point as the center and the first value as the radius. A plurality of monitoring points are set at the boundaries of the spherical areas. The center point of each line grid is located. The monitoring points are connected to the center point to obtain a visible line. If the visible line passes through at least one line grid, the visible line is deleted.

[0054] The monitoring targets include cables, cable racks and other equipment belonging to the cable structure. The three-dimensional model is then gridded and divided into multiple cubes, namely spatial grids. The division size can be set according to actual needs, such as 0.5m*0.5m*0.5m. If the spatial grid belongs to the monitoring target, it is defined as a line network. The shooting range of the camera is determined according to the parameters of the scene camera. For example, if the maximum shooting distance of the scene camera is 30 meters, the first value can be set within a range less than 30 meters, such as 20 meters, 25 meters and 30 meters. By setting multiple spherical ranges, you can try to set monitoring points at different distances from the cable. For example, there is an obstacle at 29 meters. If the camera is set at 30 meters, it will be blocked by the obstacle. If it is set at 25 meters, it will not be blocked by the obstacle.

[0055] Multiple points are selected as generation points according to the length of the line section. For example, a point is selected on the cable as a generation point every 40 meters in the line section. Three spherical areas are generated based on the generation point. The radius of each spherical area is 20 meters, 25 meters and 30 meters respectively. Then, multiple monitoring points are set in each spherical area.

[0056] The following describes the process of retaining and deleting visible lines, such as Figure 2As shown, for a certain spherical area, it is assumed that monitoring points D1 to D6 are arranged along its surface. For monitoring point D1, the center point of line grid A is Z1. The center point Z1 of line grid A is connected to monitoring point D1 to obtain visible line L1. Line grid B is located directly behind line grid A. The center point Z2 of line grid B is connected to monitoring point D1 to obtain visible line L2 (not shown in the figure). Since line grid A is directly in front of line grid B, if the center point Z2 is connected to monitoring point D1 to obtain visible line L2, then visible line L2 will inevitably pass through line grid A, and this visible line is deleted. The present invention determines the occlusion relationship according to the method of drawing visible lines. Through this method, it can be determined that line grid A can be seen from monitoring point D1 because visible line L1 does not pass through any line grids. However, when looking from D1 to monitoring point B, visible line L2 passes through line grid A, indicating that line grid A blocks line grid B. Therefore, line grid B cannot be fully seen from point D1.

[0057] Count the total number of visible lines at each monitoring point, delete the monitoring points whose total number is less than the second value, and obtain the occlusion rate of the camera at the initial installation position. If the occlusion rate is greater than the first threshold, the initial installation position is eliminated, and the retained monitoring points and the initial installation position that are not eliminated are used as alternative installation positions.

[0058] Then count the total number of visible lines at each monitoring point. If the total number decreases, it means that after the scene camera is placed at this position, no matter how its angle is adjusted, a good monitoring effect cannot be achieved, so it is deleted.

[0059] For the initial installation location where a camera has already been installed, the probability of the camera screen being obstructed by an object is calculated based on historical monitoring videos. The obstruction rate can be set as follows: a base value is set based on a predetermined duration, for example, 100. The number of times the video screen is obstructed by an object, for example, 20, is counted. The ratio of the number of obstructions to the base value is used as the obstruction rate, for example, 20 / 100 = 0.2. A first threshold is set to 0.1. When the obstruction rate exceeds the first threshold, the initial installation location is eliminated, indicating that the camera is likely to be obstructed by an object if installed there.

[0060] In this embodiment, the optimization scheme for generating a scene camera based on the alternative installation positions includes the following steps:

[0061] Set the starting value, which is the default number of scene cameras to be installed. Based on the starting value, traverse and combine the alternative installation positions to obtain multiple basic installation solutions.

[0062] A basic installation scheme is selected, and the basic installation scheme selected is simulated based on the three-dimensional model. Based on the simulation results, the effective shooting values ​​of each scene camera under the basic installation scheme are obtained. The higher the effective shooting value, the more necessary it is to install the scene camera. If the distribution of the effective shooting values ​​of the basic installation scheme meets the first condition, the number of scene cameras is reduced. If the distribution of the effective shooting values ​​meets the second condition, the number of scene cameras is increased. This step is repeated until the distribution of the effective shooting values ​​no longer meets the first and second conditions. The basic installation scheme at this time is defined as an advanced scheme.

[0063] The starting value is determined specifically based on the length of the line section and the parameters of the scene camera. In this embodiment, the starting value is set to 20, and then the alternative installation positions are traversed and combined to obtain multiple basic installation schemes, each of which includes 20 scene cameras.

[0064] The rationality of each basic installation plan is then evaluated. During this evaluation, corresponding scene cameras are set up in the 3D model based on the alternative installation locations selected in the basic installation plan. The scene cameras' shooting angles are oriented toward the center of the spherical area. The monitoring range is determined based on the scene camera parameters, and the effective shooting value of each monitoring camera is then determined based on the monitoring range. A higher effective shooting value indicates a greater need for installing that scene camera. If the effective shooting value distribution of the 20 scene cameras in the basic installation plan meets the first condition, such as the presence of a large number of scene cameras with low effective shooting values, the number of scene cameras in the basic installation plan is reduced. If the effective shooting value distribution meets the second condition, such as the presence of a large number of scene cameras with high effective shooting values, the number of scene cameras is increased. By continuously increasing or decreasing the number of scene cameras, the rationality of the basic installation plan is continuously improved.

[0065] Based on each basic installation solution, a corresponding advanced solution is obtained, and based on the installation cost and area coverage of each advanced solution, an evaluation value is obtained. Based on the evaluation value, an advanced solution is selected as the optimization solution for the scene camera.

[0066] After obtaining a corresponding advanced solution based on each basic installation plan, each advanced solution is evaluated to obtain a corresponding evaluation value. For example, the installation cost is determined based on the number of scene cameras in the advanced solution and their distance from the hub, and the area coverage is determined based on the number of line grids covered. For example, if there are 100 line grids in the area and the advanced solution can monitor 95 of them, the area coverage is 95 / 100 = 0.95. The evaluation value can be obtained by taking a weighted sum of the installation cost and the area coverage. In particular, when the weighted sum is taken, the installation cost is a negative number, so the higher the calculated evaluation value, the more reasonable the advanced solution.

[0067] Specifically, in this embodiment, obtaining the effective shooting values ​​of the cameras in each scene under the basic installation scheme based on the simulation results includes the following steps:

[0068] Based on the degree of importance, a different monitoring level is set for the surface of each line grid. Based on the size of the monitoring level, the surface of the corresponding line grid is divided into a corresponding number of monitoring areas. The shooting space of each scene camera is determined based on the camera parameters of the scene camera. The first number of monitoring areas included in the shooting space is located. The same monitoring areas in the shooting space of each scene camera are merged, and the second number of redundant monitoring areas included in each scene camera after the merger is recorded. The difference between the first number and the second number is used as the effective shooting value of the corresponding scene camera.

[0069] For example, for the line grid of the cable bracket, side 1 of its six sides can reflect the cable installation situation. Therefore, side 1 is set to monitoring level 1, and the remaining sides 2-6 are set to monitoring level 2. Side 1 is divided into two monitoring areas, and the remaining sides remain as one monitoring area.

[0070] Each monitoring area is then numbered, and the number of numbered monitoring areas that each scene camera can monitor is counted. For example, scene camera 1 is in alternative installation location 1, and its conical shooting space can monitor 100 monitoring areas. The following example illustrates how to calculate the effective shooting value. Assuming that scene camera 1 can monitor 100 monitoring areas, its first number is 100. Similarly, scene camera 2 can monitor 95 monitoring areas, with a first number of 95. Scene camera 1 can monitor 90 monitoring areas, with a first number of 90. Of the monitoring areas captured by scene cameras 1, 2, and 3, 50 are identical. Of these 50 monitoring areas, 40 also exist in the shooting space of scene camera 1. This means that 40 of the monitoring areas captured by scene camera 1 are also captured by scene cameras 2 or 3. Therefore, 40 monitoring areas of scene camera 1 are redundant, resulting in a valid shooting value of 100 - 40 = 60.

[0071] In addition to the above, this embodiment further corrects the effective shooting value based on the following steps, setting the optimal monitoring direction for each monitoring target. As described above, the surface of each line grid has a monitoring level. In the monitoring area covered by the scene camera, the monitoring area with a monitoring level greater than the fourth threshold is screened as the target area. Because the monitoring area with a higher level is more important, the optimal monitoring direction must also be oriented towards the covered monitoring area, and the optimal monitoring direction of the monitoring target is used as the optimal monitoring direction of the target area. In the shooting space of the scene camera, the optimal monitoring directions including each target area are vector-merged, thereby merging multiple optimal monitoring directions into one vector direction, obtaining the angle between the current shooting direction of the scene camera and the vector direction, determining the correction weight based on the size of the angle, and multiplying the correction weight by the calculated effective shooting value to complete the correction. For example, the angle can be cosine-calculated to obtain the correction weight.

[0072] In this embodiment, when reducing scene cameras, if in the basic installation plan, the number of scene cameras with effective shooting values ​​lower than the second threshold is greater than the first value, the first condition is met, and the scene cameras with the smallest effective shooting value are eliminated in sequence until there are greater than or equal to the third number of monitoring areas that are not included in the shooting space after elimination.

[0073] Specifically, the second threshold is set to 30, and the first value is set to 5. If, in the basic installation plan, there are five or more scene cameras with effective capture values ​​less than 30, they are removed in ascending order of their effective capture values. For example, if two scene cameras have effective capture values ​​of 20 and 30, respectively, the first scene camera is removed first. Furthermore, the third value is set to 40. If, after removing a scene camera, at least 40 monitoring areas in the route section are not included in the monitoring space of any scene camera, no further cameras are removed.

[0074] In this embodiment, increasing the number of scene cameras includes the following steps:

[0075] If in the basic installation plan, the effective shooting value is greater than the third threshold and the scene camera is less than the second value, and there are more than the third number of monitoring areas included in the shooting space, then the second condition is met, and an empty alternative installation position is selected in the three-dimensional model, and a scene camera is added in the selected alternative installation position. If the effective shooting value of the added scene camera is less than the second threshold, other alternative installation positions are selected to add scene cameras, and the alternative installation position is no longer used as the position for adding scene cameras in this basic installation plan.

[0076] If a large number of scene cameras have large effective shooting values ​​and there are many monitoring areas that are not monitored, it means that more scene cameras can be added to improve the monitoring range. At this time, select an alternative installation location, add a scene camera at the alternative installation location, and calculate the effective shooting value of the scene camera; if the effective shooting value is small, it means that under the layout of the current basic installation plan, it is not suitable to add a camera at this location. At this time, select other alternative installation locations to add cameras. When the basic installation plan is optimized later, no scene camera will be added here.

[0077] In this embodiment, generating an optimization solution for a local camera based on the candidate installation positions includes the following steps:

[0078] Obtain the occlusion rate of the current local camera. If the occlusion rate is greater than a first threshold, select an alternative installation position as an alternative position, set the local camera at the alternative position, and determine its shooting angle so that the local camera continues to shoot the original area.

[0079] If the occlusion rate of the local camera is greater than the first threshold, an alternative installation location that is not blocked by existing obstacles is selected as an alternative location, and the local camera is installed at the alternative location, while adjusting its shooting angle so that the camera can continue to monitor the original area or its nearby area.

[0080] In addition, the present invention also provides a video acquisition equipment layout optimization system for line monitoring, which is used to implement the above-mentioned method, and includes:

[0081] An input module determines the layout area of ​​the line. The input module is used to generate an initial installation plan for cameras in the layout area. The initial installation plan includes initial installation positions of scene cameras and local cameras.

[0082] A division module installs cameras based on the initial installation plan. The division module divides the deployment area into multiple line sections based on line length and terrain data. After a predetermined time interval, the module obtains impact data for each line section, including environmental data and line alarm data.

[0083] The evaluation module, based on the impact data, uses a clustering algorithm to integrate the line sections into multiple types of sub-sections, evaluates each type of sub-section, and obtains the evaluation level of each line section under the sub-section;

[0084] A construction module selects some line sections as complex sections based on the evaluation level, scans the complex sections to generate 3D point cloud data, and constructs a 3D model based on the 3D point cloud data;

[0085] The generation module is used to analyze the three-dimensional model to obtain alternative installation positions of multiple cameras, and generate an optimization plan for optimizing the positions of scene cameras and local cameras based on the initial installation position and the alternative installation positions.

[0086] The present application also provides a computer-readable storage medium having instructions stored thereon, and the instructions, when executed by a processor, implement the method described above.

[0087] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0088] The above embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

[0089] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for optimizing the layout of video acquisition equipment for line monitoring, characterized in that: include: Determine the layout area of ​​the line and generate an initial installation plan for the cameras in the layout area. The initial installation plan includes the initial installation positions of the scene cameras and local cameras. Install cameras based on the initial installation plan, divide the deployment area into multiple line sections based on line length and terrain data, and obtain impact data for each line section after a predetermined time interval, including environmental data and line alarm data; Based on the impact data, a clustering algorithm is used to integrate the line sections into multiple types of sub-sections. Each type of sub-section is evaluated to obtain the evaluation level of each line section under the sub-section. Based on the evaluation level, some line sections are selected as complex sections, the complex sections are scanned to generate 3D point cloud data, and a 3D model is constructed based on the 3D point cloud data; Analyze the three-dimensional model to obtain multiple alternative installation positions for cameras, and generate an optimization solution for optimizing the positions of scene cameras and local cameras based on the initial installation position and the alternative installation positions; Analyzing the three-dimensional model to obtain candidate installation positions for multiple cameras includes the following steps: A plurality of monitoring targets are marked in a three-dimensional model, and the three-dimensional model is gridded to obtain a plurality of spatial grids. The spatial grids are divided into line grids and non-line grids based on whether the spatial grids are components of the monitoring targets. A plurality of first values ​​are set based on parameters of a scene camera. A generation point is set in the three-dimensional model, and a plurality of spherical areas are generated with the generation point as the center and the first value as the radius. A plurality of monitoring points are set at the boundaries of the spherical areas. The center point of each line grid is located, and the monitoring points are connected to the center point to obtain a visible line. If the visible line passes through at least one line grid, the visible line is deleted. Count the total number of visible lines at each monitoring point, delete the monitoring points whose total number is less than the second value, and obtain the occlusion rate of the camera at the initial installation position. If the occlusion rate is greater than the first threshold, the initial installation position is eliminated, and the retained monitoring points and the initial installation position that are not eliminated are used as alternative installation positions.

2. The video acquisition equipment layout optimization method for line monitoring according to claim 1 is characterized in that: Generating an optimized solution for scene cameras based on alternative installation positions includes the following steps: Set the starting value, which is the default number of scene cameras to be installed. Based on the starting value, traverse and combine the alternative installation positions to obtain multiple basic installation solutions. A basic installation scheme is selected, and the selected basic installation scheme is simulated based on the 3D model. Based on the simulation results, the effective shooting value of each scene camera under the basic installation scheme is obtained. The higher the effective shooting value, the more necessary the scene camera is to be installed. If the distribution of the effective shooting values ​​of the basic installation scheme meets the first condition, the number of scene cameras is reduced. If the distribution of the effective shooting values ​​meets the second condition, the number of scene cameras is increased. This step is repeated until the distribution of the effective shooting values ​​no longer meets the first and second conditions. The basic installation scheme at this time is defined as an advanced scheme. Based on each basic installation solution, a corresponding advanced solution is obtained, and based on the installation cost and area coverage of each advanced solution, an evaluation value is obtained. Based on the evaluation value, an advanced solution is selected as the optimization solution for the scene camera.

3. The method for optimizing the layout of video acquisition equipment for line monitoring according to claim 2, characterized in that: Based on the simulation results, obtaining the effective shooting values ​​of the cameras in each scene under the basic installation scheme includes the following steps: Based on the degree of importance, a different monitoring level is set for the surface of each line grid. Based on the size of the monitoring level, the surface of the corresponding line grid is divided into a corresponding number of monitoring areas. The shooting space of each scene camera is determined based on the camera parameters of the scene camera. The first number of monitoring areas included in the shooting space is located. The same monitoring areas in the shooting space of each scene camera are merged, and the second number of redundant monitoring areas included in each scene camera after the merger is recorded. The difference between the first number and the second number is used as the effective shooting value of the corresponding scene camera.

4. The method for optimizing the layout of video acquisition equipment for line monitoring according to claim 3, characterized in that: Reducing the number of scene cameras involves the following steps: If in the basic installation plan, the number of scene cameras with effective shooting values ​​lower than the second threshold is greater than the first value, the first condition is met, and the scene cameras with the smallest effective shooting value are eliminated in sequence until there are greater than or equal to the third number of monitoring areas that are not included in the shooting space after elimination.

5. The method for optimizing the layout of video acquisition equipment for line monitoring according to claim 4, characterized in that: Increasing the number of scene cameras involves the following steps: If in the basic installation plan, the effective shooting value is greater than the third threshold and the scene camera is less than the second value, and there are more than the third number of monitoring areas included in the shooting space, then the second condition is met, and an empty alternative installation position is selected in the three-dimensional model, and a scene camera is added in the selected alternative installation position. If the effective shooting value of the added scene camera is less than the second threshold, other alternative installation positions are selected to add scene cameras, and the alternative installation position is no longer used as the position for adding scene cameras in this basic installation plan.

6. The method for optimizing the layout of video acquisition equipment for line monitoring according to claim 1, characterized in that: Generating an optimization plan for a local camera based on alternative installation positions includes the following steps: Obtain the occlusion rate of the current local camera. If the occlusion rate is greater than a first threshold, select an alternative installation position as an alternative position, set the local camera at the alternative position, and determine its shooting angle so that the local camera continues to shoot the original area.

7. A video acquisition equipment layout optimization system for line monitoring, used to implement the video acquisition equipment layout optimization method for line monitoring according to any one of claims 1 to 6, characterized in that: The system includes: An input module determines the layout area of ​​the line. The input module is used to generate an initial installation plan for cameras in the layout area. The initial installation plan includes initial installation positions of scene cameras and local cameras. A division module installs cameras based on the initial installation plan. The division module divides the deployment area into multiple line sections based on line length and terrain data. After a predetermined time interval, the module obtains impact data for each line section, including environmental data and line alarm data. The evaluation module, based on the impact data, uses a clustering algorithm to integrate the line sections into multiple types of sub-sections, evaluates each type of sub-section, and obtains the evaluation level of each line section under the sub-section; A construction module selects some line sections as complex sections based on the evaluation level, scans the complex sections to generate 3D point cloud data, and constructs a 3D model based on the 3D point cloud data; A generation module is used to analyze the three-dimensional model to obtain alternative installation positions of multiple cameras, and generate an optimization solution for optimizing the positions of scene cameras and local cameras based on the initial installation position and the alternative installation positions; Analyzing the three-dimensional model to obtain candidate installation positions for multiple cameras includes the following steps: A plurality of monitoring targets are marked in a three-dimensional model, and the three-dimensional model is gridded to obtain a plurality of spatial grids. The spatial grids are divided into line grids and non-line grids based on whether the spatial grids are components of the monitoring targets. A plurality of first values ​​are set based on parameters of a scene camera. A generation point is set in the three-dimensional model, and a plurality of spherical areas are generated with the generation point as the center and the first value as the radius. A plurality of monitoring points are set at the boundaries of the spherical areas. The center point of each line grid is located, and the monitoring points are connected to the center point to obtain a visible line. If the visible line passes through at least one line grid, the visible line is deleted. Count the total number of visible lines at each monitoring point, delete the monitoring points whose total number is less than the second value, and obtain the occlusion rate of the camera at the initial installation position. If the occlusion rate is greater than the first threshold, the initial installation position is eliminated, and the retained monitoring points and the initial installation position that are not eliminated are used as alternative installation positions.

8. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the video acquisition equipment layout optimization method for line monitoring according to any one of claims 1 to 6 is implemented.

Images