CN215642803U - Obstacle detection system for railway operation line - Google Patents
Obstacle detection system for railway operation line Download PDFInfo
- Publication number
- CN215642803U CN215642803U CN202122109898.4U CN202122109898U CN215642803U CN 215642803 U CN215642803 U CN 215642803U CN 202122109898 U CN202122109898 U CN 202122109898U CN 215642803 U CN215642803 U CN 215642803U
- Authority
- CN
- China
- Prior art keywords
- point cloud
- data
- unit
- voxel grid
- grid
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Train Traffic Observation, Control, And Security (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The present application provides an obstacle detection system for a railway line of travel, the system comprising: the system comprises a data acquisition unit, a data processing unit, a grid construction unit and a data judgment unit, wherein the data acquisition unit is used for acquiring continuous point cloud data, the data processing unit is used for processing the continuous point cloud data to obtain background point cloud data, the grid construction unit is used for constructing a voxel grid in a detection area, and the voxel grid corresponding to the background point cloud data is set to be occupied to obtain a voxel grid background model of the detection area; the data judgment unit is used for judging whether the voxel grid corresponding to the continuous point cloud data is occupied in the voxel grid background model or not and sending barrier alarm information. According to the method, the space voxel grid is constructed according to the continuous point cloud data through the three-dimensional laser radar, so that the railway operation line barrier is accurately measured, and the problems that the existing detection system is inaccurate in measurement result, poor in safety and easy to cause safety accidents are solved.
Description
Technical Field
The application relates to the technical field of laser radar detection, in particular to an obstacle detection system for a railway operation line.
Background
With the high-speed development of the transportation industry, railway transportation becomes one of the indispensable transportation modes in people's daily life, because the geographical environment along the railway route is comparatively complicated, has a large amount of mountain areas, and there are risks such as mountain rocks fall in some railway operation highway sections, consequently need carry out real-time detection to the operation highway section of railway transportation, avoids because the barrier blocks the railway, leads to taking place danger.
The traditional detection method is to use a two-dimensional laser radar to carry out real-time detection, because the two-dimensional laser radar can only scan a plane, although the detection of the barrier can be realized to a certain extent, due to the limitation of the data dimension of the two-dimensional laser radar, the three-dimensional information of the barrier can not be accurately detected and obtained, the state of the barrier can not be effectively judged, meanwhile, the barrier with too low height or hanging in the air can not be accurately detected, certain technical defects exist, and safety accidents are easily caused.
SUMMERY OF THE UTILITY MODEL
The application provides a barrier detecting system for a railway operation line, which aims to solve the problems that in the barrier detecting process of the existing railway operation line, the height of a barrier which is too low or hangs in the air cannot be accurately detected, so that the accuracy is lower and the safety is poorer.
The application provides an obstacle detection system for railway operation route, its characterized in that includes: the system comprises a data acquisition unit, a data processing unit, a grid construction unit and a data judgment unit;
the data acquisition unit is used for acquiring continuous point cloud data and sending the continuous point cloud data to the data processing unit; the output end of the data acquisition unit is connected with the input end of the data processing unit;
the data processing unit is used for processing the continuous point cloud data to obtain background point cloud data and sending the background point cloud data to the grid construction unit, and the output end of the data processing unit is connected with the input end of the grid construction unit;
the grid construction unit is used for constructing a voxel grid in a detection area, and setting the voxel grid corresponding to the background point cloud data as occupied to obtain a voxel grid background model of the detection area;
the data acquisition unit is also used for acquiring continuous point cloud data after the voxel grid background model is constructed, and sending the continuous point cloud data to the data judgment unit, wherein the output end of the data acquisition unit is connected with the input end of the data judgment unit;
the data judgment unit is used for receiving the continuous point cloud data from the data acquisition unit, judging whether a voxel grid corresponding to the continuous point cloud data is occupied in a voxel grid background model, and if not, sending obstacle alarm information.
The data processing unit comprises a point cloud judging module and a point cloud processing module;
the point cloud judging module is used for sequentially judging whether each point cloud in the continuous point cloud data is in the detection area to obtain a judgment result, and the output end of the point cloud judging module is connected with the input end of the point cloud processing module;
and the point cloud processing module is used for deleting or retaining the point cloud according to the judgment result.
The grid construction unit comprises a division step length setting unit, a voxel grid division unit and a voxel grid setting unit;
the dividing step length setting unit is used for setting the dividing step length of the voxel grid, and the output end of the dividing step length setting unit is connected with the input end of the voxel grid dividing unit;
the voxel grid dividing unit is used for dividing the detection area from any vertex P (P) according to a dividing stepx,Py,Pz) The voxel grid with the side length being the dividing step is set.
The voxel grid setting unit is used for setting the coordinates (x) of any point cloudt,yt,zt) And converting the coordinate into an index coordinate of a corresponding voxel grid, and setting the corresponding voxel grid to be occupied to obtain a voxel grid background model of the detection area.
The voxel grid setting unit is further configured to set the discontinuous unoccupied voxel grid as occupied.
When the obstacle detection system for the railway operation line provided by the application works normally, the obstacle detection system comprises the following steps:
acquiring continuous point cloud data;
processing the continuous point cloud data to obtain background point cloud data;
constructing a voxel grid in the detection area;
setting the voxel grid corresponding to the background point cloud data as occupied to obtain a voxel grid background model of a detection area;
after the voxel grid background model is constructed, continuous point cloud data is obtained;
judging whether the voxel grid corresponding to the continuous point cloud data is set to be occupied in a voxel grid background model;
and if the vehicle is set to be occupied, sending obstacle alarm information.
The processing of the continuous point cloud data to obtain background point cloud data specifically comprises the following steps:
sequentially judging whether each point cloud in the continuous point cloud data is in the detection area;
and if the point cloud is not in the detection area, deleting the point cloud.
And if the point cloud exists in the detection area, retaining the point cloud to obtain background point cloud data.
The method further comprises the steps of:
judging whether a discontinuous unoccupied voxel grid exists in the voxel grid background model;
if so, setting the discontinuous unoccupied voxel grid as occupied.
The method for constructing the voxel grid in the detection area specifically comprises the following steps:
setting a dividing step length of a voxel grid;
the detection area is divided from an arbitrary vertex P (P) according to the division stepx,Py,Pz) Setting a voxel grid with side length as a dividing step;
judging whether the final residual length is smaller than the step of division;
if the step length is smaller than the dividing step length, setting according to the actual residual length;
the index range for each voxel grid is calculated as follows:
the voxel grid index coordinate corresponding to the detection area is sequentially respectively Xindex ═ 1, x in the directions of three coordinate axesnum],Yindex=[1,ynum],Zindex=[1,znum]。
Before the voxel grid corresponding to the background point cloud data is set to be occupied, the method further comprises the step of setting the random point cloud coordinate (x)t,yt,zt) Converting into index coordinates (Xindex, yidex, Zindex) of the corresponding voxel grid;
the conversion formula is:
as can be seen from the above technical solutions, the present application provides an obstacle detection system for a railway operation line, the system including: the system comprises a data acquisition unit, a data processing unit, a grid construction unit and a data judgment unit, wherein the data acquisition unit is used for acquiring continuous point cloud data, the data processing unit is used for processing the continuous point cloud data to obtain background point cloud data, the grid construction unit is used for constructing a voxel grid in a detection area, and the voxel grid corresponding to the background point cloud data is set to be occupied to obtain a voxel grid background model of the detection area; the data judgment unit is used for judging whether the voxel grid corresponding to the continuous point cloud data is occupied in the voxel grid background model or not and sending barrier alarm information. According to the method, the space voxel grid is constructed according to the continuous point cloud data through the three-dimensional laser radar, so that the railway operation line barrier is accurately measured, and the problems that the existing detection system is inaccurate in measurement result, poor in safety and easy to cause safety accidents are solved.
Drawings
In order to more clearly explain the technical solution of the present application, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious to those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic diagram of an obstacle detection system for a railway line;
fig. 2 is a view of an application scenario of an obstacle detection system for a railway line according to the present application;
FIG. 3 is a schematic diagram of the data processing unit of the present application;
fig. 4 is a schematic structural diagram of the grid construction unit of the present application;
FIG. 5 is a flow chart of the operation of an obstacle detection system for a railroad track of the present application;
FIG. 6 is a flowchart of the operation of one embodiment of the present application.
The system comprises a data acquisition unit 1, a data processing unit 2, a point cloud judgment module 21, a point cloud processing module 22, a grid construction unit 3, a division step length setting unit 31, a voxel grid division unit 32, a voxel grid setting unit 33 and a data judgment unit 4.
Detailed Description
Reference will now be made in detail to embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following examples do not represent all embodiments consistent with the present application. But merely as exemplifications of systems and methods consistent with certain aspects of the application, as recited in the claims.
According to the application scene graph of the obstacle detection system for the railway running line, the three-dimensional laser radar scans the detection area to obtain point cloud data of the detection area, and the point cloud data comprises ground, railway facilities such as rails, railway ballasts and track plates. And accumulating and learning facility point clouds such as the ground, the rails, the railway ballasts, the track plates and the like into background point cloud data from the point cloud data by processing the point cloud data. The detection area is divided into space voxel grids, and a point cloud down-sampling mode is adopted to establish a voxel grid background model of an irregular curved surface area of a road surface where a railway is located. After the voxel grid background model is established, the point clouds in the monitored area can be distinguished by combining the real-time scanning data of the three-dimensional radar and the three-dimensional background information of the monitored area, the background point clouds are separated from the scanning point clouds to obtain the point clouds of the obstacles, and finally the detection of the obstacles is realized.
Referring to fig. 1, the present application provides an obstacle detection system for a railway line of travel, comprising: the system comprises a data acquisition unit 1, a data processing unit 2, a grid construction unit 3 and a data judgment unit 4;
the data acquisition unit 1 is used for acquiring continuous point cloud data and sending the continuous point cloud data to the data processing unit 2; the output end of the data acquisition unit 1 is connected with the input end of the data processing unit 2;
the data processing unit 2 is configured to process the continuous point cloud data to obtain background point cloud data, and send the background point cloud data to the grid construction unit 3, where an output end of the data processing unit 2 is connected to an input end of the grid construction unit 3;
the grid construction unit 3 is used for constructing a voxel grid in a detection area, and setting the voxel grid corresponding to the background point cloud data as occupied to obtain a voxel grid background model of the detection area;
it should be noted that the detection area may be any shape of convex geometry containing the railway line, and the height direction may be any, and is not lower than the height of the passing train. For example, referring to fig. 2, the detection area is a three-dimensional defense area a, and the cubic defense area a includes all rail facilities and the ground downwards and is higher than the height of the train upwards. The coordinate system of the three-dimensional radar scanning point cloud and the coordinate system of the three-dimensional defense area are taken as a reference, and the length, the width and the height of the cube are respectively L, W and H.
The data acquisition unit 1 is further configured to acquire continuous point cloud data after the voxel grid background model is constructed, and send the continuous point cloud data to the data judgment unit 4, and an output end of the data acquisition unit 1 is connected with an input end of the data judgment unit 4;
the data judgment unit 4 is configured to receive the continuous point cloud data from the data acquisition unit 1, judge whether a voxel grid corresponding to the continuous point cloud data is occupied in a voxel grid background model, and send obstacle alarm information if the voxel grid corresponding to the continuous point cloud data is not occupied.
In some embodiments, referring to fig. 3, the data processing unit 2 includes a point cloud determining module 21 and a point cloud processing module 22;
the point cloud judging module 21 is configured to sequentially judge whether each point cloud in the continuous point cloud data is in the detection area, so as to obtain a judgment result, and an output end of the point cloud judging module 21 is connected with an input end of the point cloud processing module 22;
the point cloud processing module 22 is configured to delete or retain the point cloud according to the determination result.
In some embodiments, referring to fig. 4, the mesh construction unit 3 includes a division step setting unit 31, a voxel mesh division unit 32, and a voxel mesh setting unit 33;
the dividing step setting unit 31 is configured to set a dividing step of a voxel grid, and an output end of the dividing step setting unit 31 is connected to an input end of the voxel grid dividing unit 32;
the voxel grid dividing unit 32 is used for dividing the detection area from an arbitrary vertex P (P) according to a division stepx,Py,Pz) The voxel grid with the side length being the dividing step is set.
The voxel grid setting unit 33 is used for setting the coordinates (x) of any point cloudt,yt,zt) And converting the coordinate into an index coordinate of a corresponding voxel grid, and setting the corresponding voxel grid to be occupied to obtain a voxel grid background model of the detection area.
In some embodiments, the voxel grid setting unit 33 is further configured to set the discontinuous unoccupied voxel grid to occupied.
In a second aspect, referring to fig. 5, the present application provides an obstacle detection method for a railway track, comprising the steps of:
s100: acquiring continuous point cloud data;
s110: processing the continuous point cloud data to obtain background point cloud data;
more specifically, the background point cloud data corresponds to infrastructure entities necessary in the railway operation line.
S120: constructing a voxel grid in the detection area;
s130: setting the voxel grid corresponding to the background point cloud data as occupied to obtain a voxel grid background model of a detection area;
it should be noted that, the construction of the voxel grid background model of the railway scene is completed, and the voxel grids in the detection area are divided into two types, one type is set as an occupied voxel grid, and the other type is set as an unoccupied voxel grid. The occupied voxel grid corresponds to the physical scene surface in the detection area and its subsurface foundation. The unoccupied voxel grid corresponds to the real area to be monitored in the detection area.
S140: after the voxel grid background model is constructed, continuous point cloud data is obtained;
s150: judging whether the voxel grid corresponding to the continuous point cloud data is set to be occupied in a voxel grid background model;
s160: and if the vehicle is not set to be occupied, sending obstacle alarm information.
After the voxel grid background model is built, point cloud data scanned by the laser radar can be judged and analyzed, and if the voxel grid corresponding to the point cloud data is not occupied, an obstacle exists in a monitoring area. Subsequent obstacle judgment, alarm and other modes can be carried out. It should be noted that the alarm information may be an acoustic alarm or an optical alarm, and may also be transmitted to a remote server through remote communication transmission.
In some embodiments, referring to fig. 6, the processing the continuous point cloud data to obtain background point cloud data specifically includes the following steps:
s200: sequentially judging whether each point cloud in the continuous point cloud data is in the detection area;
s210: and if the point cloud is not in the detection area, deleting the point cloud.
S220: and if the point cloud exists in the detection area, retaining the point cloud to obtain background point cloud data.
In some embodiments, the method further comprises the steps of:
judging whether a discontinuous unoccupied voxel grid exists in the voxel grid background model;
if so, setting the discontinuous unoccupied voxel grid as occupied.
It should be noted that in a detection scene of a real railway operation line, an actual railway, a railway ballast, a track slab and the ground form a continuously completed real physical scene, and the background is a monitoring area above the background and a ground foundation below the background. Due to the laser scanning characteristic, partial shielding exists in scanning point cloud of a real scene, and discontinuous holes of background features appear in a voxel grid of a detection area. And filling the voxel grids of the detection area by adopting an interpolation mode. And after filling is complete, setting the voxel grids corresponding to the ground foundation below the background surface voxel grid as occupied. The problem that detection results are inaccurate due to shielding when the laser radar acquires point cloud data can be effectively solved.
More specifically, the setting of the discontinuous unoccupied voxel grid as occupied specifically includes: and judging all the voxel grids divided in the detection area, sequentially judging the index coordinates in the x direction and the y direction without the voxel grids, arbitrarily setting the index coordinates as occupied voxel grid coordinates (xindex, yindex and zindex), and setting the voxel grid corresponding to the x index and the y index and with the z index smaller than the zindex as occupied, wherein the occupied voxel grid corresponds to the ground foundation in the detection area.
In some embodiments, the constructing a voxel grid in the detection region specifically includes the following steps:
setting a dividing step length of a voxel grid;
the detection area is divided from an arbitrary vertex P (P) according to the division stepx,Py,Pz) Setting a voxel grid with side length as a dividing step;
judging whether the final residual length is smaller than the step of division;
if the step length is smaller than the dividing step length, setting according to the actual residual length;
the index range for each voxel grid is calculated as follows:
the voxel grid index coordinate corresponding to the detection area is sequentially respectively Xindex ═ 1, x in the directions of three coordinate axesnum],Yindex=[1,ynum],Zindex=[1,znum]。
In some embodiments, before setting the voxel grid corresponding to the background point cloud data to be occupied, the method further comprises setting the arbitrary point cloud coordinates (x) to be occupiedt,yt,zt) Converting into index coordinates of the corresponding voxel grid;
the conversion formula is:
according to the technical scheme provided by the application, the application provides an obstacle detection system and method for a railway operation line, and the system comprises: the system comprises a data acquisition unit, a data processing unit, a grid construction unit and a data judgment unit, wherein the data acquisition unit is used for acquiring continuous point cloud data, the data processing unit is used for processing the continuous point cloud data to obtain background point cloud data, the grid construction unit is used for constructing a voxel grid in a detection area, and the voxel grid corresponding to the background point cloud data is set to be occupied to obtain a voxel grid background model of the detection area; the data judgment unit is used for judging whether the voxel grid corresponding to the continuous point cloud data is occupied in the voxel grid background model or not and sending barrier alarm information. According to the method, the space voxel grid is constructed according to the continuous point cloud data through the three-dimensional laser radar, so that the railway operation line barrier is accurately measured, and the problems that the existing detection system is inaccurate in measurement result, poor in safety and easy to cause safety accidents are solved.
Furthermore, according to the technical scheme, the three-dimensional laser radar scanning point cloud data are utilized, the complex surface of the railway scene can be comprehensively scanned, the three-dimensional monitoring area of the irregular surface boundary of the railway scene is established, the three-dimensional monitoring of the railway scene is realized, and the method and the device can be popularized to other monitoring applications similar to fixed scenes.
The embodiments provided in the present application are only a few examples of the general concept of the present application, and do not limit the scope of the present application. Any other embodiments extended according to the scheme of the present application without inventive efforts will be within the scope of protection of the present application for a person skilled in the art.
Claims (4)
1. An obstacle detection system for a railroad track, comprising: the system comprises a data acquisition unit (1), a data processing unit (2), a grid construction unit (3) and a data judgment unit (4);
the data acquisition unit (1) is used for acquiring continuous point cloud data and sending the continuous point cloud data to the data processing unit (2); the output end of the data acquisition unit (1) is connected with the input end of the data processing unit (2);
the data processing unit (2) is used for processing the continuous point cloud data to obtain background point cloud data, and sending the background point cloud data to the grid construction unit (3), wherein the output end of the data processing unit (2) is connected with the input end of the grid construction unit (3);
the grid construction unit (3) is used for constructing a voxel grid in a detection area, and setting the voxel grid corresponding to the background point cloud data as occupied to obtain a voxel grid background model of the detection area;
the data acquisition unit (1) is further used for acquiring continuous point cloud data after the voxel grid background model is constructed, and sending the continuous point cloud data to the data judgment unit (4), wherein the output end of the data acquisition unit (1) is connected with the input end of the data judgment unit (4);
the data judgment unit (4) is used for receiving the continuous point cloud data from the data acquisition unit (1), judging whether a voxel grid corresponding to the continuous point cloud data is occupied in a voxel grid background model, and if not, sending obstacle alarm information.
2. Obstacle detection system for a railway line according to claim 1, characterized in that said data processing unit (2) comprises a point cloud judging module (21) and a point cloud processing module (22);
the point cloud judging module (21) is used for sequentially judging whether each point cloud in the continuous point cloud data is in the detection area to obtain a judgment result, and the output end of the point cloud judging module (21) is connected with the input end of the point cloud processing module (22);
the point cloud processing module (22) is used for deleting or retaining the point cloud according to the judgment result.
3. The obstacle detection system for a railway running line according to claim 2, wherein the grid construction unit (3) includes a division step setting unit (31), a voxel grid division unit (32), and a voxel grid setting unit (33);
the dividing step length setting unit (31) is used for setting the dividing step length of the voxel grid, and the output end of the dividing step length setting unit (31) is connected with the input end of the voxel grid dividing unit (32);
the voxel grid dividing unit (32) is used for dividing the detection area from an arbitrary vertex P (P) according to a dividing stepx,Py,Pz) Setting a voxel grid with side length as a dividing step;
the voxel grid setting unit (33) is used for setting the coordinates (x) of any point cloudt,yt,zt) And converting the coordinate into an index coordinate of a corresponding voxel grid, and setting the corresponding voxel grid to be occupied to obtain a voxel grid background model of the detection area.
4. Obstacle detection system for a railway line according to claim 3, characterized in that the voxel grid setting unit (33) is further adapted to set the discontinuous unoccupied voxel grid as occupied.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202122109898.4U CN215642803U (en) | 2021-09-02 | 2021-09-02 | Obstacle detection system for railway operation line |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202122109898.4U CN215642803U (en) | 2021-09-02 | 2021-09-02 | Obstacle detection system for railway operation line |
Publications (1)
Publication Number | Publication Date |
---|---|
CN215642803U true CN215642803U (en) | 2022-01-25 |
Family
ID=79906185
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202122109898.4U Active CN215642803U (en) | 2021-09-02 | 2021-09-02 | Obstacle detection system for railway operation line |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN215642803U (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116796210A (en) * | 2023-08-25 | 2023-09-22 | 山东莱恩光电科技股份有限公司 | Barrier detection method based on laser radar |
-
2021
- 2021-09-02 CN CN202122109898.4U patent/CN215642803U/en active Active
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116796210A (en) * | 2023-08-25 | 2023-09-22 | 山东莱恩光电科技股份有限公司 | Barrier detection method based on laser radar |
CN116796210B (en) * | 2023-08-25 | 2023-11-28 | 山东莱恩光电科技股份有限公司 | Barrier detection method based on laser radar |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106887020A (en) | A kind of road vertical and horizontal section acquisition methods based on LiDAR point cloud | |
CN115761722A (en) | Obstacle detection system and method for railway operation line | |
WO2022141911A1 (en) | Roadside sensing unit-based method for quick recognition of dynamic target point cloud and point cloud segmentation | |
CN109489584B (en) | Tunnel clearance detection system and tunnel clearance identification method based on 3D technology | |
CN104143194A (en) | Point cloud partition method and device | |
CN110133669B (en) | Three-dimensional laser foreign matter intrusion monitoring method and system | |
CN105761305B (en) | A kind of vehicle traveling detection method and system | |
CN109828282A (en) | A kind of vehicle overall dimension automatic checkout system and method based on laser radar | |
CN108765974A (en) | A kind of traffic conditions monitoring device, monitoring method and system | |
CN114312877B (en) | Railway comprehensive inspection system | |
CN102819951A (en) | Road target detection system, road target detection method and road target detection device | |
CN215642803U (en) | Obstacle detection system for railway operation line | |
CN112945096B (en) | Tunnel disease monitoring system and method for high-speed train | |
CN114379598B (en) | Railway comprehensive inspection system | |
CN114379607B (en) | Comprehensive railway inspection method | |
JP2023078053A (en) | Device and method for developing mode with combination of gradient and velocity | |
CN115014224A (en) | LiDAR point cloud and oblique aerial image-based ground surface deformation monitoring method | |
CN102829759A (en) | Remote intelligent three-dimensional digital early warning method and system for subgrade surface settlement | |
CN115830805A (en) | Highway slope monitoring and early warning system and equipment based on digital twin | |
CN114295069A (en) | Side slope deformation monitoring method and system for unmanned aerial vehicle carrying three-dimensional laser scanner | |
CN110782628B (en) | Mountain landslide monitoring system based on Beidou system and monitoring method thereof | |
CN111881566B (en) | Landslide displacement detection method and device based on live-action simulation | |
CN115220035B (en) | Road void detection system and detection method | |
CN116884198A (en) | Intelligent digital twin management system for railway subgrade deformation monitoring and disaster early warning | |
CN204196909U (en) | A kind of quick, intelligent inspection vehicle in tunnel surface crack based on image procossing |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
GR01 | Patent grant | ||
GR01 | Patent grant |