Railway foreign matter detection device, detection system and detection method

Abstract

The application discloses a track foreign matter detection device, a detection system and a detection method, wherein the detection device comprises a three-dimensional laser radar module, which is used for real-time scanning of a monitoring area through laser; an image acquisition module, which is linked with the three-dimensional laser radar module, and is used for image acquisition of foreign matter when the three-dimensional laser radar module scans the foreign matter; and a data processing module, which is used for obtaining and storing information obtained by the three-dimensional laser radar module and the image acquisition module, comparing two-phase laser point cloud data obtained by the three-dimensional laser radar module before and after the foreign matter invades the track, and judging whether there is an influential foreign matter.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of monitoring, more particularly, to a track foreign matter detection device, a detection system and a detection method. BACKGROUND

[0002] The railway industry in China is in a key period of rapid development, and safe operation has always been the fundamental of the healthy development of the railway industry. The railway line has a large span and a long distance, and passes through many types of natural disasters such as strong winds, precipitation, and earthquakes. In particular, through special environments such as bridges and culverts, it is easy to cause foreign matter invasion events such as landslides, mudslides, and rolling stones, which seriously affect the normal running of trains. In view of the above problems, the railway still adopts the method of periodic manual inspection of the track, and installs video monitoring, sensors and other technical defense means in special areas, but due to the limitations of the technology itself and the application mode, the effect is not ideal in the aspect of track foreign matter monitoring, and it is difficult to ensure the accuracy and timeliness of track foreign matter monitoring. At present, the stress sensing technology is relatively mature in the monitoring of natural disasters, and is mostly deployed in mountainous areas, slope surfaces and culverts along the track, which can effectively monitor natural disasters such as landslides and mudslides near the track, but it is difficult to determine whether a foreign matter invasion event has occurred.

[0003] Therefore, how to accurately determine a foreign matter invasion event is a problem that needs to be solved urgently. SUMMARY

[0004] The purpose of the present application is to provide a track foreign matter detection device, a detection system and a detection method, which can accurately determine a foreign matter invasion event.

[0005] In order to achieve the above purpose, the present application provides a track foreign matter detection device, which comprises:

[0006] A three-dimensional laser radar module is used to scan a monitoring area in real time through laser;

[0007] An image acquisition module is linked with the three-dimensional laser radar module, and is used to acquire images of foreign matter when the three-dimensional laser radar module scans the foreign matter;

[0008] A data processing module is used to acquire and store information obtained by the three-dimensional laser radar module and the image acquisition module, compare two-phase laser point cloud data obtained by the three-dimensional laser radar module before and after the invasion of foreign matter, and determine whether there is an influential foreign matter.

[0009] In an optional solution, the device further comprises an alarm module, which is used to alarm on site when the data processing module determines that there is an influential foreign matter.

[0010] Optionally, the device comprises a housing, the three-dimensional laser radar module, the image acquisition module and the data processing module are arranged inside the housing, and the alarm module is installed above the top wall of the housing.

[0011] Optionally, the alarm module is an audible and visual alarm lamp.

[0012] The application further provides a track foreign matter detection system, comprising:

[0013] A plurality of track foreign matter detection devices described above are provided with a communication unit.

[0014] A background processing module is in communication connection with the communication unit, receives information obtained by the three-dimensional laser radar module and the image acquisition module of the track foreign matter detection device, and analyzes and processes the information.

[0015] Optionally, the background processing module analyzes image information obtained by the image acquisition module based on a neural network, identifies foreign matters and classifies them.

[0016] The application further provides a track foreign matter detection method, comprising:

[0017] Real-time scanning of a monitoring area is performed by a three-dimensional laser radar module;

[0018] When the three-dimensional laser radar module scans a foreign matter, an image acquisition module is triggered to acquire images of the foreign matter;

[0019] A data processing module determines whether there is a foreign matter having an influence based on foreign matter information obtained by the three-dimensional laser radar module, and gives an alarm prompt when it is determined that there is a foreign matter having an influence.

[0020] Optionally, the method for determining whether there is a foreign matter having an influence by the data processing module comprises:

[0021] The shape, area, distance and position of the foreign matter are analyzed by comparing two-phase laser point cloud data before and after the foreign matter invades a line, and it is determined whether the foreign matter is a foreign matter having an influence according to the influence weight proportion of the shape, area, distance and position.

[0022] Optionally, the method further comprises:

[0023] Information obtained by the three-dimensional laser radar module and the image acquisition module is sent to a background processing module;

[0024] The background processing module determines whether an alarm is needed based on the received information.

[0025] In an optional solution, the background processing module analyzes image information obtained by the image acquisition module based on a neural network, identifies foreign matters and classifies them, and determines whether to alarm according to the identification and classification results.

[0026] The present application has the advantages that:

[0027] The three-dimensional laser radar module has the advantages of high detection precision, small volume, visualization, strong anti-active interference capability, etc., and in combination with the image acquisition module, on-site video and picture data can be obtained to assist in judging the authenticity of target objects.

[0028] The system of the present application has other characteristics and advantages, which will be apparent or will be described in detail in the accompanying drawings and the subsequent detailed description incorporated herein, which together serve to explain the specific principles of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0029] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description of exemplary embodiments of the present application taken in conjunction with the accompanying drawings.

[0030] Figure 1 A schematic diagram of a track foreign matter detection device according to an embodiment of the present application is shown.

[0031] Figure 2 A partial flowchart of a track foreign matter detection method according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0032] The present application will be described in more detail below. Although preferred embodiments are provided, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present application is more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0033] Example 1

[0034] Reference Figure 1 An embodiment of the present application provides a track foreign matter detection device, which comprises:

[0035] A three-dimensional laser radar module for real-time scanning of a monitoring area by laser;

[0036] An image acquisition module linked with the three-dimensional laser radar module for image acquisition of foreign matters when the three-dimensional laser radar module scans foreign matters;

[0037] The data processing module is configured to acquire and store information obtained by the three-dimensional laser radar module and the image acquisition module, compare two-phase laser point cloud data obtained by the three-dimensional laser radar module before and after foreign matter invading the line, and determine whether there is foreign matter that has an influence.

[0038] In this embodiment, the device further comprises an alarm module, which is configured to perform on-site alarm when the data processing module determines that there is foreign matter that has an influence.

[0039] In this embodiment, the device comprises a shell, the three-dimensional laser radar module, the image acquisition module and the data processing module are arranged inside the shell, and the alarm module is installed above the top wall of the shell. The alarm module is an audible and visual alarm lamp.

[0040] The three-dimensional laser radar module is based on active laser detection technology. By emitting laser beams outward, and then comparing the received signals reflected from the target with the transmitted signals, the information related to the target, such as distance, position, area, height, etc. can be obtained. The laser radar has all-day and three-dimensional imaging characteristics, can emit a large amount of optical energy at one time, complete accurate monitoring and three-dimensional imaging of the target, and generate point cloud data with attribute information. This module can detect target objects of 10cm x 10cm size within a range of 100 meters, and analyze the distance, position and area of the target object.

[0041] The image acquisition module realizes the acquisition of on-site image data, including video and picture, and is used for secondary confirmation of the target on the track surface. When the three-dimensional laser radar detects the target, the picture of the target on the track surface is captured, and the picture information is uploaded to the user terminal in the later period. This module supports large-angle, high-resolution video picture acquisition capability, supports digital zoom capability, and ensures that the target can be captured.

[0042] The data processing module provides functions of processing, analyzing and target recognition and determination of three-dimensional laser radar data, supports updating of a track surface reference model, supports storage of three-dimensional radar data, model data and picture data obtained by the image acquisition module, supports target weight algorithm updating and maintenance, and supports data transmission and communication functions.

[0043] Reference Figure 2The principle of the present application is to use high-sensitivity detection and three-dimensional imaging technology of laser radar to obtain two periods of laser point cloud data before and after the target invades the line, to compare the two periods of laser point cloud data after processing the laser point cloud data to determine whether there is redundant material, and to determine whether there is a target invasion. First, the three-dimensional laser radar obtains the basic point cloud data of the track surface, and pre-processes the point cloud data to remove outlier and invalid point cloud data, extracts the basic data of the track surface, and completes the modeling of the track reference surface; second, the three-dimensional laser radar completes repetitive scanning of the track surface at intervals of 10 seconds, and completes rapid processing of the three-dimensional laser radar data currently obtained, establishes an octree-based spatial change detection algorithm, compares the track reference model with the current track surface laser point cloud data, marks the change area by analyzing the geometric relationship of each point with the track reference model, and determines the area of interest; finally, the point cloud data of the area of interest is analyzed and determined, the shape, area, distance and position of the target object are analyzed, and the authenticity of the invasion target is determined by weight.

[0044] The three-dimensional laser radar module of the present application has the advantages of high detection accuracy, small size, visualization, strong anti-active interference capability, etc., and can obtain on-site video and picture data in combination with the image acquisition module to assist in determining the authenticity of the target object. The present application can effectively monitor the target objects appearing on the track surface, and can detect and identify falling rocks, light objects, color steel tiles and other target objects, thereby improving the track foreign object monitoring capability.

[0045] The three-dimensional laser radar module of the present application has the advantages of high detection accuracy, small size, visualization, strong anti-active interference capability, etc., and can obtain on-site video and picture data in combination with the image acquisition module to assist in determining the authenticity of the target object. The present application can effectively monitor the target objects appearing on the track surface, and can detect and identify falling rocks, light objects, color steel tiles and other target objects, thereby improving the track foreign object monitoring capability.

[0046] Embodiment 2

[0047] The present embodiment provides a track foreign object detection system, comprising:

[0048] A plurality of track foreign object detection devices as described above are provided with a communication unit;

[0049] A background processing module is in communication connection with the communication unit, receives information obtained by the three-dimensional laser radar module and the image acquisition module of the track foreign object detection device, and analyzes and processes the information.

[0050] In the present embodiment, the background processing module analyzes the image information obtained by the image acquisition module based on a neural network, identifies foreign objects and classifies them.

[0051] The background processing module provides data storage and management display, device management, notification prompt and other functions, and the user can directly use the software to check the target object, can display the three-dimensional laser point cloud image and the picture of the target object, and can intuitively show the effect of the target object in the three-dimensional laser radar and the picture, so as to better confirm the target object.

[0052] The system is mainly applied to railway track surface foreign matter monitoring, can be combined with acousto-optic and electric equipment to assist in prompting, so as to facilitate the timely discovery of the patrol personnel and nearby personnel. The device adopts integrated design, can be used in various ways such as vertical rod and wall hanging, and the detection angle covers the whole monitoring track surface as much as possible. At the same time, the system can also be applied to tunnel, slope and other area monitoring.

[0053] Embodiment 3

[0054] The embodiment provides a track foreign matter detection method, which comprises the following steps:

[0055] The three-dimensional laser radar module is used for real-time scanning of a monitoring area;

[0056] When the three-dimensional laser radar module scans a foreign matter, the image acquisition module is triggered to acquire an image of the foreign matter;

[0057] The data processing module judges whether there is a foreign matter having an influence based on the foreign matter information acquired by the three-dimensional laser radar module, and gives an alarm prompt when it is judged that there is a foreign matter having an influence.

[0058] In the embodiment, the method for the data processing module to judge whether there is a foreign matter having an influence comprises: comparing the two-period laser point cloud data of the foreign matter before and after the foreign matter intrudes into a line, analyzing the shape, area, distance and position of the foreign matter, and judging whether the foreign matter is a foreign matter having an influence according to the influence weight proportion of the shape, area, distance and position.

[0059] In the embodiment, the method further comprises: sending the information acquired by the three-dimensional laser radar module and the image acquisition module to a background processing module; and the background processing module judges whether an alarm is needed based on the received information.

[0060] In the embodiment, the background processing module analyzes the image information acquired by the image acquisition module based on a neural network, identifies the foreign matter and classifies the foreign matter, and judges whether an alarm is needed according to the identification and classification results.

[0061] The implementation process of the method is referred to the first two embodiments, which will not be repeated here.

[0062] The above has described the embodiments of the present application, and the above description is exemplary, is not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes are obvious to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A method for detecting foreign objects on a railway track, utilizing a railway foreign object detection system, characterized in that, The track foreign object detection system includes multiple track foreign object detection devices, each of which includes: A 3D lidar module is used to monitor an area in real time by scanning with a laser. The image acquisition module, which is linked with the three-dimensional lidar module, is used to acquire images of the foreign object when the three-dimensional lidar module scans it. The data processing module is used to acquire and store the information obtained by the three-dimensional lidar module and the image acquisition module, and compare the laser point cloud data before and after the foreign object intrusion obtained by the three-dimensional lidar module to determine whether there is an influential foreign object. The orbital foreign object detection method includes: The monitoring area is scanned in real time using a 3D lidar module; When the three-dimensional lidar module detects a foreign object, it triggers the image acquisition module to acquire an image of the foreign object. The data processing module determines whether there are any foreign objects that could have an impact based on the foreign object information obtained by the three-dimensional lidar module, and issues an alarm when it determines that there are any foreign objects that could have an impact. Specifically, determining whether an influential foreign object exists based on the foreign object information acquired by the three-dimensional lidar module includes: The basic laser point cloud data of the track surface is obtained by acquiring the basic laser point cloud data of the track surface based on the three-dimensional lidar, and the basic laser point cloud data is preprocessed to obtain the basic data of the track surface. Establish a track surface reference model based on the track surface basic data; The three-dimensional lidar is controlled to repeatedly scan the track surface at 10-second intervals to acquire the latest laser point cloud data in real time and perform preprocessing to obtain real-time laser point cloud data. The octree-based spatial change detection algorithm compares real-time laser point cloud data with the orbital surface reference model point by point. By analyzing the geometric relationship between each point and the orbital surface reference model, it identifies all change regions that are inconsistent with the orbital reference model and marks them as regions of interest. Based on the laser point cloud data of the region of interest, the shape, area, distance and position of the intruding target are analyzed, and the authenticity of the intruding target is determined by weighting.

2. The method for detecting foreign objects on a track according to claim 1, characterized in that, The device also includes an alarm module, which is used to issue an on-site alarm when the data processing module determines that there is an influential foreign object.

3. The method for detecting foreign objects on a track according to claim 2, characterized in that, The device also includes a housing, inside which the three-dimensional lidar module, the image acquisition module, and the data processing module are disposed, and the alarm module is installed on the top wall of the housing.

4. The method for detecting foreign objects on a track according to claim 2, characterized in that, The alarm module is an audible and visual alarm light.

5. The method for detecting foreign objects on a track according to claim 1, characterized in that, The system also includes: Communication unit; The background processing module is connected to the communication unit, receives information from the three-dimensional lidar module and image acquisition module transmitted by the track foreign object detection device, and analyzes and processes the information.

6. The method for detecting foreign objects on a track according to claim 5, characterized in that, The method further includes: The information obtained by the three-dimensional lidar module and the image acquisition module is sent to the background processing module; The background processing module determines whether an alarm is needed based on the received information.

7. The method for detecting foreign objects on a track according to claim 6, characterized in that, The background processing module analyzes the image information obtained by the image acquisition module based on a neural network, identifies and classifies foreign objects, and determines whether an alarm is needed based on the identification and classification results.

Images