Train state recognition method and device, electronic equipment and storage medium

Abstract

The invention provides a train state recognition method and device, electronic equipment and a storage medium, and belongs to the technical field of rail transit. Comprising the steps that image information collected by a visual sensor on a platform in the rail direction, train speed information collected by a millimeter wave radar sensor, first distance information between a train and the platform and second distance information collected by an infrared sensor in the direction perpendicular to the rail direction are obtained; according to the image information, the train speed information, the first distance information and the second distance information, the arrival and departure station state and the marshalling length of the train are recognized. Through the multi-sensor fusion technology, the train state is comprehensively recognized from multiple dimensions such as the train image, the train distance and the train speed, and the train departure state and the marshalling length can be accurately, reliably and efficiently recognized.

Description

Technical Field

[0001] This invention relates to the field of rail transit technology, and in particular to a train status identification method, device, electronic device, and storage medium. Background Technology

[0002] In railway transportation systems, accurately knowing the arrival and departure status of trains and their train formation length is crucial for ensuring safe operation and improving transportation efficiency. However, existing technologies have many shortcomings in this regard.

[0003] Currently, train arrival and departure status identification mainly relies on manual observation, traditional track circuits, and some simple sensor devices. Manual observation is limited by human attention and reaction speed, making it prone to oversights and misjudgments. While traditional track circuits can detect whether a train occupies the track, they struggle to comprehensively and accurately identify the train's precise location, direction of travel, and specific arrival / departure status (such as whether it is preparing to stop or has fully entered the station). Regarding train formation length identification, existing methods suffer from high cost, low accuracy, or limited applicability. For example, calculating train length by installing positioning devices on the locomotive and tail car requires manual installation by train crew on the tail car after formation, which is not only labor-intensive but also inconvenient to maintain. Calculating train length by installing active electronic tags on the locomotive and each car is costly and complex to maintain. Calculating train length by continuously measuring the displacement of the train's head and tail through insulating joints requires track circuits installed along the trackside, limiting its application and making it unsuitable for complex track environments.

[0004] As railway transportation develops towards intelligence and efficiency, there is an urgent need for a technology that can accurately, reliably, and efficiently identify the arrival and departure status of trains and the length of train formations. Summary of the Invention

[0005] This invention provides a train status identification method, device, electronic device, and storage medium to solve the technical problem of how to accurately, reliably, and efficiently identify the arrival and departure status of trains and the length of train formations.

[0006] This invention provides a train status identification method, comprising: The system acquires image information collected by visual sensors on the platform in the direction of the track, train speed information collected by millimeter-wave radar sensors, first distance information between the train and the platform, and second distance information collected by infrared sensors in the direction perpendicular to the track. Based on the image information, the train speed information, the first distance information, and the second distance information, the arrival and departure station status and train formation length of the train are identified.

[0007] According to a train status recognition method provided by the present invention, the step of recognizing the arrival / departure station status and train formation length of a train based on the image information, the train speed information, the first distance information, and the second distance information includes: The image information, the train speed information, the first distance information, and the second distance information are input into the train status recognition model to obtain the arrival / departure station status and the train formation length output by the train status recognition model. The training set of the train status recognition model includes samples of the image information, samples of the train speed information, samples of the first distance information, samples of the second distance information, and corresponding samples of the arrival / departure station status and samples of the train formation length.

[0008] According to a train status recognition method provided by the present invention, before recognizing the arrival / departure station status and train formation length based on the image information, the train speed information, the first distance information, and the second distance information, the method further includes: Obtain the visibility of the area where the platform is located; Based on the principle that the weight of the image information in the train status recognition model is positively correlated with the visibility, the weight of the image information is determined according to the visibility. Based on the principle that the weights of the train speed information and the first distance information in the train state recognition model are negatively correlated with the visibility, the weights of the train speed information and the first distance information are determined according to the visibility.

[0009] According to a train status recognition method provided by the present invention, the step of recognizing the arrival / departure station status and train formation length of a train based on the image information, the train speed information, the first distance information, and the second distance information includes: The arrival / departure station status is identified based on the image information, the train speed information, and the first distance information; The train formation length is identified based on the image information, the train speed information, the first distance information, and the second distance information.

[0010] According to a train status recognition method provided by the present invention, the step of recognizing the arrival / departure station status based on the image information, the train speed information, and the first distance information includes: If the area occupied by the train image in the image information gradually increases, and the train speed information and the first distance information gradually decrease, then the arrival / departure station status is determined to be the station entry status. If the image information and the first distance information remain unchanged, and the train speed information is zero, then the arrival / departure station status is determined to be a stopped status. If the area occupied by the train image in the image information gradually decreases, and the train speed information and the first distance information gradually increase, then the arrival / departure station status is determined to be the departure status.

[0011] According to a train status recognition method provided by the present invention, the step of recognizing the train formation length based on the image information, the train speed information, the first distance information, and the second distance information includes: The train formation length is calculated based on the train speed information and the first distance information; The image information is identified to obtain the number of train carriages and the number of connection features between the carriages; The train formation is determined based on the second distance information; The train formation length is verified based on the number of carriages, the number of connecting features between the carriages, and the train formation type.

[0012] According to a train status recognition method provided by the present invention, the second distance information includes a third distance information collected by the infrared sensor at the 1 / 4 position of the platform and a fourth distance information collected at the 3 / 4 position of the platform; Determining the train formation based on the second distance information includes: If both the third distance information and the fourth distance information are less than a preset threshold, then the train formation is determined to be a long formation. If one of the third distance information and the fourth distance information is greater than a preset threshold and the other is less than the preset threshold, then the train formation is determined to be a short formation.

[0013] The present invention also provides a train status identification device, comprising: The acquisition module is used to acquire image information collected by the visual sensor on the platform in the direction of the track, train speed information collected by the millimeter-wave radar sensor, first distance information between the train and the platform, and second distance information collected by the infrared sensor in the direction perpendicular to the track. The identification module is used to identify the arrival / departure station status and train formation length of the train based on the image information, the train speed information, the first distance information, and the second distance information.

[0014] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the train status recognition method as described above.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the train status identification method as described above.

[0016] The train status identification method, device, electronic equipment, and storage medium provided by this invention, through multi-sensor fusion technology, comprehensively identify the train status from multiple dimensions such as train image, train distance, and train speed, and can accurately, reliably, and efficiently identify the train departure status and train formation length. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the train status recognition method provided by the present invention.

[0019] Figure 2 This is a schematic diagram showing the installation positions of the millimeter-wave radar sensor and infrared sensor provided by the present invention.

[0020] Figure 3 This is a schematic diagram of the image recognition principle provided by the present invention.

[0021] Figure 4 This is a schematic diagram of the speed changes of trains entering and leaving stations provided by the present invention.

[0022] Figure 5 This is a schematic diagram of the intelligent detection system provided by the present invention.

[0023] Figure 6 This is a schematic diagram of the train status recognition device provided by the present invention.

[0024] Figure 7 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0026] The following is combined Figures 1-7This invention describes the train status identification method, apparatus, electronic device, and storage medium provided by the present invention.

[0027] Currently, in terms of system-level control, platform screen door systems primarily connect to the signaling system's command and status feedback interfaces for automatic door opening and closing signals, with the sliding doors automatically controlled by the signaling system. The signaling system issues door opening and closing commands when trains enter or leave the station. A major problem with system-level control is that most older train models operating on lines already equipped with platform screen doors lack ATO (Automatic Train Control) functionality and cannot be retrofitted. Therefore, the signaling system lacks the capability to link with the platform screen door system. Platform screen door control is entirely achieved through manual vehicle identification and operation of buttons on a local control panel. With manual operation, if vehicle identification is incorrect or button operation is erroneous, causing the platform screen door to open outside the vehicle's range, it will pose a significant safety hazard. In summary, without the involvement of a signaling system, researching train arrival / departure status and train formation length recognition is crucial to reducing station operating costs, improving operational efficiency, and ensuring passenger and train safety.

[0028] Figure 1 This is a flowchart illustrating the train status recognition method provided by the present invention, as shown below. Figure 1 As shown, this includes, but is not limited to, steps S1 and S2.

[0029] Step S1: Acquire image information collected by the visual sensor on the platform in the direction of the track, train speed information collected by the millimeter-wave radar sensor, first distance information between the train and the platform, and second distance information collected by the infrared sensor in the direction perpendicular to the track.

[0030] This invention allows for the installation of high-definition cameras as visual sensors on pillars at both ends of a station platform, selecting locations with a wide field of view and clear capture of the entire train. The camera's installation height and angle should be precisely calculated and adjusted to ensure coverage of the entire process of the train entering, stopping, and leaving the station. After installation, the visual sensor is calibrated and its parameters are set, including image resolution, frame rate, and exposure time, to ensure clear and accurate image quality. When a train enters the visual sensor's field of view, the image information captured by the sensor includes the train image; when the train leaves the station and exits the visual sensor's field of view, the image information captured by the sensor does not include the train image.

[0031] Millimeter-wave radar sensors can measure the distance and speed between the train and the sensor (station) in real time. Millimeter-wave radar has advantages such as high precision and strong anti-interference capabilities, and can operate stably even in complex weather conditions (such as rain, fog, and dust storms). Figure 2As shown, this invention allows for the installation of millimeter-wave radar sensors at suitable locations at both ends of the platform. The installation height should be slightly higher than the train roof to ensure that the radar waves can cover the entire length of the train. The installation angle of the millimeter-wave radar sensors needs to be precisely adjusted to accurately measure the distance between the train and the millimeter-wave radar sensor (platform) and the train speed. After installation, the millimeter-wave radar sensors are calibrated and tested, including verifying parameters such as distance measurement accuracy and speed measurement accuracy, to ensure that the millimeter-wave radar sensors can function properly.

[0032] like Figure 2 As shown, this invention allows for the installation of infrared sensors at two suitable locations, one before and one after, on the platform side (on the platform screen door). The infrared sensors collect second distance information in a direction perpendicular to the track. When a train is directly in front of the infrared sensor, the second distance information is relatively small; when there is no train directly in front of the infrared sensor, the second distance information is relatively large.

[0033] Step S2: Based on image information, train speed information, first distance information, and second distance information, identify the train's arrival / departure station status and train formation length.

[0034] The arrival and departure station status of a train includes the station entry status, the station stop status, and the station departure status.

[0035] Understandably, if the train is approaching the station, the area occupied by the train image in the image information will gradually increase as the train gets closer to the visual sensor; if the train is stopped, the area occupied by the train image in the image information will remain unchanged; if the train is leaving the station, the area occupied by the train image in the image information will gradually decrease as the train moves away from the visual sensor. Therefore, the arrival and departure status of the train can be obtained through image recognition.

[0036] Understandably, since a train is composed of interconnected carriages, the image information includes images of each carriage and the connecting features between them. Therefore, the number of carriages in a train can be determined through image recognition, which helps in determining the train's formation length. For example, if the image information shows 8 carriages and 7 connecting features, then the train can be identified as an 8-car train.

[0037] Understandably, if a train is approaching a station, its speed will gradually decrease; if the train is stopped, its speed will remain at zero; and if the train is departing from a station, its speed will gradually increase. Therefore, the arrival and departure status of a train can be identified based on its speed information.

[0038] Understandably, if the train is approaching the station, the initial distance information will gradually decrease as the train gets closer to the millimeter-wave radar sensor; if the train is stopped, the initial distance information will remain unchanged; and if the train is leaving the station, the initial distance information will gradually increase as the train moves away from the millimeter-wave radar sensor. Therefore, the arrival and departure status of the train can be identified based on the initial distance information.

[0039] As mentioned above, the second distance information is smaller when the train is directly in front of the infrared sensor, and larger when there is no train directly in front of the infrared sensor. Therefore, the train formation (long or short) can be identified based on the magnitude of the second distance information collected by the infrared sensors at the two locations, thus helping to determine the train's formation length.

[0040] This invention uses multi-sensor fusion technology to comprehensively identify train status from multiple dimensions such as image, distance, and speed, and can accurately, reliably, and efficiently identify train departure status and train formation length.

[0041] In one embodiment, step S2 of the present invention may specifically include: The image information, train speed information, first distance information and second distance information are input into the train status recognition model to obtain the arrival and departure station status and train formation length output by the train status recognition model. The training set for the train status recognition model includes samples of image information, train speed information, first distance information, second distance information, and corresponding arrival / departure station status and train formation length.

[0042] This invention can collect a large amount of data on different types of trains under various operating conditions (entering, stopping, and leaving the station) in actual railway operation environments as training data, including image information collected by visual sensors, distance information collected by infrared sensors, and distance and speed information collected by millimeter-wave radar sensors. Simultaneously, the corresponding train arrival and departure station status and train formation length are recorded as labels for the training data.

[0043] The training data is then divided into training, validation, and test sets. The training set is used to train the train status recognition model. By continuously adjusting the model's parameters (such as the weights and thresholds of the neural network), the model can accurately identify the arrival and departure status of trains and calculate the train formation length. During training, the validation set is used to evaluate the model's performance, and the training strategy is adjusted in a timely manner to prevent overfitting or underfitting.

[0044] Based on the test results on the test set, the model is further optimized. For example, by improving data preprocessing methods, adjusting the model structure, and increasing the amount of training data, the model's accuracy, stability, and generalization ability can be improved. Simultaneously, the model is regularly updated and maintained to adapt to the ever-changing situation in the railway transportation system (such as the introduction of new trains and the renovation of station facilities).

[0045] After the model is trained, the collected image information, train speed information, first distance information and second distance information are input into the train status recognition model to obtain the train's arrival and departure station status and train formation length.

[0046] This allows for accurate, reliable, and efficient identification of train arrival and departure station status and train formation length using a large model.

[0047] In one embodiment, prior to step S2, the train status identification method of the present invention may further include: Obtain the visibility of the area where the platform is located; Based on the principle that the weight of image information in the train status recognition model is positively correlated with visibility, the weight of image information is determined according to visibility. Based on the principle that the weights of train speed information and first distance information in the train state recognition model are negatively correlated with visibility, the weights of train speed information and first distance information are determined according to visibility.

[0048] Because different sensors adapt differently to various environmental conditions, the reliability of data collected by these sensors varies under different environments. Visual sensors provide reliable image information in good weather conditions, while millimeter-wave radar sensors provide reliable distance and speed information in adverse weather conditions (such as rain, fog, and sandstorms). Therefore, in clear weather with good visibility, the visibility in the platform area is high, and appropriately increasing the weight of visual sensor data in the train status recognition model can improve the accuracy of train status recognition. Conversely, in adverse weather conditions with low visibility in the platform area, the reliability of visual sensor data is low, and appropriately increasing the weight of millimeter-wave radar sensor data can improve the accuracy of train status recognition.

[0049] This invention enables the model to maintain high performance under various environmental conditions by dynamically adjusting the weights of data from each sensor, thereby enhancing the model's environmental adaptability.

[0050] The present invention can also identify train status through methods other than large models; that is, in one embodiment, step S2 may specifically include: The arrival and departure station status is identified based on image information, train speed information, and initial distance information; The train formation length is identified based on image information, train speed information, first distance information, and second distance information.

[0051] In one embodiment, the present invention identifies the arrival / departure station status based on image information, train speed information, and first distance information, which may specifically include: If the area occupied by the train image in the image information gradually increases while the train speed information and the first distance information gradually decrease, then the arrival / departure station status is determined to be the station entry status. If the image information and the first distance information remain unchanged, and the train speed information is zero, then the arrival / departure station status is determined to be a stopped status. If the area occupied by the train image in the image information gradually decreases while the train speed information and the first distance information gradually increase, then the arrival / departure station status is determined to be the departure status.

[0052] The image recognition principle of this invention is as follows: Figure 3 As shown, the train speeds under different arrival and departure station conditions are as follows: Figure 4 As shown. The principle of identifying the arrival and departure station status based on image information, train speed information, and first distance information has been introduced earlier, and will not be repeated here.

[0053] Judging the arrival and departure status of a train by comprehensively considering the train image, train speed, and train distance can improve the accuracy of train arrival and departure status identification.

[0054] Of course, the present invention can also identify the train's start-up preparation actions based on the image information of the visual sensor, such as changes in the headlights and changes in the lighting status inside the carriages, in order to identify the train's departure status.

[0055] In one embodiment, the present invention identifies the train formation length based on image information, train speed information, first distance information, and second distance information, which may specifically include: The train formation length is calculated based on train speed information and first distance information; The image information is identified to obtain the number of train carriages and the number of connection features between the carriages; The train formation is determined based on the second distance information; The train formation length is verified based on the number of carriages, the number of connecting features between carriages, and the train formation format.

[0056] The first distance information of the present invention may include the distance between the front of the train and the platform, the distance between the rear of the train and the platform, and the train's formation length can be calculated by combining the train's running direction, train speed information and the first distance information.

[0057] Since current train formations are limited to 8, 16, and 17 cars, this invention allows for the installation of infrared sensors at the 1 / 4 and 3 / 4 positions on the platform. Furthermore, to ensure reliable detection, two sets of infrared sensors can be installed at each location. Figure 2 As shown. After confirming that the train is in a stopped state, the train formation can be determined based on the second distance information.

[0058] Therefore, the second distance information includes the third distance information collected by the infrared sensor at 1 / 4 position on the platform and the fourth distance information collected at 3 / 4 position on the platform; The train formation is determined based on the second distance information, specifically including: If both the third and fourth distance information are less than the preset threshold, the train formation is determined to be a long formation. If either the third distance information or the fourth distance information is greater than a preset threshold and the other is less than a preset threshold, then the train formation is determined to be a short formation.

[0059] If both the third and fourth distance information are less than the preset threshold, it means that there are trains at both the 1 / 4 and 3 / 4 positions of the platform, and the train formation can be determined to be a long formation. If one of the third and fourth distance information is greater than the preset threshold and the other is less than the preset threshold, for example, if the third distance information is less than the preset threshold and the fourth distance information is greater than the preset threshold, it means that there are trains at the 1 / 4 position of the platform and no trains at the 3 / 4 position, and the train formation can be determined to be a short formation.

[0060] The train formation length is verified based on the number of carriages, the number of connecting features between carriages, and the train formation type. Specifically, if the number of carriages, the number of connecting features between carriages, and the train formation type all match the calculated formation length, it means that the calculated formation length is accurate; otherwise, it is necessary to further verify whether the calculated formation length is accurate.

[0061] Verifying train formation length using image information and second distance information can improve the accuracy of train formation length.

[0062] As described above, compared to traditional methods for identifying train arrival / departure status and train formation length, this invention reduces the need for manual intervention and the installation and maintenance of complex equipment. For example, it eliminates the need for train operators to manually install tail-end positioning devices or active electronic tags in each carriage, reducing labor and equipment costs. Simultaneously, the improved accuracy and reliability of the system reduce operational accidents and losses caused by misjudgments, further lowering operating costs. Accurate identification of train arrival / departure status and train formation length provides more timely and accurate information to the railway dispatching system, helping to optimize train operation plans and scheduling schemes, reduce train waiting times and delays, and improve the overall efficiency of railway transportation.

[0063] The intelligent detection system of the present invention, such as Figure 5 As shown, based on Figure 5 This invention enables train-to-ground coordination upon detecting a train's stop: if a train door opening action is detected, the platform door corresponding to the train door opens; if a train door closing action is detected, the platform door corresponding to the train door closes. Synchronizing the opening and closing of train doors and platform doors based on train length and arrival / departure station status greatly reduces passenger waiting risks and ensures travel safety.

[0064] like Figure 6 As shown, the present invention provides a train status identification device, comprising: The acquisition module is used to acquire image information collected by the visual sensor on the platform in the direction of the track, train speed information collected by the millimeter-wave radar sensor, first distance information between the train and the platform, and second distance information collected by the infrared sensor in the direction perpendicular to the track. The recognition module is used to identify the arrival and departure station status and train formation length of the train based on image information, train speed information, first distance information and second distance information.

[0065] It should be noted that the train status identification device provided by the present invention can execute the train status identification method of any of the above embodiments during specific operation, which will not be elaborated in this embodiment.

[0066] Furthermore, the identification module can be used for: The image information, train speed information, first distance information and second distance information are input into the train status recognition model to obtain the arrival and departure station status and train formation length output by the train status recognition model. The training set for the train status recognition model includes samples of image information, train speed information, first distance information, second distance information, and corresponding arrival / departure station status and train formation length.

[0067] Furthermore, the train status identification device may also include a determination module for: Obtain the visibility of the area where the platform is located; Based on the principle that the weight of image information in the train status recognition model is positively correlated with visibility, the weight of image information is determined according to visibility. Based on the principle that the weights of train speed information and first distance information in the train state recognition model are negatively correlated with visibility, the weights of train speed information and first distance information are determined according to visibility.

[0068] Furthermore, the identification module can be used for: The arrival and departure station status is identified based on image information, train speed information, and initial distance information; The train formation length is identified based on image information, train speed information, first distance information, and second distance information.

[0069] Furthermore, the identification module can be used for: If the area occupied by the train image in the image information gradually increases while the train speed information and the first distance information gradually decrease, then the arrival / departure station status is determined to be the station entry status. If the image information and the first distance information remain unchanged, and the train speed information is zero, then the arrival / departure station status is determined to be a stopped status. If the area occupied by the train image in the image information gradually decreases while the train speed information and the first distance information gradually increase, then the arrival / departure station status is determined to be the departure status.

[0070] Furthermore, the identification module can be used for: The train formation length is calculated based on train speed information and first distance information; The image information is identified to obtain the number of train carriages and the number of connection features between the carriages; The train formation is determined based on the second distance information; The train formation length is verified based on the number of carriages, the number of connecting features between carriages, and the train formation format.

[0071] Furthermore, the second distance information includes the third distance information collected by the infrared sensor at 1 / 4 position on the platform and the fourth distance information collected at 3 / 4 position on the platform; The recognition module can be used for: If both the third and fourth distance information are less than the preset threshold, the train formation is determined to be a long formation. If either the third distance information or the fourth distance information is greater than a preset threshold and the other is less than a preset threshold, then the train formation is determined to be a short formation.

[0072] Figure 7 This is a schematic diagram of the structure of the electronic device provided by the present invention, such as... Figure 7As shown, the electronic device may include a processor, a communications interface, a memory, and a communication bus, wherein the processor, communications interface, and memory communicate with each other via the communication bus. The processor can call logical instructions in the memory to execute a train status recognition method, which includes: acquiring image information collected by a visual sensor on the platform in the direction of the track, train speed information collected by a millimeter-wave radar sensor, a first distance information between the train and the platform, and a second distance information collected by an infrared sensor in the direction perpendicular to the track; and identifying the train's arrival / departure status and train formation length based on the image information, train speed information, first distance information, and second distance information.

[0073] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0074] On the other hand, the present invention also provides a computer program product, the computer program product including a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, when the program instructions are executed by a computer, the computer is able to execute the train status recognition method provided in the above embodiments, the method including: acquiring image information collected by a visual sensor on the platform in the direction of the track, train speed information collected by a millimeter-wave radar sensor and a first distance information between the train and the platform, and a second distance information collected by an infrared sensor in the direction perpendicular to the track; and identifying the arrival and departure station status and train formation length of the train based on the image information, train speed information, first distance information and second distance information.

[0075] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the train status recognition method provided in the above embodiments. The method includes: acquiring image information collected by a visual sensor on the platform in the direction of the track, train speed information collected by a millimeter-wave radar sensor, a first distance information between the train and the platform, and a second distance information collected by an infrared sensor in the direction perpendicular to the track; and identifying the arrival / departure status and train formation length of the train based on the image information, train speed information, first distance information, and second distance information.

[0076] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0077] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0078] 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 the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A train status recognition method, characterized in that, include: The system acquires image information collected by visual sensors on the platform in the direction of the track, train speed information collected by millimeter-wave radar sensors, first distance information between the train and the platform, and second distance information collected by infrared sensors in the direction perpendicular to the track. Based on the image information, the train speed information, the first distance information, and the second distance information, the arrival and departure station status and train formation length of the train are identified.

2. The train status identification method according to claim 1, characterized in that, The step of identifying the train's arrival / departure station status and train formation length based on the image information, the train speed information, the first distance information, and the second distance information includes: The image information, the train speed information, the first distance information, and the second distance information are input into the train status recognition model to obtain the arrival / departure station status and the train formation length output by the train status recognition model. The training set of the train status recognition model includes samples of the image information, samples of the train speed information, samples of the first distance information, samples of the second distance information, and corresponding samples of the arrival / departure station status and samples of the train formation length.

3. The train status identification method according to claim 2, characterized in that, Before identifying the train's arrival / departure station status and train formation length based on the image information, the train speed information, the first distance information, and the second distance information, the method further includes: Obtain the visibility of the area where the platform is located; Based on the principle that the weight of the image information in the train status recognition model is positively correlated with the visibility, the weight of the image information is determined according to the visibility. Based on the principle that the weights of the train speed information and the first distance information in the train state recognition model are negatively correlated with the visibility, the weights of the train speed information and the first distance information are determined according to the visibility.

4. The train status identification method according to claim 1, characterized in that, The step of identifying the train's arrival / departure station status and train formation length based on the image information, the train speed information, the first distance information, and the second distance information includes: The arrival / departure station status is identified based on the image information, the train speed information, and the first distance information; The train formation length is identified based on the image information, the train speed information, the first distance information, and the second distance information.

5. The train status identification method according to claim 4, characterized in that, The step of identifying the arrival / departure station status based on the image information, the train speed information, and the first distance information includes: If the area occupied by the train image in the image information gradually increases, and the train speed information and the first distance information gradually decrease, then the arrival / departure station status is determined to be the station entry status. If the image information and the first distance information remain unchanged, and the train speed information is zero, then the arrival / departure station status is determined to be a stopped status. If the area occupied by the train image in the image information gradually decreases, and the train speed information and the first distance information gradually increase, then the arrival / departure station status is determined to be the departure status.

6. The train status identification method according to claim 4, characterized in that, The step of identifying the train formation length based on the image information, the train speed information, the first distance information, and the second distance information includes: The train formation length is calculated based on the train speed information and the first distance information; The image information is identified to obtain the number of train carriages and the number of connection features between the carriages; The train formation is determined based on the second distance information; The train formation length is verified based on the number of carriages, the number of connecting features between the carriages, and the train formation type.

7. The train status identification method according to claim 6, characterized in that, The second distance information includes the third distance information collected by the infrared sensor at the 1 / 4 position of the platform and the fourth distance information collected at the 3 / 4 position of the platform; Determining the train formation based on the second distance information includes: If both the third distance information and the fourth distance information are less than a preset threshold, then the train formation is determined to be a long formation. If one of the third distance information and the fourth distance information is greater than a preset threshold and the other is less than the preset threshold, then the train formation is determined to be a short formation.

8. A train status identification device, characterized in that, include: The acquisition module is used to acquire image information collected by the visual sensor on the platform in the direction of the track, train speed information collected by the millimeter-wave radar sensor, first distance information between the train and the platform, and second distance information collected by the infrared sensor in the direction perpendicular to the track. The identification module is used to identify the arrival / departure station status and train formation length of the train based on the image information, the train speed information, the first distance information, and the second distance information.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the train status recognition method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the train status recognition method as described in any one of claims 1 to 7.

Images