Train data identification and display method and device, electronic equipment and medium

The continuous image of the train passing through the gate is obtained through the line scan camera, the confidence score is calculated and the accurate data is selected for display, which solves the problem of inaccurate data when passing through the gate at high speed, and realizes efficient and intuitive monitoring of the train status.

CN118918543APending Publication Date: 2024-11-08SHENZHEN PINGFANG SCI & TECH
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202410982836.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The existing train operating status detection and identification system is uneven when the train passes through the gate at high speed, and cannot accurately obtain and intuitively display train data, resulting in poor data display effect.

Method used

The line scan camera obtains two consecutive frames of driving images when the train passes through the gate, and recognizes them based on the first and second frame images, calculates the confidence score, selects the target train data with higher confidence for display, and generates data reports and web pages Interface to display accurate train information.

Benefits of technology

It improves the accuracy and display effect of train data, realizes reliable monitoring of train status, and ensures the accuracy and intuitiveness of data identification and display when passing through high-speed gates.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118918543A_ABST
    Figure CN118918543A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of railway transportation safety, and provides a train data identification and display method and device, electronic equipment and a medium. The method comprises the following steps: acquiring two continuous frames of driving images during train lockage through a line scan camera, performing identification based on the first frame of driving image to obtain first train data, determining a first confidence score of the first frame of driving image according to the first train data, determining the accuracy of the first train data, and performing identification based on the second frame of driving image to obtain the accuracy of the first train data. And determining a first confidence score of the first frame of driving image according to the first train data to obtain second train data, determining a second confidence score of the second frame of driving image according to the second train data to obtain the accuracy of the second train data, and finally determining target train data with higher confidence from the first train data and the second train data through the first confidence score and the second confidence score. And displaying the target train data. The technical problem that train data cannot be accurately obtained and displayed during high-speed lockage of a train is solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of railway transportation safety technology, and in particular to a method, device, electronic equipment and medium for identifying and displaying train data. Background Art

[0002] Train operation status detection and identification technology has become an important pillar in the field of railway transportation. With the help of sensors, cameras and complex algorithms, it can monitor various parameters of the train operation in real time to ensure that it can perform transportation tasks safely and efficiently.

[0003] At present, there are many train operation status detection and identification application systems on the market. They mainly rely on algorithms to verify the carriage data, and use the images captured by the camera for auxiliary identification to monitor the vehicle passing through the gate. They can also generate corresponding report statistics and historical records for users to query.

[0004] However, the accuracy of train data verification in the existing system needs to be improved, especially when the train passes through the gate at high speed. The quality of the captured images is uneven, the display effects are cross-cutting, and the train data and gate status cannot be displayed intuitively and efficiently. Summary of the invention

[0005] In view of this, the present application provides a method, device, display screen and storage medium for identifying and displaying train data, which are used to solve the technical problem that train data may not be accurately obtained and displayed when a train passes through a gate at high speed.

[0006] In the first aspect, the present application provides a method for identifying and displaying train data, which adopts the following technical solution: Use a line scan camera to obtain two consecutive frames of running images when the train passes through the gate; Perform recognition based on the first frame of the driving image to obtain first train data; Determining a first confidence score of the first frame of driving image according to the first train data; Perform recognition based on the second frame of driving image to obtain second train data; Determining a second confidence score of the second frame of driving image according to the second train data; Determining target train data from the first train data and the second train data according to the first confidence score and the second confidence score; The target train data is displayed.

[0007] Optionally, the performing recognition based on the first frame of the driving image to obtain the first train data includes: Performing text recognition on the first frame of driving image to obtain first text recognition data; Performing image recognition on the first frame of driving image to obtain first image recognition data; The first text recognition data and the first image recognition data are used as the first train data.

[0008] Optionally, performing text recognition on the first frame of driving image to obtain first text recognition data includes: Performing text detection and text positioning on the first frame of the driving image through a text detection network to obtain a first text region; Performing text recognition on the first text area through a convolutional recurrent neural network to obtain first text data; The first text recognition data is obtained by performing data matching between the first text data and a preset train database.

[0009] Optionally, performing image recognition on the first frame of driving image to obtain first image recognition data includes: Performing unsupervised segmentation on the first frame of driving image to obtain a first frame of driving sub-image set; Performing Hall transformation detection on the first frame driving sub-image set to obtain a first target image set; Performing color space conversion on the first target image set to obtain a second target image set; Performing feature extraction on the second target image set to obtain a first feature information set; The first image recognition data is obtained by performing feature matching with the preset train database according to the first feature information set.

[0010] Optionally, determining a first confidence score of the first frame of driving image according to the first train data includes: Performing matching degree recognition on the first text recognition data through a first preset matching degree recognition model to obtain a first train data matching degree; Performing matching degree recognition on the first image recognition data through a second preset matching degree recognition model to obtain a second train data matching degree; The first confidence score is obtained by performing a weighted calculation based on the first train data matching degree and the second train data matching degree.

[0011] Optionally, determining the target train data from the first train data and the second train data according to the first confidence score and the second confidence score includes: scoring the first confidence score and the second confidence score; When the first confidence score is greater than the second confidence score, determining the first train data as the target train data; When the second confidence score is greater than the first confidence score, the second train data is determined to be the target train data.

[0012] Optionally, displaying the target train data includes: Generate a data report based on the target train data; A web page interface is generated based on the data report and the web page interface is displayed.

[0013] In a second aspect, the present application provides a train data identification and display device, the device comprising: An image acquisition module is used to extract any two consecutive frames of running images from the running video of the train passing through the gate; A first recognition module, used for performing recognition based on a first frame of driving image to obtain first train data; A first scoring module, configured to determine a first confidence score of the first frame of driving image according to the first train data; A second recognition module is used to perform recognition based on the second frame of the driving image to obtain second train data; A second scoring module, configured to determine a second confidence score of the second frame of driving image according to the second train data; a data determination module, configured to determine target train data from the first train data and the second train data according to the first confidence score and the second confidence score; The data display module is used to display the target train data.

[0014] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the train data identification and display method when executing the computer program.

[0015] In a fourth aspect, the present application provides a computer-readable storage medium, comprising a computer program, which, when executed by a processor, implements the steps of the method for identifying and displaying train data.

[0016] The present application obtains two consecutive frames of driving images when a train passes through a gate through a line scan camera, identifies the first frame of the driving image to obtain the first train data, and determines the first confidence score of the first frame of the driving image. The first confidence score is used to reflect the accuracy of the first train data. The second frame of the driving image is identified to obtain the second train data, and the second confidence score of the second frame of the driving image is determined. The second confidence score is used to reflect the accuracy of the second train data. By comparing the first confidence score with the second confidence score, the target train data with a higher confidence score is obtained and a data report is generated for display, thereby improving the accuracy of the train data, and displaying the accurate data intuitively, and reliable monitoring of the train status can be achieved through accurate train data. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flow chart of a method for identifying and displaying train data according to an embodiment of the present application; Figure 2 This is a functional module diagram of a train data identification and display device shown in an embodiment of the present application; Figure 3 A structural diagram of an electronic device shown in an embodiment of the present application. DETAILED DESCRIPTION

[0018] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification of the present application, the singular expressions "one", "a kind of", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations comprising one or more of the listed items.

[0019] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0020] See also Figure 1 As shown, the train data identification and display method shown in the embodiment of the present application includes the following steps.

[0021] S11, obtaining two consecutive frames of running images of the train passing through the gate through a line scan camera.

[0022] Line scan cameras are used to capture high-resolution images of continuously moving objects. Unlike ordinary area scan cameras, line scan cameras do not capture the entire scene at once, but instead continuously acquire image data line by line along the direction of the object being photographed. The camera lens captures a line or a series of lines of the moving object, which gradually form a complete image as the object moves.

[0023] When a train passes through the gate, the line scan camera equipped in the gate captures the real-time driving image of the train, wherein the first captured image is defined as the first frame driving image, the second captured image is defined as the second frame driving image, and the first frame driving image and the second frame driving image are defined as two consecutive frames of driving images. The driving image can be a front image and a carriage image. Different train data can be obtained based on different train images. For example, the front image can include the shape of the headlights and the logo, and the carriage image can include the carriage code and the connection method of the carriages.

[0024] In the above embodiment, by extracting any two consecutive frames of video images from the travel video of the train passing through the lock, data capture and analysis can be performed on the two consecutive frames of travel images, the train data in the travel image corresponding to each frame can be obtained, and the higher quality train data in the two frames of video images can be determined, thereby improving the accuracy of train data recognition.

[0025] S12, performing recognition based on the first frame of the driving image to obtain first train data.

[0026] After obtaining any two frames of driving images of the train, the system will first perform image preprocessing on the first frame of the driving image, perform conventional denoising, enhancement and other operations on the first frame of the driving image, and improve the quality of the first frame of the driving image. The first frame of the driving image is then image recognized to obtain the first train data of the first frame of the train image. Among them, image recognition is mainly divided into two categories. One is based on the recognition of text information in the first frame of the driving image to obtain the text features of the train, such as the head text information, the number information of the carriage, etc.; the other is based on the recognition of image information in the first frame of the driving image to obtain the structural features of the train, such as the shape of the headlights, the design of the carriage doors and windows, etc.

[0027] In the above embodiment, the text information data and image information data on the first frame of driving image are obtained by identifying the first frame of driving image, and the train data of the first frame of driving image is obtained from multiple angles, thereby improving the accuracy of train data identification.

[0028] In an optional embodiment, the step of performing recognition based on the first frame of the driving image to obtain the first train data includes: Performing text recognition on the first frame of driving image to obtain first text recognition data; Performing image recognition on the first frame of driving image to obtain first image recognition data; The first text recognition data and the first image recognition data are used as the first train data.

[0029] After acquiring the first frame of the driving image, the text recognition technology can accurately extract the first text recognition information from the first frame of the driving image, such as the head text information and the car number. This information usually contains the logo and basic identification information of the train, such as the type, number and unit of the vehicle. The image recognition technology is then used to further obtain the first image recognition information in the first frame of the driving image, such as the head structure shape, car door and window design and other structural features. These image features can provide information about the external design and structure of the train.

[0030] After integrating the first text recognition data and the first image recognition data, these information are integrated into the first train data for storage and recording.

[0031] In this embodiment, text and image recognition is performed based on the first frame of the driving image, so that the first text recognition data and the first image recognition data of the train can be effectively obtained, the accuracy and reliability of train recognition can be improved, and real-time and accurate data support can be provided for train operation to ensure the safe operation of the train.

[0032] In an optional embodiment, performing text recognition on the first frame of driving image to obtain first text recognition data includes: Performing text detection and text positioning on the first frame of the driving image through a text detection network to obtain a first text region; Performing text recognition on the first text area through a convolutional recurrent neural network to obtain first text data; The first text recognition data is obtained by performing data matching between the first text data and a preset train database.

[0033] Among them, the text detection network uses CTPN (Connectionist Text Proposal Network) to perform text detection and text positioning on the first frame of the driving image. The connectionist text proposal network is composed of a convolutional neural network and a bidirectional long short-term memory network. By using the convolutional neural network to extract features from the first frame of the driving image, the text features that may exist in the first frame of the driving image are obtained, and a set of equal-width text marker lines are used to mark the corresponding areas where text features exist. The bidirectional long short-term memory network is used to capture text in the corresponding areas, and information classification is performed based on the text information obtained in the steps. The text marker lines are adjusted based on the captured text information, and the marker lines of adjacent texts are connected into complete text lines through the existing text line construction algorithm, thereby obtaining the first text area.

[0034] After determining the first text region, the first text region is subjected to text recognition through a convolutional recurrent neural network, and the convolutional recurrent neural network includes a convolutional layer, a recurrent layer, and a transcription layer. The deep text features extracted from the first text region can be modeled through the recurrent layer to capture the contextual dependencies of the text sequence, and the text or symbol at each position is iteratively predicted through the recurrent layer. Finally, the text information of the recurrent layer is output as a text sequence through the transcription layer to obtain the first text data.

[0035] After determining the first text data, the first text data is accurately matched with a preset train database. The train database contains standardized train data, such as train model, train structure and other data. Based on the text information in the first text data and the data matching with the preset vehicle database, the fuzzy text or erroneous text in the first text data can be corrected and matched to obtain the correct text data, and some recognition text data can be matched to a standard format to achieve further conversion of the first text data and obtain the first text recognition data.

[0036] In the above embodiment, by connecting the text proposal network, the text area in the driving image can be accurately captured to ensure the accuracy and completeness of text detection. The convolutional recurrent neural network can deeply mine the features of the text area and accurately identify the text content, avoiding errors and omissions in the recognition method of text information. Finally, by accurately matching with the preset train database, possible text errors can be corrected, and the recognition results can be standardized, greatly improving the availability and accuracy of the data.

[0037] In an optional embodiment, the performing image recognition on the first frame of driving image to obtain first image recognition data includes: Performing unsupervised segmentation on the first frame of driving image to obtain a first frame of driving sub-image set; Performing Hall transformation detection on the first frame driving sub-image set to obtain a first target image set; Performing color space conversion on the first target image set to obtain a second target image set; Performing feature extraction on the second target image set to obtain a first feature information set; The first image recognition data is obtained by performing feature matching with the preset train database according to the first feature information set.

[0038] Unsupervised segmentation refers to the random segmentation of the first frame of driving images without manual annotation or pre-setting. The first frame of driving images is divided into different regions, namely the first frame of driving sub-image sets, so as to facilitate the subsequent more detailed processing and analysis of these sub-images and improve the accuracy of data recognition.

[0039] Hall transform detection is used to detect specific shapes in images, such as straight lines or circles. By performing Hall transform detection on the first frame of the driving sub-image set, the train outline or other important features can be identified, and the position and parameters of the straight lines or other geometric shapes in the image can be determined, thereby effectively obtaining the shape structure information under the train and determining the first target image set with structural features.

[0040] Performing color space conversion on each image in the first target image set. Converting the first target image from one color representation to another, such as from RGB (red, green, blue) color space to HSV (hue, saturation, brightness) color space, in which the characteristic structure of the vehicle can be more accurately identified.

[0041] After the color space conversion, feature extraction is performed on the second target image set. Features can be any measurable characteristics related to the train in the image, including body color, texture features, edge features, etc. According to the feature information extracted from the second target image set, it is matched with the preset train database, and the corresponding train data is matched based on the image in the preset train database to prevent some detail errors in the recognition process from causing recognition errors or recognition accidents.

[0042] In the above embodiment, the image information of the train can be accurately identified from the first driving image through multiple steps such as image segmentation, Hall transformation, color space conversion, feature extraction and feature matching.

[0043] S13: Determine a first confidence score of the first frame of driving image according to the first train data.

[0044] After obtaining the first train data corresponding to the first frame of the driving image, the confidence score of the first driving image can be calculated based on the first train data. The first confidence score is used to measure the accuracy of the recognized first train data. The higher the first confidence score, the better the recognition effect of the first frame of the driving image and the more accurate the first train data.

[0045] In the above embodiment, by performing a first confidence score on the first train data, the accuracy of the first train data can be determined, which can be used for subsequent comparison with the confidence score of the second frame of driving image to determine a corresponding driving image with better quality.

[0046] In an optional embodiment, determining a first confidence score of the first frame of driving image according to the first train data includes: Performing matching degree recognition on the first text recognition data through a first preset matching degree recognition model to obtain a first train data matching degree; Performing matching degree recognition on the first image recognition data through a second preset matching degree recognition model to obtain a second train data matching degree; The first confidence score is obtained by performing a weighted calculation based on the first train data matching degree and the second train data matching degree.

[0047] The first preset matching degree recognition model is a pre-set database for text data on the train. By extracting the first text recognition data in the first train data, such as the vehicle identification number, date and time, etc., these text information are input into the first preset matching degree recognition model. The first preset matching degree recognition model matches the key information extracted from the text data with the corresponding train text information and outputs the first train data matching degree, which reflects the accuracy and matching degree of the text information in the image.

[0048] The second preset matching recognition model is a pre-set database for image recognition data. For the structural features in the first frame of the driving image, such as the outline and external structure of the train, the second preset matching recognition model is used to extract and match specific train data from the first frame of the display image to generate a matching score. The second train data matching degree reflects the accuracy and matching degree of the structural features in the image.

[0049] After obtaining the matching degree of the first train data and the matching degree of the second train data, they can be calculated using a preset weighted ratio, and the overall confidence of the image can be comprehensively evaluated based on the relative weights of the matching degree of the first train data and the matching degree of the second train data.

[0050] In the above embodiment, by performing double matching on the first text recognition data and the first image recognition data through the first preset matching degree recognition model and the second preset matching degree recognition model, the analysis capability and accuracy of the train data are improved.

[0051] S14, performing recognition based on the second frame of driving image to obtain second train data.

[0052] After the first frame of the driving image is recognized, the second frame of the driving image will be recognized immediately. The second frame of the driving image may differ from the first frame of the driving image in certain details, such as differences in clarity and differences caused by sudden changes in external lighting. The second frame of the driving image is recognized, and the specific recognition method is the same as the recognition method for the first frame of the driving image in S12. The second train data can be obtained by performing image recognition on the second frame of the driving image.

[0053] In the above embodiment, by obtaining the second train data corresponding to the second frame of the driving image, it can be used to compare with the first train data of the first frame of the driving image to obtain more accurate train data, so that the user can continuously obtain more accurate train data based on each frame.

[0054] S15: Determine a second confidence score of the second frame of driving image according to the second train data.

[0055] After the second train data is determined, the same confidence scoring method as in S13 is executed to obtain a second confidence score of the second frame of the driving image.

[0056] By obtaining the second confidence score of the second frame of the driving image, the first confidence score and the second confidence score can be compared to obtain the train data with better confidence to improve the recognition accuracy of the train data.

[0057] S16: Determine target train data from the first train data and the second train data according to the first confidence score and the second confidence score.

[0058] According to the first confidence score and the second confidence score corresponding to the first train data and the second train data, the data accuracy of the first train data and the second train data can be judged, and the train data with higher accuracy can be determined as the target train data.

[0059] It should be noted that in this application, the extraction and recognition of two consecutive frames of driving images when the train passes through the lock are real-time and continuous, that is, after obtaining the first two frames of driving images, the subsequent recognized images between each two frames will be continuously obtained, thereby continuously updating the target train data, so that the accuracy of the train data continues to increase.

[0060] In an optional embodiment, determining the target train data from the first train data and the second train data according to the first confidence score and the second confidence score includes: scoring the first confidence score and the second confidence score; When the first confidence score is greater than the second confidence score, determining the first train data as the target train data; When the second confidence score is greater than the first confidence score, the second train data is determined to be the target train data.

[0061] After obtaining the first confidence score and the second confidence score, the first confidence score and the second confidence score are compared. When the first confidence score is greater than the second confidence score, it indicates that the first train data extracted from the first frame of the driving image is accurate, and the first train data is identified as the target train data. When the second confidence score is greater than the first confidence score, it indicates that the second train data extracted from the second frame of the driving image is accurate, and the second train data is identified as the target train data.

[0062] Through the above embodiment, the more accurate target train data in the train data of two adjacent frames is selected according to the confidence score.

[0063] S17, displaying the target train data.

[0064] After obtaining the target train data of two adjacent frames, the target train data and the driving pictures of the corresponding frames will be output and can be displayed in the form of data reports. The displayed content can include the text recognition data and image recognition data of the train.

[0065] Through the above-mentioned embodiments, the text and image recognition data of the train are displayed in the form of data reports, providing efficient and intuitive information support for users to monitor and analyze the status of the train.

[0066] In an optional embodiment, displaying the target train data includes: Generate a data report based on the target train data; A web page interface is generated based on the data report and the web page interface is displayed.

[0067] First, after acquiring the target train data of two adjacent frames, a detailed data report will be automatically generated based on the target train data. The data report can provide users with a comprehensive and accurate data overview of key information such as train number, number of carriages, and time stamp of passing the gate.

[0068] After the data report is generated, the system will further use the data to generate an interactive web interface. The web interface can be generated through HTML, CSS and JavaScript technology, allowing users to easily browse and understand the train data.

[0069] In the web interface, in addition to displaying the content of the data report, the system will also present the entire process of the train passing through the gate in a dynamic or static form based on information such as timestamps and driving images. This function can be achieved by embedding video players, picture carousels, etc., so that users can intuitively experience the actual situation of train operation.

[0070] In addition, the web page interface should also provide necessary interactive functions, such as data filtering, sorting, exporting, etc., to meet the different needs of users. At the same time, the system should also ensure the stability and security of the web page interface to prevent data leakage or illegal access.

[0071] In the above embodiment, the target train data is displayed to the user in an intuitive and easy-to-understand manner by generating data reports and web page interfaces, providing the user with convenient and efficient train data viewing, and the train status can be obtained based on the corresponding data.

[0072] In summary, the present application uses a line scan camera to obtain two consecutive frames of driving images when a train passes through a gate, identifies the first frame of the driving image to obtain the first train data, and determines the first confidence score of the first frame of the driving image, and reflects the accuracy of the first train data through the first confidence score. The second frame of the driving image is identified to obtain the second train data, and the second confidence score of the second frame of the driving image is determined, and the accuracy of the second train data is reflected through the second confidence score. By comparing the first confidence score with the second confidence score, the target train data with a higher confidence score is obtained and a data report is generated for display, thereby improving the accuracy of the train data, and displaying the accurate data intuitively, and reliable monitoring of the train status can be achieved through accurate train data.

[0073] See also Figure 2 The above is a functional module diagram of the train data identification and display device provided in an embodiment of the present application.

[0074] In some embodiments, the train data recognition and display device 20 may include a plurality of functional modules composed of computer program segments. The computer programs of the various program segments in the train data recognition and display device 20 may be stored in a memory of the train data recognition and display device and executed by at least one processor to perform (see Figure 1 Description) Function of the method for identifying and displaying train data.

[0075] In this embodiment, the train data recognition and display device 20 can be divided into multiple functional modules according to the functions it performs. The functional modules can include: an image acquisition module 201, a first recognition module 202, a first scoring module 203, a second recognition module 204, a second scoring module 205, a data determination module 206, and a data display module 207.

[0076] The image acquisition module 201 is used to acquire two consecutive frames of running images of a train passing through a gate through a line scan camera.

[0077] The first recognition module 202 is used to perform recognition based on the first frame of driving image to obtain first train data.

[0078] The first scoring module 203 is used to determine a first confidence score of the first frame of driving image according to the first train data.

[0079] The second recognition module 204 is used to perform recognition based on the second frame of driving image to obtain second train data.

[0080] The second scoring module 205 is used to determine a second confidence score of the second frame of driving image according to the second train data.

[0081] The data determination module 206 is configured to determine target train data from the first train data and the second train data according to the first confidence score and the second confidence score.

[0082] The data display module 207 is used to display the target train data.

[0083] Through the above functional modules, the train data can be accurately acquired and displayed when the train passes through the lock.

[0084] An embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, all or part of the steps of a method for identifying and displaying train data are implemented.

[0085] See also Figure 3 , which is a functional module diagram of an electronic device according to an embodiment of the present application. In a preferred embodiment of the present application, the electronic device 3 includes a memory 31, at least one processor 32, and at least one communication bus 33.

[0086] Those skilled in the art should understand that Figure 3 The structure of the electronic device shown does not constitute a limitation of the embodiments of the present application, and may be either a bus structure or a star structure. The printer may also include more or less other hardware or software than shown in the figure, or a different arrangement of components.

[0087] The embodiment of the present application also provides a computer-readable storage medium. The computer-readable storage medium is, for example, a non-volatile memory, such as: magnetic media (such as hard disks, floppy disks and tapes), optical media (such as CDROM disks and DVDs), magneto-optical media (such as optical disks) and hardware devices specially constructed to store and execute computer-executable instructions (such as read-only memory (ROM), random access memory (RAM), flash memory, etc.). Computer-executable instructions are stored on the computer-readable storage medium. The computer-readable storage medium can be used by one or more processors or processing devices to execute computer-executable instructions to implement the train data recognition and display method in the aforementioned embodiment.

[0088] In addition, it can be understood that the aforementioned embodiments are merely illustrative descriptions of the present application. Under the premise that the technical features do not conflict, the structures do not contradict, and the purpose of the invention of the present application is not violated, the technical solutions of the various embodiments can be arbitrarily combined and used in combination.

[0089] In the several embodiments provided in the present application, it should be understood that the disclosed modules and methods can be implemented in other ways. For example, the system or device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the system or unit can be electrical, mechanical or other forms.

[0090] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0091] In addition, each functional unit / module in each embodiment of the present application may be integrated into one processing unit / module, or each unit / module may exist physically separately, or two or more units / modules may be integrated into one unit / module. The above-mentioned integrated unit / module may be implemented in the form of hardware or in the form of hardware plus software functional units / modules.

[0092] The above-mentioned integrated unit / module implemented in the form of a software functional unit / module can be stored in a computer-readable storage medium. The above-mentioned software functional unit is stored in a storage medium, including a number of instructions for enabling one or more processors of a computer device (which can be a personal computer, a server, or a network device, etc.) to perform some steps of the methods of each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store program codes.

[0093] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for identifying and displaying train data, characterized in that: include: Use a line scan camera to obtain two consecutive frames of running images when the train passes through the gate; Perform recognition based on the first frame of the driving image to obtain first train data; Determining a first confidence score of the first frame of driving image according to the first train data; Perform recognition based on the second frame of the driving image to obtain second train data; Determining a second confidence score of the second frame of driving image according to the second train data; Determining target train data from the first train data and the second train data according to the first confidence score and the second confidence score; The target train data is displayed.

2. The train data identification and display method according to claim 1, characterized in that: The obtaining of the first train data by performing recognition based on the first frame of the driving image includes: Performing text recognition on the first frame of driving image to obtain first text recognition data; Performing image recognition on the first frame of driving image to obtain first image recognition data; The first text recognition data and the first image recognition data are used as the first train data.

3. The train data identification and display method according to claim 2, characterized in that: The performing text recognition on the first frame of driving image to obtain first text recognition data includes: Performing text detection and text positioning on the first frame of the driving image through a text detection network to obtain a first text region; Performing text recognition on the first text area through a convolutional recurrent neural network to obtain first text data; The first text recognition data is obtained by performing data matching between the first text data and a preset train database.

4. The train data identification and display method according to claim 3, characterized in that: The performing image recognition on the first frame of driving image to obtain first image recognition data includes: Performing unsupervised segmentation on the first frame of driving image to obtain a first frame of driving sub-image set; Performing Hall transformation detection on the first frame driving sub-image set to obtain a first target image set; Performing color space conversion on the first target image set to obtain a second target image set; Performing feature extraction on the second target image set to obtain a first feature information set; The first image recognition data is obtained by performing feature matching with the preset train database according to the first feature information set.

5. The train data identification and display method according to claim 4, characterized in that: Determining a first confidence score of the first frame of driving image according to the first train data includes: Performing matching degree recognition on the first text recognition data through a first preset matching degree recognition model to obtain a first train data matching degree; Performing matching degree recognition on the first image recognition data through a second preset matching degree recognition model to obtain a second train data matching degree; The first confidence score is obtained by performing a weighted calculation based on the first train data matching degree and the second train data matching degree.

6. The train data identification and display method according to claim 5, characterized in that: The determining the target train data from the first train data and the second train data according to the first confidence score and the second confidence score comprises: scoring the first confidence score and the second confidence score; When the first confidence score is greater than the second confidence score, determining the first train data as the target train data; When the second confidence score is greater than the first confidence score, the second train data is determined to be the target train data.

7. The train data identification and display method according to any one of claims 1 to 6, characterized in that: The displaying of the target train data comprises: Generate a data report based on the target train data; A web page interface is generated based on the data report and the web page interface is displayed.

8. A train data identification and display device, characterized in that: The device comprises: An image acquisition module is used to extract any two consecutive frames of running images from the running video of the train passing through the gate; A first recognition module, used for performing recognition based on a first frame of driving image to obtain first train data; A first scoring module, configured to determine a first confidence score of the first frame of driving image according to the first train data; A second recognition module is used to perform recognition based on the second frame of the driving image to obtain second train data; A second scoring module, configured to determine a second confidence score of the second frame of driving image according to the second train data; a data determination module, configured to determine target train data from the first train data and the second train data according to the first confidence score and the second confidence score; The data display module is used to display the target train data.

9. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the train data identification and display method as claimed in any one of claims 1 to 7 are implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for realizing train data recognition and display according to any one of claims 1 to 7 are implemented.

Citation Information

Patent Citations

  • A system for identifying the train number of urban rail train

    CN109241971A

  • Train number detection and recognition method based on deep learning

    CN110472633A

  • Information identification method, device and system, electronic equipment and storage medium

    CN111353434A

  • Image recognition method, device and equipment

    CN117636316A