A method and device for assisting railway route handling based on video linkage processing of poor branching

Abstract

The application provides a kind of railway route auxiliary handling method based on video linkage processing shunt bad, comprising the following steps: step 1, according to shunting operation plan generation route, and the route is checked, whether there is shunt bad section;Step 2, when there is shunt bad section, based on the route area where shunt bad section is located finds camera and route mapping table, obtains associated camera address and magnification;Step 3, by camera address and magnification, the route area video that corresponding camera real-time shooting is consulted;Step 4, vehicle occupancy detection is carried out to the route area video of camera by video analysis server;Step 5, output video detection result, to provide dispatch personnel with video analysis result and carry out the handling of route.

Description

Technical Field

[0001] This invention relates to the field of railway dispatching and route technology, and more specifically, to a method and device for auxiliary handling of railway routes with poor route division based on video linkage. Background Technology

[0002] The railway intelligent dispatching system comprehensively utilizes various technologies such as railway signal control, information processing, data communication, and automation to achieve full-process control and intelligent dispatching of train and shunting operations, enabling efficient, accurate, and coordinated dispatching of railway transportation resources. It includes resource optimization, predictive decision-making, train route processing, shunting route processing, real-time locomotive monitoring, data analysis, and emergency response, effectively ensuring railway transportation safety and improving operational efficiency and service quality.

[0003] A faulty track shunting condition refers to a situation where, when a train or vehicle occupies a section of track, poor contact between the wheels and the rails or the presence of insulating material on the rail surface prevents the track relay from deactivating properly, causing the computer interlocking system to fail to correctly display the red light strip or track occupancy status of the track circuit. This manifests primarily as the occupied section still displaying as free, unstable occupancy status display, and the red light strip not disappearing after the train leaves. Faulty track shunting has a significant impact on signal control, directly affecting the safety of the signal control system. With the rapid development of video technology, many stations have implemented video coverage. This has led to the development of new technical methods for confirming faulty track shunting sections. For example, CN115984200A provides a video analysis-based method for detecting faulty track shunting. This method uses video to check the track occupancy clearance status and compares it with the corresponding track occupancy clearance status sent by the computer interlocking system to determine the faulty track shunting condition.

[0004] However, in practice, dispatchers still need to confirm the status of faulty track sections when processing train and shunting routes. Specifically, dispatchers must follow a series of rigorous steps. First, before preparing a route, they must personally or assign someone else to confirm that all faulty track sections related to the route are in an unobstructed state. Only then can the route be processed, and the switches in the faulty track sections be locked individually. When conducting shunting operations, it must first be confirmed that the locomotive and rolling stock have completely left the faulty track section of the switch circuit, and this must be reported by the shunting personnel or driver. Only after confirmation can the route be opened.

[0005] In order to solve the above problems, people have been seeking an ideal technological solution. Summary of the Invention

[0006] Therefore, it is necessary to provide a method and device for assisting in the handling of railway routes with poor routing based on video linkage to address the above-mentioned technical problems. For sections with poor routing, the video screen of the route to be sorted is automatically switched to the current scene for automatic identification, eliminating the need for manual selection of a camera and zooming, which is fast and efficient.

[0007] To achieve the above objectives, the first aspect of the present invention provides a method for auxiliary processing of railway routes with poor routing based on video linkage, comprising the following steps:

[0008] Step 1: Generate the approach route according to the shunting operation plan, and check the approach route to determine if there are any sections with poor routing.

[0009] Step 2: When there is a faulty routing section, look up the camera-to-routes mapping table based on the route area where the faulty routing section is located to obtain the associated camera address and magnification.

[0010] Step 3: Access the real-time video of the route area captured by the corresponding camera using the camera address and magnification.

[0011] Step 4: Use the video analytics server to detect vehicle occupancy in the camera's access area video.

[0012] Step 5: Output the video detection results so that dispatchers can process routes based on the video analysis results.

[0013] To achieve the above objectives, a second aspect of the present invention provides a railway route auxiliary processing device based on video linkage processing for poor routing, comprising:

[0014] The intelligent dispatching system is configured to generate entry routes based on the shunting operation plan and check the entry routes to determine if there are any sections with poor routing.

[0015] The faulty splitter video linkage module is configured to, when a faulty splitter section exists, look up the camera-path mapping table based on the path area where the faulty splitter section is located, obtain the associated camera address and magnification, and then retrieve the real-time video of the path area captured by the corresponding camera using the camera address and magnification.

[0016] The vehicle occupancy detection module is configured to perform vehicle occupancy detection on the video of the camera's access area retrieved by the faulty video linkage module via a video analysis server.

[0017] The dual-layer display includes an interlocking display and a video display. The interlocking display is associated with the intelligent dispatching system and is used to display the route and faulty sections generated by the intelligent dispatching system. The video display is associated with the faulty section video linkage module and the vehicle occupancy detection module and is configured to display the accessed route area video and vehicle occupancy detection results.

[0018] The beneficial effects of this invention are as follows:

[0019] When a faulty road segment is detected during the route planning process, this invention automatically searches the camera-route mapping table based on the route area where the faulty segment is located. It obtains the associated camera address and magnification, and then retrieves the real-time video of the route area captured by the corresponding camera using the camera address and magnification. This automatically switches the video feed of the route to be planned to the current scene, eliminating the need for manual selection of a camera and zooming, making it faster and more efficient. Furthermore, a video analysis server performs vehicle occupancy detection on the camera's route area video; manual verification of the detection results is all that's needed, significantly improving the level of intelligence. The organic integration of the intelligent dispatch system's signal control and the video system improves the efficiency of automatic route planning for faulty road segments, solves the problem of information silos, and greatly enhances the convenience and practicality for dispatchers using both the intelligent dispatch system and the video system. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the auxiliary processing method for railway routes with poor routing based on video linkage according to the present invention.

[0021] Figure 2 It is a key process for handling auxiliary procedures for railway routes with poor routing based on video linkage;

[0022] Figure 3 This is a schematic diagram of the mapping relationship in this invention;

[0023] Figure 4 This is a diagram illustrating a one-to-one relationship;

[0024] Figure 5 This is a diagram illustrating a one-to-many relationship;

[0025] Figure 6 This is a diagram illustrating a many-to-many relationship;

[0026] Figure 7 This is a schematic diagram of the video linkage display for faulty splitter channels.

[0027] Figure 8 This is a process for analyzing defective videos using AI. Detailed Implementation

[0028] The technical solution of the present invention will be further described in detail below through specific embodiments.

[0029] Example 1

[0030] This embodiment provides a method for assisting in handling railway route misalignment based on video linkage, such as... Figure 1 and Figure 2 As shown, it includes the following steps:

[0031] Step 1: Generate the approach route based on the shunting operation plan, and check the approach route to determine if there are any sections with poor routing.

[0032] Specifically, route processing includes shunting routes and train routes, both of which require a series of condition checks during processing, including faulty track alignment. A faulty track alignment section appears as empty in the interlocking system, but in reality, there may be cars or trains that have not completely departed, posing a serious safety hazard.

[0033] Defective track switches are not detected dynamically in real time. Instead, known defective track switches are recorded in the interlocking database through routine maintenance, testing, and manual marking. Therefore, when a route is generated, on-site measurement is not required. Instead, all track sections traversed by the route are retrieved from the route table. For each section, its track switch status field is searched in the interlocking database to see if it has been marked as a defective track switch.

[0034] It is understandable that the management of faulty branch sections is an important part of the daily maintenance work of the signal specialist. When a new faulty branch section is found, it needs to be included in the management in a timely manner, and the faulty branch section should be eliminated in a timely manner after rectification.

[0035] Step 2: When there is a faulty routing section, look up the camera and route mapping table based on the route area where the faulty routing section is located to obtain the associated camera address and magnification.

[0036] Under normal circumstances, a poorly routed section may be covered by more than two cameras. To achieve the best effect, a certain magnification factor is considered. This data is incorporated into the intelligent scheduling management as a data dictionary. When using it, you can directly look up the camera-to-route mapping table based on the route area where the poorly routed section is located.

[0037] Specifically, based on the positional relationships of the basic interlocking route table covered by the station cameras, a mapping relationship is established between cameras (C) and route areas (P). A schematic diagram of the overall mapping is shown below. Figure 3 As shown.

[0038] The basic interlocking route table is the core data file of the station's interlocking system, specifying all possible train routes and shunting routes within the station and their interlocking relationships. The interlocking system uses this table to perform logical operations such as route selection, turnout locking, and signal opening.

[0039] In practice, based on the path corresponding to the camera, the mapping relationship between the camera (C) and the path area (P) can be divided into several cases such as one-to-one, one-to-many, and many-to-many.

[0040] (1) One-to-one relationship: one camera corresponds to one basic path, such as Figure 4 As shown, the camera covers D1 to D3.

[0041] (2) One-to-many relationship: One camera corresponds to multiple basic routes. This situation is mainly distributed in the turnout area, such as Figure 5 As shown, the camera covers four basic paths: D1 to D2, D1 to D4 (D4 to D1), D5 to D2, and D5 to D4. Among them, D1 to D4 is the same as D4 to D1.

[0042] (2) Many-to-many relationship: Many-to-many relationships are generally when the route is long, and one camera cannot cover the entire route, requiring two or more cameras to cover the entire route. For example Figure 6 As shown, the route from D3 to II requires coverage by cameras #1, #2, #3, #4, and #5. Simultaneously, these five cameras also cover the route from D3 to I.

[0043] It's understandable that after establishing the mapping relationship, it's necessary to incorporate the mapping correspondence into the data dictionary. When arranging routes, video feeds from different camera addresses are automatically retrieved.

[0044] Specifically, for one-to-one and one-to-many scenarios, simply using a single-view video is sufficient for the dispatcher to view the site. For many-to-many scenarios, the videos captured by the cameras need to be stitched together according to the order of their addresses to display the complete video of the route area, allowing the dispatcher to grasp the overall situation on site.

[0045] It's understandable that different camera types employ different recall mechanisms. For bullet cameras, the field of view is fixed, but they typically support motorized zoom. When planning a route, the required field of view is calculated based on the route's range, and the bullet camera's focal length is adjusted (zoom in) to cover the entire route or key sections. For PTZ cameras, preset positions can be set in advance; multiple preset positions can be set, each corresponding to a fixed horizontal / vertical angle and focal length. When planning a route, the preset position associated with that route is automatically recalled based on the route information. The PTZ camera can then quickly rotate to that position and automatically focus, achieving coverage of the live video feed. This method offers fast response and accurate positioning.

[0046] Step 3: Access the real-time video of the route area captured by the corresponding camera by using the camera address and magnification.

[0047] By displaying the video footage corresponding to the faulty section of the route using the camera address and magnification, the dispatcher can intuitively grasp the on-site situation and determine whether there are vehicles occupying the faulty section.

[0048] For a specific diagram illustrating the linked display of faulty video feeds, please refer to [link / reference]. Figure 7 .

[0049] As can be seen, a dual-layer display is used to assist in route processing. One layer of the display is associated with the intelligent dispatch system and is used to display the route and faulty sections generated by the intelligent dispatch system. It can be understood that when a faulty section is identified, the faulty section can be marked with a special color or mark, thereby providing the dispatcher with comprehensive safety monitoring information. The other layer of the display is associated with the faulty section video linkage module and the vehicle occupancy detection module and is configured to display the accessed route area video and vehicle occupancy detection results.

[0050] Understandably, during this process, the video feed can be intelligently and automatically switched in real time based on the route being processed. By intelligently linking route arrangement with on-site video, dispatchers can gain a comprehensive understanding of the on-site operation process from both signal and video perspectives.

[0051] Step 4: Use the video analytics server to detect vehicle occupancy in the camera's access area.

[0052] It is understandable that the video analytics server uses AI technology to analyze in real time whether there are vehicles occupying poorly routed sections, and will issue a real-time warning when a vehicle occupies the route.

[0053] It is understandable that the faulty splitter sections are relatively fixed sections, and the camera installation positions are also fixed to achieve fast and efficient detection. Therefore, when streaming data frames from cameras covering the faulty splitter sections, noise reduction processing is required to account for factors such as changes in lighting, rain and fog interference, and sensor noise.

[0054] In this embodiment, after retrieving the real-time video of the route area captured by the corresponding camera by camera address and magnification, the median filtering algorithm is used to clean the video, eliminate random noise in the image, remove invalid image frames such as blurry or overexposed images, and improve the overall quality of the dataset.

[0055] Furthermore, in this embodiment, when the video analysis server detects vehicle occupancy in the camera's approach area video, it employs a classification defect model trained on the YOLO V8 base model. This model is fast and efficient in specific scenarios involving vehicle occupancy in defective routes, and pre-defines prevention zones for each defective route, resulting in very high detection efficiency. After detection, the output points with vehicle occupancy are represented by status and image data. Specifically, the AI ​​analysis process for defective route video is as follows: Figure 8 As shown.

[0056] Step 5: Output the video detection results so that dispatchers can process routes based on the video analysis results.

[0057] In the above scheme, when a poorly routed section is found in the route planning process, the system automatically searches the camera-route mapping table based on the route area where the poorly routed section is located, obtains the associated camera address and magnification, and then retrieves and displays the real-time video of the route area captured by the corresponding camera using the camera address and magnification. This achieves automatic linkage switching of the video feed of the route to be planned to the current scene, eliminating the need for manual selection of a camera and zooming, making it quick and efficient. Furthermore, the video analysis server performs vehicle occupancy detection on the video feed of the camera's route area, and the operator only needs to confirm the detection results, greatly improving the level of intelligence. Through the organic integration of the intelligent dispatch system's signal control and the video system, the efficiency of automatic route planning for poorly routed sections is improved, the problem of information silos is solved, and the convenience and practicality of dispatchers using the intelligent dispatch system and the video system are greatly improved.

[0058] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0059] Example 2

[0060] Based on the same inventive concept, this application also provides a device for implementing the above-described method for assisting in the processing of faulty railway routes based on video-linked processing. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the device for assisting in the processing of faulty railway routes based on video-linked processing provided below can be found in the limitations of the method for assisting in the processing of faulty railway routes based on video-linked processing described above, and will not be repeated here.

[0061] Specifically, the video-linked auxiliary processing device for railway route misalignment includes:

[0062] The intelligent dispatching system is configured to generate entry routes based on the shunting operation plan and check the entry routes to determine if there are any sections with poor routing.

[0063] The faulty splitter video linkage module is configured to, when a faulty splitter section exists, look up the camera-path mapping table based on the path area where the faulty splitter section is located, obtain the associated camera address and magnification, and then retrieve the real-time video of the path area captured by the corresponding camera using the camera address and magnification.

[0064] The vehicle occupancy detection module is configured to perform vehicle occupancy detection on the video of the camera's access area retrieved by the faulty video linkage module via a video analysis server.

[0065] The dual-layer display includes an interlocking display and a video display. The interlocking display is associated with the intelligent dispatching system and is used to display the route and faulty sections generated by the intelligent dispatching system. The video display is associated with the faulty section video linkage module and the vehicle occupancy detection module and is configured to display the accessed route area video and vehicle occupancy detection results.

[0066] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them; although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications can still be made to the specific implementation of the present invention or equivalent substitutions can be made to some technical features without departing from the spirit of the technical solutions of the present invention, and all such modifications and substitutions should be covered within the scope of the technical solutions claimed in the present invention.

Claims

1. A method for assisting in handling railway route misalignment based on video-linked processing, characterized in that, Includes the following steps: Step 1: Generate the approach route according to the shunting operation plan, and check the approach route to determine if there are any sections with poor route division. Step 2: When there is a faulty routing section, look up the camera-to-routes mapping table based on the route area where the faulty routing section is located to obtain the associated camera address and magnification. Step 3: Access and display the real-time video of the route area captured by the corresponding camera using the camera address and magnification. Step 4: Use the video analytics server to detect vehicle occupancy in the camera's access area video. Step 5: Output the video detection results so that dispatchers can process routes based on the video analysis results.

2. The method for auxiliary processing of railway routes with poor routing based on video linkage processing according to claim 1, characterized in that, The mapping relationship between cameras and routes can be one-to-one, one-to-many, or many-to-many. The term "one-to-one" means that one camera corresponds to one basic route; The term "one-to-many" refers to one camera corresponding to multiple basic paths. The term "many-to-many" refers to multiple cameras corresponding to cover a single route.

3. The method for auxiliary processing of railway routes with poor routing based on video linkage processing according to claim 2, characterized in that, When accessing real-time video of the access area captured by the corresponding camera using the camera address and magnification, if it is a many-to-many relationship, the videos captured by the cameras need to be stitched together according to the order of the camera addresses to display the complete access area video.

4. The method for auxiliary processing of railway routes with poor routing based on video linkage processing according to claim 3, characterized in that, Different calling mechanisms are adopted according to different camera types; for bullet cameras, a scaling strategy is used to cover the specified path; for PTZ cameras, preset positions are set in advance, and during the path arrangement process, the video covers the on-site video screen by loading the preset positions.

5. The method for auxiliary processing of railway routes with poor routing based on video linkage processing according to claim 3, characterized in that, After retrieving the real-time video of the route area captured by the corresponding camera by camera address and magnification, the median filtering algorithm is used to clean and remove noise from the video.

6. A method for assisting in handling railway route defects based on video-linked processing according to claim 1 or 2, characterized in that, When the video analytics server performs vehicle occupancy detection on the video feed of the camera's approach area, it uses a poorly classified model trained on YOLO V8 as the base model.

7. A railway route auxiliary processing device based on video linkage for handling faulty routing, characterized in that, include: The intelligent dispatching system is configured to generate entry routes based on the shunting operation plan and check the entry routes to determine if there are any sections with poor routing. The faulty splitter video linkage module is configured to, when a faulty splitter section exists, look up the camera-path mapping table based on the path area where the faulty splitter section is located, obtain the associated camera address and magnification, and then retrieve the real-time video of the path area captured by the corresponding camera using the camera address and magnification. The vehicle occupancy detection module is configured to perform vehicle occupancy detection on the video of the camera's access area retrieved by the faulty video linkage module via a video analysis server. The dual-layer display includes an interlocking display and a video display. The interlocking display is associated with the intelligent dispatching system and is used to display the approach path and faulty branch sections generated by the intelligent dispatching system. The video display is associated with the faulty routing video linkage module and the vehicle occupancy detection module, and is configured to display the access route area video and vehicle occupancy detection results.

Images