Rail transit protection area division method and system
Through the camera and drone collaboratively obtaining feature point coordinates, combined with GIS maps, the boundaries of rail transit protection areas are depicted in real time, solving the problems of low monitoring efficiency and unclear boundaries in the existing technology, and improving the safety of urban rail transit patrol and efficiency.
Patent Information
- Application Number
- CN202410950843.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-16
- Publication Date
- 2024-11-12
AI Technical Summary
In the prior art, video surveillance in rail transit protection areas relies on naked eye positioning, resulting in low surveillance and inaccurate monitoring, and the inability to clearly define protection boundaries, resulting in safety hazards, construction misjudgment and disputes, affecting incident investigation and resolution.
Use the camera to obtain real scene maps, select feature points to obtain two-dimensional pixel coordinates, use drones to obtain three-dimensional spatial coordinates, build transformation relationships, combine GIS maps to determine the boundaries of the protected area, draw protection edges in real time, and clearly divide the boundaries of the protected area.
It improves the efficiency of video surveillance in urban rail transit inspections, reduces construction misjudgment and misjudgment, ensures the timeliness of safety responses, clearly divides the boundaries of the protection zones, and reduces safety hazards.
Smart Images

Figure CN118941987A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of traffic protection zone division, and in particular relates to a rail transit protection zone division method and system. Background Art
[0002] The rail transit protection zone generally includes the top surface of the vehicle tunnel and the substation, transmission or ventilation facilities on the surface layer. The video surveillance technology in the rail transit protection zone plays a vital role in the safety of the urban rail transit system. This type of technology can detect safety incidents in a timely manner through real-time monitoring, so that rail maintenance personnel can respond.
[0003] In the existing technology, monitoring personnel mainly rely on the naked eye to locate the protection boundary and identify key areas, and the monitoring efficiency is low and inaccurate. If it is not possible to divide the rail transit into clear protection boundaries based on monitoring images, there will be insufficient monitoring of the protection areas without clear signs, making it impossible to detect security loopholes in time, posing a safety hazard; in addition, the failure to divide clear protection boundaries may also lead to misjudgment or omission of construction around the track, causing misunderstandings and disputes, and affecting the investigation and resolution of the incident. For example, on January 22, 2021, a sudden limit violation occurred in a certain city’s subway. When a construction unit was conducting on-site surveys and construction, due to misjudgment of the construction protection boundary, a 9-centimeter-diameter drill penetrated the subway tunnel and penetrated 98 centimeters into the tunnel, colliding with the train that was in operation at the time, causing the subway to be closed for a short time, with direct economic losses of 3.0665 million yuan. Summary of the invention
[0004] The present invention provides a rail transit protection zone division method and system, which are used to solve the problems existing in the prior art.
[0005] In a first aspect, in order to achieve the above-mentioned object, the present invention provides a method for dividing a rail transit protection zone, which comprises:
[0006] S1: Use a camera to obtain a real scene image;
[0007] S2: Select multiple feature points of the real scene image and obtain the two-dimensional pixel coordinates of each feature point;
[0008] S3: Use drone to obtain the three-dimensional spatial coordinates of feature points;
[0009] S4: construct the transformation relationship between the two-dimensional pixel coordinates, the three-dimensional space coordinates and the three-dimensional space coordinates of the camera photography center S;
[0010] S5: Determine the division area where the real scene image is located in the GIS map according to the device space coordinates and shooting angle of the camera;
[0011] S6: according to the transformation relationship, the boundary coordinates of the subway protection zone of the divided area are converted into boundary pixel coordinates;
[0012] S7: Place the boundary pixel coordinates into the real scene image, and draw the protection boundary line in the real scene image in real time.
[0013] Through the above technical scheme, the present invention proposes a method for demarcating a rail transit protection zone. The method can quickly locate the feature points of the real-scene image based on the real-scene image obtained by the camera on the surface, and depict the protection boundary in the real-scene image in combination with the GIS map and the unique transformation relationship of the present invention, thereby clearly demarcating the protection boundary, reducing the possibility of misjudgment or omission of construction around the track due to unclear protection boundaries, and improving the efficiency of video surveillance in urban rail transit inspections.
[0014] Preferably, the transformation relationship of the present invention is:
[0015]
[0016] Among them, in the ground photogrammetry coordinate system D-XYZ, the three-dimensional spatial coordinates of the feature point are (X, Y, Z), and the three-dimensional spatial coordinates of the camera photography center S are (X S , Y S , Z S ); in formulas (1)-(2), the x and y coordinates are the x and y coordinates in the image plane, and f is the principal distance of the camera; a i 、b i 、c i (i=1, 2, 3) is the direction cosine, the two-dimensional pixel coordinates of the feature point of the real scene photo are (u, v), and the two-dimensional pixel coordinates are used as the pixel coordinates of the feature point, (u 0 , v 0 ) is the two-dimensional pixel origin coordinate of the origin of the image plane coordinate system in the pixel coordinate system, and dx and dy are the physical sizes of a pixel in the image taken by the real-life photo.
[0017] Preferably, step S4 comprises:
[0018] Three feature points are randomly selected. After the camera captures the real scene, the three-dimensional spatial coordinates of the three feature points, the three-dimensional spatial coordinates of the camera photography center S, and the two-dimensional pixel coordinates of the three feature points are substituted into formulas (1)-(4) to obtain a i 、b i 、c i (i=1, 2, 3).
[0019] Preferably, before step S1, the method of the present invention further comprises:
[0020] Adjust the camera's intrinsic and extrinsic matrix parameters until they reach the reference range;
[0021] Among them, the intrinsic matrix parameters include focal length and optical center position, and the extrinsic matrix parameters include device space coordinates.
[0022] Preferably, the multiple feature points of the real scene image are selected including:
[0023] S21: using a feature detection algorithm to preliminarily select detection points of the real scene image;
[0024] S22: Using consistency check, repeatability check or feature point descriptor method to screen the detection points and obtain multiple feature points.
[0025] Preferably, the feature detection algorithm is SIFT, SURF or ORB.
[0026] Preferably, step S3 includes:
[0027] Use a drone to fly to the feature point to obtain the three-dimensional spatial coordinates of the feature point;
[0028] Among them, feature points include building corners, stone pier endpoints and telephone pole vertices.
[0029] In a second aspect, in order to achieve the above-mentioned object, the present invention provides a rail transit protection zone division system, which comprises:
[0030] Camera, used to obtain real-life images;
[0031] A two-dimensional pixel coordinate acquisition module is used to select multiple feature points of the real scene image and obtain the two-dimensional pixel coordinates of each feature point;
[0032] UAV, used to obtain the three-dimensional spatial coordinates of feature points;
[0033] A transformation relationship building module is used to build the transformation relationship between two-dimensional pixel coordinates, three-dimensional space coordinates and the three-dimensional space coordinates of the camera photography center S;
[0034] The real-view image positioning module is used to determine the division area where the real-view image is located in the GIS map according to the device space coordinates and shooting angle of the camera;
[0035] A coordinate conversion module, used for converting the boundary coordinates of the subway protection zone of the divided area into boundary pixel coordinates according to the transformation relationship;
[0036] The protection boundary line drawing module is used to place the boundary pixel coordinates into the real scene image and draw the protection boundary line in the real scene image in real time.
[0037] The present invention has at least the following advantages:
[0038] 1) Compared with the prior art, the present invention reduces the possibility of untimely safety response and misjudgment or omission of illegal operations due to unclear protection boundaries of real-view images during urban rail inspections, and improves the efficiency of video surveillance in urban rail transit inspections.
[0039] 2) Compared with the prior art, the technical solution of the present invention can be applied to the delineation of the protective boundary of the surface layer.
[0040] 3) Compared with the prior art, the present invention adopts a unique transformation relationship, and formulas (1)-(4) can more accurately express the spatial affine relationship of the three-dimensional spatial structure in the two-dimensional real scene plane. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0042] Figure 1 This is a flow chart of the method for dividing a rail transit protection zone in Embodiment 1 of the present invention.
[0043] Figure 2 This is a schematic diagram of drawing a protection boundary after the camera and the drone work together in embodiment 1 of the present invention.
[0044] Figure 3 It is a structural diagram of the rail transit protection zone division system in the second embodiment of the present invention. DETAILED DESCRIPTION
[0045] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0046] In the description of the present invention, it should be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc. indicate directions or positional relationships based on the directions or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, they cannot be understood as limitations on the present invention.
[0047] This embodiment discloses a rail transit protection zone division method and system, which are used to solve the problems existing in the prior art.
[0048] Embodiment 1
[0049] like Figure 1 As shown, a method for dividing a rail transit protection zone includes:
[0050] S1: Use a camera to obtain a real scene image.
[0051] In this embodiment, the camera is disposed on the ground, such as on a utility pole, an elevated bridge or a high-rise building.
[0052] Before step S1, the method of the present invention further comprises:
[0053] Adjust the camera's intrinsic and extrinsic matrix parameters until they reach the reference range.
[0054] It should be noted that the intrinsic matrix parameters include focal length and optical center position, while the extrinsic matrix parameters include device space coordinates (X S , Y S , Z S ). The camera parameter setting should ensure that the intrinsic and extrinsic matrix are accurate, at least within the reference range. The intrinsic matrix parameters are used to map the points of the camera coordinate system to the image coordinate system. If the intrinsic matrix is inaccurate, it will cause image distortion, feature point extraction errors, and increased spatial coordinate calculation errors. The extrinsic matrix parameters are used to describe the position and orientation of the camera in the world coordinate system. They are key parameters for mapping points in the world coordinate system to the camera coordinate system. If the extrinsic matrix is inaccurate, it will cause target positioning deviations and spatial coordinate calculation errors.
[0055] S2: Select multiple feature points of the real scene image and obtain the two-dimensional pixel coordinates of each feature point.
[0056] The multiple feature points of the real scene image selected in step S2 include:
[0057] S21: using a feature detection algorithm to preliminarily select detection points of the real scene image;
[0058] S22: Using consistency check, repeatability check or feature point descriptor method to screen the detection points and obtain multiple feature points.
[0059] In this embodiment, the above-mentioned feature detection algorithm can adopt SIFT, SURF or ORB algorithm. When selecting feature points, it is necessary to collect representative and stable feature points in the real scene image, wherein the visual uniqueness needs to be considered when selecting feature points, and the feature points should be unique in the real scene image, that is, they have a strong degree of distinction relative to other points in the surrounding environment. This can be evaluated by calculating the visual features such as the gradient or texture of the pixels around the feature points, and can be specifically evaluated by SIFT, SURF or ORB algorithm, and can be achieved by executing step S21. After the preliminary selection of the uniqueness of the feature points, it is also necessary to check the stability of the feature points (that is, to check whether a feature point can still be accurately detected or matched under different lighting conditions or viewing angles), which requires the execution of step S22 to be achieved. The feature point takes the upper left corner of the edge of the real scene image as the origin, the horizontal right is the positive direction of the u axis, the vertical downward is the positive direction of the v axis, and the two-dimensional pixel coordinates are (u, v).
[0060] Through the above steps S21-S22, the feature points selected from the real scene image are more representative and stable, which is conducive to improving the calculation accuracy of the subsequent model.
[0061] Preferably, for subway power supply transformers located on the surface, the above-mentioned characteristic points may be corner points of high-rise buildings, end points of stone piers, end points of boundary stakes and vertices of electric poles.
[0062] S3: Use drones to obtain the three-dimensional spatial coordinates of feature points.
[0063] Specifically, the above step S3 includes:
[0064] Use a drone to fly to the feature points and obtain the three-dimensional spatial coordinates of the feature points.
[0065] It should be noted that when using a drone to fly to the above-mentioned feature points, the flight environment and flight time of the drone need to be considered. In practical applications, after the drone flies to the feature points, it uploads its own three-dimensional space coordinates to the system of the present invention.
[0066] Preferably, the camera and the drone need to work together to remotely confirm that the three-dimensional spatial coordinates obtained by the drone are accurate.
[0067] In this embodiment, the three-dimensional space coordinates are (x, y, z), which can be the GPS coordinates of WGS-84 or the three-dimensional coordinates under the local coordinate system. Each feature point has a unique three-dimensional space coordinate, and the unit accuracy of the three-dimensional space coordinate is set according to actual conditions.
[0068] S4: Construct the transformation relationship between the two-dimensional pixel coordinates, the three-dimensional space coordinates and the three-dimensional space coordinates of the camera photography center S.
[0069] The above transformation relationship is:
[0070]
[0071]
[0072] It should be noted that if Figure 2 As shown, the above step S4 is based on the principle of photogrammetry, wherein, in the ground photogrammetry coordinate system D-XYZ, the three-dimensional spatial coordinates of the feature point are (X, Y, Z), and the three-dimensional spatial coordinates of the camera photography center S are (X S , Y S , Z S ); in formulas (1)-(2), the x and y coordinates are the x and y coordinates in the image plane, and f is the principal distance of the camera; a i 、b i 、c i (i=1, 2, 3) are direction cosines, and the two-dimensional pixel coordinates of the feature points are (u, v), (u 0 , v 0 ) is the two-dimensional pixel origin coordinate of the origin of the image plane coordinate system in the pixel coordinate system, and dx and dy are the physical dimensions of a pixel in the image captured by the real-life photo.
[0073] Specifically, the above step S4 includes:
[0074] Three feature points are randomly selected. After the camera takes a photo, the three-dimensional spatial coordinates of the three feature points, the three-dimensional spatial coordinates of the camera photography center S, and the two-dimensional pixel coordinates of the three feature points are substituted into formulas (1)-(4) to obtain a i 、b i 、c i (i=1, 2, 3).
[0075] It should be noted that each camera obtains different real-life images due to its different position and shooting direction, and the transformation parameters of the corresponding monitoring area finally calculated are different.
[0076] S5: Determine the division area where the real scene image is located in the GIS map according to the device space coordinates and the shooting angle of the camera.
[0077] It should be noted that the GIS map is a relatively rough satellite map, which is used to locate the actual location of the real-life image so that the staff can know which area of the city each real-life image is located in. The real-life image is a local surveillance image.
[0078] In this embodiment, the device space coordinates include longitude, latitude and height relative to the horizontal plane, and are unique coordinates of the world coordinate system. The shooting angle refers to the monitoring direction of the camera.
[0079] Specifically, the above step S5 is used to determine the geographical location of the real scene image in GIS. According to the spatial coordinates and shooting angle of the camera, it is known in which area of the city the real scene image is located, and it is preliminarily matched with the subway boundary range on the GIS map. Then, the three-dimensional spatial coordinates of the feature points measured by the drone are compared with the GIS map to realize the combination of the real scene image and the GIS map.
[0080] S6: According to the transformation relationship, the boundary coordinates of the subway protection zone of the divided area are converted into boundary pixel coordinates.
[0081] Specifically, when the subway protection boundary is found in the matching GIS map, several points with uniform intervals are selected along the boundary to obtain their boundary coordinates, and the coordinates of the points on the boundary are converted into pixel coordinates using formulas (1)-(4).
[0082] S7: Place the boundary pixel coordinates into the real scene image, and draw the protection boundary line in the real scene image in real time.
[0083] Through steps S1-S7, the present invention can draw protection boundaries in the real-life map of each section of rail transit, quickly locate feature points, and combine GIS maps and transformation relationships to depict the protection boundaries in the real-life map, thereby clearly dividing the protection boundaries, reducing the possibility of misjudgment or omission of construction around the track due to unclear protection boundaries, and improving the efficiency of video surveillance in urban rail transit inspections.
[0084] Embodiment 2
[0085] Reference Figure 3 This embodiment discloses a rail transit protection zone division system, which includes:
[0086] Camera, used to obtain real-life images;
[0087] A two-dimensional pixel coordinate acquisition module is used to select multiple feature points of the real scene image and obtain the two-dimensional pixel coordinates of each feature point;
[0088] UAV, used to obtain the three-dimensional spatial coordinates of feature points;
[0089] A transformation relationship building module is used to build the transformation relationship between two-dimensional pixel coordinates, three-dimensional space coordinates and the three-dimensional space coordinates of the camera photography center S;
[0090] The real-view image positioning module is used to determine the division area where the real-view image is located in the GIS map according to the device space coordinates and shooting angle of the camera;
[0091] A coordinate conversion module, used for converting the boundary coordinates of the subway protection zone of the divided area into boundary pixel coordinates according to the transformation relationship;
[0092] The protection boundary line drawing module is used to place the boundary pixel coordinates into the real scene image and draw the protection boundary line in the real scene image in real time.
[0093] It can be understood by those skilled in the art that all or part of the steps of the above-mentioned embodiments can be implemented by hardware related to program instructions. The above-mentioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiment are executed. The above-mentioned storage medium includes: ROM, RAM, magnetic disk or optical disk, etc., various media that can store program codes.
[0094] For those skilled in the art, various other corresponding changes and deformations can be made according to the technical solutions and concepts described above, and all of these changes and deformations should fall within the protection scope of the claims of the present invention.
Claims
1. A method for dividing a rail transit protection zone, characterized in that: include: S1: Use a camera to obtain a real scene image; S2: Select multiple feature points of the real scene image and obtain the two-dimensional pixel coordinates of each feature point; S3: using a drone to obtain the three-dimensional spatial coordinates of the feature points; S4: constructing a transformation relationship between the two-dimensional pixel coordinates, the three-dimensional space coordinates and the three-dimensional space coordinates of the camera photography center S; S5: Determine the division area where the real scene image is located in the GIS map according to the device space coordinates and shooting angle of the camera; S6: According to the transformation relationship, the boundary coordinates of the subway protection zone of the divided area are converted into boundary pixel coordinates; S7: placing the boundary pixel coordinates into the real scene image, and drawing a protection boundary line in the real scene image in real time.
2. The method for dividing a rail transit protection zone according to claim 1, characterized in that: The transformation relationship expression is: Among them, in the ground photogrammetry coordinate system D-XYZ, the three-dimensional spatial coordinates of the feature point are (X, Y, Z), and the three-dimensional spatial coordinates of the camera photography center S are (X s , Y S , Z S ); in formulas (1)-(2), the x and y coordinates are the x and y coordinates in the image plane, and f is the principal distance of the camera; a i , b i 、c i (i=1, 2, 3) are direction cosines, the two-dimensional pixel coordinates of the feature point are (u, v), (u0, v0) are the two-dimensional pixel origin coordinates of the origin of the image plane coordinate system in the pixel coordinate system, and dt and dy are the physical dimensions of a pixel in the captured image.
3. The method for dividing a rail transit protection zone according to claim 2, characterized in that: The step S4 comprises: Three feature points are randomly selected. After the camera captures the real scene, the three-dimensional spatial coordinates of the three feature points, the three-dimensional spatial coordinates of the camera photography center S, and the two-dimensional pixel coordinates of the three feature points are substituted into formulas (1)-(4) to obtain a i 、b i 、c i (i=1, 2, 3).
4. The rail transit protection zone division method according to claim 1 is characterized in that: Before step S1, the method further includes: Adjusting the intrinsic matrix parameters and the extrinsic matrix parameters of the camera until they reach a reference range; The intrinsic matrix parameters include focal length and optical center position, and the extrinsic matrix parameters include the device space coordinates.
5. The rail transit protection zone division method according to claim 1 is characterized in that: The selecting of the plurality of feature points of the real scene image comprises: S21: using a feature detection algorithm to preliminarily select detection points of the real scene image; S22: Using consistency check, repeatability check or feature point descriptor method to screen the detection points and obtain multiple feature points.
6. The method for dividing a rail transit protection zone according to claim 5, characterized in that: The feature detection algorithm is SIFT, SURF or ORB.
7. The method for dividing a rail transit protection zone according to claim 1, characterized in that: The step S3 comprises: Use a drone to fly to the feature point to obtain the three-dimensional spatial coordinates of the feature point; The characteristic points include building corners, stone pier endpoints and utility pole vertices.
8. A rail transit protection zone demarcation system, characterized in that: include: Camera, used to obtain real-life images; A two-dimensional pixel coordinate acquisition module is used to select multiple feature points of the real scene image and obtain the two-dimensional pixel coordinates of each feature point; A drone, used to obtain the three-dimensional spatial coordinates of the feature points; A transformation relationship building module, used to build a transformation relationship between the two-dimensional pixel coordinates, the three-dimensional space coordinates and the three-dimensional space coordinates of the camera photography center S; A real-view image positioning module is used to determine the division area where the real-view image is located in the GIS map according to the device space coordinates and shooting angle of the camera; A coordinate conversion module, used for converting the boundary coordinates of the subway protection zone of the divided area into boundary pixel coordinates according to the transformation relationship; The protection boundary line drawing module is used to place the boundary pixel coordinates into the real scene image and draw the protection boundary line in the real scene image in real time.
Citation Information
Cited By
Electric power gallery air hidden danger wide-area monitoring method based on low earth orbit satellite internet and related device
CN119382350A
