Operation monitoring method and system of LED display screen
By using a pre-trained display abnormality recognition model, combined with image data and operating parameters, the accurate prediction and early warning of LED display faults is achieved, and the problem of fault detection delay in the existing technology is solved, which reduces maintenance costs and increases User satisfaction.
Patent Information
- Application Number
- CN202411413746.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-11
- Publication Date
- 2024-11-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The fault detection of existing LED display screens relies on manual experience and lacks effective fault prediction methods, resulting in delays in fault detection, high maintenance costs and reduced user satisfaction.
The pre-trained display abnormality recognition model is adopted, combined with image data and operating parameters (temperature, voltage, current), and the convolutional neural network and the time loop neural network are used to predict faults, generate fault prediction probability, and output fault warning information and Protection operation.
Accurate prediction of LED display failures is achieved, reducing maintenance difficulty and cost, and improving user satisfaction and equipment reliability.
Smart Images

Figure CN118918806A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of data processing technology, and in particular to an operation monitoring method and system for an LED display screen. Background Art
[0002] In existing LED display technology, display failure is a common problem. These failures are often unpredictable and bring a lot of inconvenience to the use of the display. Display failures may be caused by a variety of factors. These failures not only affect the normal operation of the display, but also lead to increased maintenance costs and reduced user satisfaction.
[0003] At present, the detection of display screen faults mainly relies on manual experience and technical support. Due to the lack of effective fault prediction methods, many faults are not discovered until after they occur, making the maintenance of LED screens difficult and costly. Summary of the invention
[0004] The technical problem to be solved by this application is to provide an operation monitoring method and system for an LED display screen, which can accurately obtain the failure prediction probability of the LED display screen. The specific scheme is as follows: A method for monitoring the operation of an LED display screen, comprising: Get the image to be displayed; Controlling the LED display screen to display the image; Using a preset image acquisition device to photograph the LED display screen, obtain image data of the LED display screen, and collect operating parameters of the LED display screen within a preset time period; the operating parameters include at least one of temperature, voltage and current; Using a pre-trained display screen abnormality recognition model based on the image data and the operating parameters, an abnormality recognition result of the LED display screen is obtained; the abnormality recognition result includes a failure prediction probability of the LED display screen; Output the abnormality recognition result of the LED display screen.
[0005] In the above method, optionally, the display screen abnormality recognition model includes a convolutional neural network, a time recurrent neural network and an output module; The method of using a pre-trained LED display screen abnormality recognition model to obtain an abnormality recognition result of the LED display screen based on the image data and the operating parameters includes: Preprocessing the image data and the operating parameters within the preset time period respectively; Inputting the preprocessed image data into the convolutional neural network to obtain a first eigenvector output by the convolutional neural network; Inputting the preprocessed operating parameters within the preset time period into the time recurrent neural network to obtain a second feature vector output by the time recurrent neural network; Merge the first eigenvector and the second eigenvector to obtain a eigenvector corresponding to the LED display screen; The feature vector corresponding to the LED display screen is input into the output module to obtain an abnormality recognition result of the LED display screen.
[0006] In the above method, optionally, the training process of the display screen abnormality recognition model includes: Obtain a training data set and an initial display screen abnormality recognition model to be trained; the training data set includes multiple training samples and a sample label corresponding to each training sample; each training sample includes historical image data of an LED display screen and historical operating parameters in a historical time period, and the sample label of each training sample is used to indicate whether the LED display screen has a fault; Selecting a target training sample currently used for training from the training data set; Inputting the target training sample into the initial display screen abnormality recognition model to obtain a recognition result corresponding to the target training sample; Based on a preset binary cross entropy loss function, the recognition result corresponding to the target training sample and the sample label, a loss function value of the initial display screen abnormality recognition model is calculated; the loss function value represents the degree of difference between the recognition result corresponding to the target training sample and the sample label; Updating the model parameters of the initial display screen abnormality recognition model according to the loss function value; When the initial display screen abnormality recognition model after updating the model parameters does not meet the preset training completion condition, returning to the step of selecting a target training sample currently used for training from the training data set; When the initial display screen abnormality recognition model after updating the model parameters satisfies the preset training completion condition, the initial display screen abnormality recognition model that satisfies the training completion condition is determined as the trained display screen abnormality recognition model.
[0007] The above method, optionally, after obtaining the abnormality recognition result of the LED display screen, further comprises: If the fault prediction probability is greater than a preset probability threshold, fault warning information of the LED display screen is generated; The fault warning information is output.
[0008] The above method, optionally, after generating the fault warning information of the LED display screen, further comprises: Detecting whether the temperature of the LED display screen is greater than a preset temperature threshold; When the temperature of the LED display screen is greater than a preset temperature threshold, a preset protection operation is performed; the protection operation includes at least one of reducing the brightness of the LED display screen and closing a portion of the display area of the LED display screen.
[0009] An operation monitoring system for an LED display screen, comprising: An acquisition unit, used for acquiring an image to be displayed; A control unit, used to control the LED display screen to display the image; A first execution unit is used to use a preset image acquisition device to shoot the LED display screen, obtain image data of the LED display screen, and collect operating parameters of the LED display screen within a preset time period; the operating parameters include at least one of temperature, voltage and current; A second execution unit is used to obtain an abnormality recognition result of the LED display screen based on the image data and the operating parameters using a pre-trained display screen abnormality recognition model; the abnormality recognition result includes a failure prediction probability of the LED display screen; The first output unit is used to output the abnormality recognition result of the LED display screen.
[0010] In the above system, optionally, the display screen abnormality recognition model includes a convolutional neural network, a time recurrent neural network and an output module; The second execution unit includes: A preprocessing subunit, used to preprocess the image data and the operating parameters within the preset time period respectively; A first execution subunit, used for inputting the preprocessed image data into the convolutional neural network to obtain a first feature vector output by the convolutional neural network; A second execution subunit is used to input the preprocessed operating parameters within the preset time period into the time recurrent neural network to obtain a second feature vector output by the time recurrent neural network; A third execution subunit is used to merge the first feature vector and the second feature vector to obtain a feature vector corresponding to the LED display screen; The fourth execution subunit is used to input the feature vector corresponding to the LED display screen into the output module to obtain the abnormality recognition result of the LED display screen.
[0011] In the above system, optionally, the second execution unit includes: An acquisition subunit is used to acquire a training data set and an initial display screen abnormality recognition model to be trained; the training data set includes a plurality of training samples and a sample label corresponding to each training sample; each training sample includes historical image data of an LED display screen and historical operating parameters in a historical time period, and the sample label of each training sample is used to indicate whether a fault occurs in the LED display screen; A selection subunit is used to select a target training sample currently used for training from the training data set; The recognition subunit is used to input the target training sample into the initial display screen abnormality recognition model to obtain a recognition result corresponding to the target training sample; A calculation subunit, configured to calculate a loss function value of the initial display screen abnormality recognition model based on a preset binary cross entropy loss function, the recognition result corresponding to the target training sample, and the sample label; the loss function value represents the degree of difference between the recognition result corresponding to the target training sample and the sample label; An updating subunit, used for updating the model parameters of the initial display screen abnormality recognition model according to the loss function value; A fifth execution subunit, configured to trigger the selection subunit to reselect a target training sample currently used for training from the training data set when the initial display screen abnormality recognition model after updating the model parameters does not meet a preset training completion condition; The determination subunit is used to determine the initial display screen abnormality recognition model that meets the preset training completion condition as the trained display screen abnormality recognition model when the initial display screen abnormality recognition model after updating the model parameters meets the preset training completion condition.
[0012] The above system may optionally further include: A generating unit, configured to generate fault warning information of the LED display screen if the fault prediction probability is greater than a preset probability threshold; The second output unit is used to output the fault warning information.
[0013] The above system may optionally further include: A detection unit, used to detect whether the temperature of the LED display screen is greater than a preset temperature threshold; The third execution unit is used to execute a preset protection operation when the temperature of the LED display screen is greater than a preset temperature threshold; the protection operation includes at least one of reducing the brightness of the LED display screen and closing a portion of the display area of the LED display screen.
[0014] Based on the above-mentioned implementation of the present application, a method and system for monitoring the operation of an LED display screen is provided, the method comprising: acquiring an image to be displayed; controlling the LED display screen to display the image; photographing the LED display screen using a preset image acquisition device to obtain image data of the LED display screen, and collecting operating parameters of the LED display screen within a preset time period; the operating parameters include at least one of temperature, voltage and current; obtaining an abnormal recognition result of the LED display screen based on the image data and the operating parameters using a pre-trained display screen abnormality recognition model; the abnormal recognition result includes a predicted probability of failure of the LED display screen; and outputting the abnormal recognition result of the LED display screen. By applying the method provided in the embodiment of the present application, the predicted probability of failure of the LED display screen can be accurately obtained, thereby helping users evaluate the health status of the equipment and the risk of possible failure in the future. By predicting the probability of possible failure of the display screen, users can take preventive measures before the failure actually occurs, effectively reducing the difficulty and cost of maintenance. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.
[0016] Figure 1 A method flow chart of an operation monitoring method of an LED display screen provided in this application; Figure 2 This is a structural example diagram of a display screen abnormality recognition model provided in this application; Figure 3 A flowchart of a process for obtaining abnormal identification results of an LED display screen provided by the present application; Figure 4 A flowchart of a training process of a display screen abnormality recognition model provided in this application; Figure 5 An example diagram of an implementation scenario provided for this application; Figure 6 A schematic diagram of the structure of an operation monitoring system for an LED display screen provided in this application. DETAILED DESCRIPTION
[0017] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0018] In this application, the terms "comprises", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element.
[0019] Based on this, the embodiment of the present application provides an operation monitoring method of an LED display screen, which can be applied to an electronic device, and the electronic device can be a computer, a controller of an LED display screen, a tablet, a smart phone or a smart wearable device, etc. The method flow chart of the method is as follows: Figure 1 As shown, specifically including: S101: Acquire an image to be displayed.
[0020] In this embodiment, the image to be displayed may be a preset test image, or may be any video frame in a video stream that the LED display needs to display.
[0021] S102: Control the LED display screen to display the image.
[0022] Optionally, a display instruction may be generated according to the image to be displayed, and the display instruction may be sent to the LED display screen so that the LED display screen displays the image.
[0023] In this embodiment, the number of the LED display screens may be one or more.
[0024] S103: photographing the LED display screen by using a preset image acquisition device to obtain image data of the LED display screen, and collecting operating parameters of the LED display screen within a preset time period; the operating parameters include at least one of temperature, voltage and current.
[0025] In this embodiment, the image acquisition device can be a camera, a camera or other device; the image acquisition device can be set in front of the LED display screen, and when the LED display screen displays an image, the LED display screen can be photographed to obtain image data of the LED display screen.
[0026] Optionally, the temperature of the LED display screen is collected by a temperature sensor; the voltage and current of the LED display screen can be collected by voltage and current sensors; wherein the number of temperature sensors can be one or more; when the number of temperature sensors is one, the temperature sensor can be deployed at any position of the LED display screen; when the number of temperature sensors is multiple, the area where the LED display screen is located can be divided into multiple temperature detection areas, and a temperature sensor is arranged at the center point of each temperature detection area.
[0027] In this embodiment, the preset time period may include the current time, which is the time for photographing the LED display screen. The length of the preset time period may be set to any length, for example, it may be set to 1 minute, 5 minutes, 1 hour, 1 day, etc.
[0028] S104: using a pre-trained display screen abnormality recognition model to obtain an abnormality recognition result of the LED display screen based on the image data and the operating parameters; the abnormality recognition result includes a fault prediction probability of the LED display screen.
[0029] In this embodiment, the fault prediction probability indicates the possibility of a failure of the LED display screen. The fault prediction probability can be a value in the interval of [0, 1], where 0 indicates that there is no possibility of a failure, and 1 indicates that a failure will definitely occur. Any value between 0 and 1 indicates the probability of a failure. The larger the value, the higher the possibility of a failure.
[0030] S105: Outputting the abnormality recognition result of the LED display screen.
[0031] Optionally, the abnormality recognition result of the LED display screen can be displayed on a preset monitoring interface.
[0032] By applying the method provided in the embodiment of the present application, decision makers can be helped to evaluate the health status of the equipment and the risk of possible failure in the future by outputting the failure prediction probability. By predicting the probability of failure of the LED display screen, preventive measures can be taken before the failure actually occurs.
[0033] In some embodiments, if there are multiple LED display screens, after obtaining the abnormality identification result of each LED display screen, the maintenance priority of each LED display screen can be determined according to the fault prediction probability of each LED display screen; maintenance plan information is generated according to the maintenance priority of each LED display screen, and the maintenance plan information is output; the maintenance plan information may include the LED display screens that need to be repaired and the maintenance order of each LED display screen that needs to be repaired.
[0034] In this embodiment, by outputting the maintenance plan information, the user can give priority to repairing LED display screens with a high probability of failure, so that the components that may fail can be replaced in advance.
[0035] By using the method provided in the embodiment of the present application, problems can be discovered early to avoid long downtime caused by sudden failures; and preventive maintenance is usually less expensive than emergency repairs.
[0036] Optionally, the probability of fault prediction can help maintenance personnel quickly locate the display screen or component that may have problems. Based on the fault identification results, high-risk displays can be prioritized and maintenance resources can be allocated reasonably. By monitoring the working status of the display screen, early wear caused by overuse can be avoided. Predictive maintenance can ensure that key components are replaced before failure, extending the life of the entire display system. In addition, display interruptions caused by faults can be reduced, improving the viewing experience.
[0037] In one embodiment provided in the present application, a display screen abnormality recognition model includes a convolutional neural network, a time recurrent neural network, and an output module.
[0038] Optional, see Figure 2 , is a structural example diagram of a display screen abnormality recognition model provided in an embodiment of the present application. The display screen abnormality recognition model may specifically include an input layer, a feature extraction layer, a feature fusion layer, and an output module. The output module includes a fully connected layer and an output layer. Among them, the feature extraction layer includes a convolutional neural network CNN and a time training neural network LSTM.
[0039] In some embodiments, the process of obtaining the abnormality recognition result of the LED display screen based on the image data and the operating parameters using the pre-trained LED display screen abnormality recognition model in S104 is as follows: Figure 3 As shown, including: S301: Preprocess the image data and the operating parameters within the preset time period respectively.
[0040] In this embodiment, the process of preprocessing the image data may include: cropping the image data to remove the image content other than the LED display screen in the image data, then converting the cropped image data into RGB format, and adjusting the size of the image data; normalizing the image data after the size adjustment, thereby completing the preprocessing of the image data, as follows:
[0041] In the formula, X1 is the original pixel value, x a : Normalized pixel value, the value 255 is the maximum value of the RGB color space.
[0042] Optionally, the size of the image data may be W×H×3, where W is the width and H is the height.
[0043] In this embodiment, the operation parameters may be pre-processed by normalizing the operation parameters within a preset time period, as follows: in, is any parameter among voltage, current and temperature, is the normalized parameter; The minimum value of the parameters in the preset time period; It is the maximum value among the parameters in the preset time period.
[0044] S302: Input the preprocessed image data into the convolutional neural network to obtain a first eigenvector output by the convolutional neural network.
[0045] Optionally, the first feature vector is a convolution feature of the image data.
[0046] S303: Input the preprocessed operating parameters within the preset time period into the time recurrent neural network to obtain a second feature vector output by the time recurrent neural network.
[0047] Optionally, the second feature vector may be a time feature of an operating parameter within a preset time period.
[0048] S304: Merge the first feature vector and the second feature vector to obtain a feature vector corresponding to the LED display screen.
[0049] In this embodiment, the first feature vector and the second feature vector may be input into a feature fusion layer to merge the first feature vector and the second feature vector to obtain a feature vector corresponding to the LED display screen.
[0050] S305: Inputting the feature vector corresponding to the LED display screen into the output module to obtain an abnormality recognition result of the LED display screen.
[0051] In this embodiment, the output module includes a fully connected layer and an output layer, and the output layer uses a Sigmoid activation function. The Sigmoid activation function is expressed as follows: Among them, z is the linear output of the fully connected layer (that is, the result of multiplying the weight by the input plus the bias term), and σ(z) is the output after the Sigmoid function.
[0052] Optionally, after obtaining the abnormality recognition result of the LED display screen, the image data of the LED display screen, the operating parameters within a preset time period, and the abnormality recognition result can be associated and stored, or the model parameters of the LED display screen abnormality recognition model can be optimized using the image data of the LED display screen, the operating parameters within a preset time period, and the abnormality recognition result.
[0053] In one embodiment provided in the present application, the training process of the display screen abnormality recognition model is as follows: Figure 4 As shown, including: S401: Obtain a training data set and an initial display screen abnormality recognition model to be trained; the training data set includes multiple training samples and a sample label corresponding to each training sample; each training sample includes historical image data of an LED display screen and historical operating parameters within a historical time period, and the sample label of each training sample is used to indicate whether a LED display screen fails.
[0054] Optionally, the length of the historical time period is consistent with the length of the above-mentioned preset time period.
[0055] In this embodiment, the historical image data of the LED display screen may be image data obtained by photographing the LED display screen during a historical period; the historical operating parameters may be one or more parameters such as temperature, voltage, current, etc. collected during the historical period.
[0056] Optionally, the historical time periods of different training samples may be different.
[0057] S402: Selecting target training samples currently used for training from the training data set.
[0058] In this embodiment, the target training sample currently used for training may be selected according to the arrangement order of the training samples in the training data set, or a training sample may be randomly selected from the training data set as the target training sample.
[0059] S403: Input the target training sample into the initial display screen abnormality recognition model to obtain a recognition result corresponding to the target training sample.
[0060] Optionally, the recognition result may be an abnormality prediction probability corresponding to the target training sample.
[0061] S404: Based on a preset binary cross entropy loss function, the recognition results corresponding to the target training samples and the sample labels, a loss function value of the initial display screen abnormality recognition model is calculated; the loss function value represents the degree of difference between the recognition results corresponding to the target training samples and the sample labels.
[0062] In this embodiment, the process of calculating the loss function value is as follows: in, represents the actual sample label (0 or 1), Represents the abnormal prediction probability corresponding to the target training sample.
[0063] S405: updating model parameters of the initial display screen abnormality recognition model according to the loss function value.
[0064] S406: Determine whether the initial display screen abnormality recognition model after updating the model parameters meets the preset training completion conditions; if not, return to execute S402; if so, execute S407.
[0065] In this embodiment, the training completion condition may be that the training times of the initial display screen abnormality model reaches a preset training times threshold, the loss function converges, or the prediction accuracy of the initial display screen abnormality model is greater than a preset accuracy threshold.
[0066] S407: Determine the initial display screen abnormality recognition model that meets the training completion condition as the trained display screen abnormality recognition model.
[0067] In an embodiment provided by the present application, after obtaining the abnormality recognition result of the LED display screen in S104, the method further includes: If the fault prediction probability is greater than a preset probability threshold, fault warning information of the LED display screen is generated; The fault warning information is output.
[0068] In this embodiment, the warning level of the LED display screen can be determined according to the fault prediction probability; a fault warning information template corresponding to the warning level is obtained; the identification, image data, operating parameters and fault prediction probability of the LED display screen are filled into the fault warning information template to obtain the fault warning information corresponding to the warning level; and the fault warning information is output.
[0069] In this embodiment, the fault warning information may be sent to a preset terminal device, or the fault warning information may be output and displayed in a preset monitoring interface.
[0070] In this embodiment, the warning level can be determined according to the fault prediction probability, thereby ensuring that LED display screens with a high probability of failure receive priority attention.
[0071] In one embodiment provided in the present application, if the fault prediction probability is greater than a preset probability threshold, the configuration file of the LED display screen can be backed up, and the configuration file may include one or more of resolution configuration, refresh rate configuration, brightness configuration, color difference calibration configuration, content playback configuration, run time configuration, access configuration, security configuration, fault recovery configuration, and maintenance configuration.
[0072] Optionally, after a LED display screen fails, the backed-up configuration file can be loaded into a new LED display screen to complete the rapid switching of the LED display screen; or after the failed LED display screen is repaired, the backed-up configuration file can be reloaded into the repaired LED display screen, thereby saving the configuration time of the LED display screen.
[0073] In an embodiment provided by the present application, after generating the fault warning information of the LED display screen, the method further includes: Detecting whether the temperature of the LED display screen is greater than a preset temperature threshold; When the temperature of the LED display screen is greater than a preset temperature threshold, a preset protection operation is performed; the protection operation includes at least one of reducing the brightness of the LED display screen and closing a portion of the display area of the LED display screen.
[0074] In this embodiment, when it is detected that the temperature of the LED display screen exceeds a preset temperature threshold, a protection operation can be performed to effectively reduce the power consumption of the LED display screen, thereby reducing the temperature of the LED display screen.
[0075] In this embodiment, the display parameters of the LED display screen can be adjusted according to the working state of the LED display screen, thereby ensuring the best viewing effect.
[0076] See also Figure 5 , is an example diagram of an implementation scenario provided in the present application, showing an implementation scenario including an LED display management center, N LED display screens, a sensor component corresponding to each LED display screen, and an image acquisition device, wherein N is a positive integer; the LED online screen management center may include a control device and a computer for displaying a monitoring interface.
[0077] Optionally, for each LED display screen, the control device can obtain an image to be displayed; control the LED display screen to display the image; use a preset image acquisition device to shoot the LED display screen to obtain image data of the LED display screen, and collect operating parameters of the LED display screen within a preset time period through a sensor acquisition component; the operating parameters include at least one of temperature, voltage and current; use a pre-trained display screen abnormality recognition model to obtain an abnormality recognition result of the LED display screen based on the image data and the operating parameters; the abnormality recognition result includes a failure prediction probability of the LED display screen; and control the computer device to output and display the abnormality recognition result of the LED display screen on an online monitoring interface.
[0078] Optionally, when the predicted failure probability of the LED display screen is greater than a preset failure probability threshold, the maintenance personnel may be notified by SMS, email, phone call, etc. to perform maintenance on the LED display screen.
[0079] and Figure 1 Corresponding to the method described above, the embodiment of the present application also provides an operation monitoring system of an LED display screen for monitoring Figure 1 The specific implementation of the method in the system is shown in the following figure. Figure 6 As shown, including: An acquisition unit 601 is used to acquire an image to be displayed; A control unit 602 is used to control the LED display screen to display the image; The first execution unit 603 is used to use a preset image acquisition device to shoot the LED display screen, obtain image data of the LED display screen, and collect operating parameters of the LED display screen within a preset time period; the operating parameters include at least one of temperature, voltage and current; The second execution unit 604 is used to obtain an abnormality recognition result of the LED display screen based on the image data and the operating parameters using a pre-trained display screen abnormality recognition model; the abnormality recognition result includes a fault prediction probability of the LED display screen; The first output unit 605 is used to output the abnormality recognition result of the LED display screen.
[0080] In an embodiment provided in the present application, based on the above solution, optionally, the display screen abnormality recognition model includes a convolutional neural network, a time recurrent neural network and an output module; Accordingly, the second execution unit 604 includes: A preprocessing subunit, used to preprocess the image data and the operating parameters within the preset time period respectively; A first execution subunit, used for inputting the preprocessed image data into the convolutional neural network to obtain a first feature vector output by the convolutional neural network; A second execution subunit is used to input the preprocessed operating parameters within the preset time period into the time recurrent neural network to obtain a second feature vector output by the time recurrent neural network; A third execution subunit is used to merge the first feature vector and the second feature vector to obtain a feature vector corresponding to the LED display screen; The fourth execution subunit is used to input the feature vector corresponding to the LED display screen into the output module to obtain the abnormality recognition result of the LED display screen.
[0081] In an embodiment provided in the present application, based on the above solution, optionally, the second execution unit includes: An acquisition subunit is used to acquire a training data set and an initial display screen abnormality recognition model to be trained; the training data set includes a plurality of training samples and a sample label corresponding to each training sample; each training sample includes historical image data of an LED display screen and historical operating parameters in a historical time period, and the sample label of each training sample is used to indicate whether a fault occurs in the LED display screen; A selection subunit is used to select a target training sample currently used for training from the training data set; The recognition subunit is used to input the target training sample into the initial display screen abnormality recognition model to obtain a recognition result corresponding to the target training sample; A calculation subunit, configured to calculate a loss function value of the initial display screen abnormality recognition model based on a preset binary cross entropy loss function, the recognition result corresponding to the target training sample, and the sample label; the loss function value represents the degree of difference between the recognition result corresponding to the target training sample and the sample label; An updating subunit, used for updating the model parameters of the initial display screen abnormality recognition model according to the loss function value; A fifth execution subunit, configured to trigger the selection subunit to reselect a target training sample currently used for training from the training data set when the initial display screen abnormality recognition model after updating the model parameters does not meet a preset training completion condition; The determination subunit is used to determine the initial display screen abnormality recognition model that meets the preset training completion condition as the trained display screen abnormality recognition model when the initial display screen abnormality recognition model after updating the model parameters meets the preset training completion condition.
[0082] In an embodiment provided in the present application, based on the above solution, optionally, it further includes: A generating unit, configured to generate fault warning information of the LED display screen if the fault prediction probability is greater than a preset probability threshold; The second output unit is used to output the fault warning information.
[0083] In an embodiment provided in the present application, based on the above solution, optionally, it further includes: A detection unit, used to detect whether the temperature of the LED display screen is greater than a preset temperature threshold; The third execution unit is used to execute a preset protection operation when the temperature of the LED display screen is greater than a preset temperature threshold; the protection operation includes at least one of reducing the brightness of the LED display screen and closing a portion of the display area of the LED display screen.
[0084] The specific principles and execution processes of each unit and module in the operation monitoring system of the LED display screen disclosed in the above-mentioned embodiment of the present application are the same as the operation monitoring method of the LED display screen disclosed in the above-mentioned embodiment of the present application. Please refer to the corresponding parts of the operation monitoring method of the LED display screen provided in the above-mentioned embodiment of the present application, and will not be repeated here.
[0085] It should be noted that each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.
[0086] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations.
[0087] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.
[0088] It can be known from the description of the above implementation methods that those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present application or certain parts of the embodiments.
[0089] The above is a detailed introduction to the operation monitoring method of an LED display screen provided by the present application. This article uses specific examples to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, according to the idea of the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.
Claims
1. A method for monitoring the operation of an LED display screen, characterized in that: include: Get the image to be displayed; Controlling the LED display screen to display the image; Using a preset image acquisition device to photograph the LED display screen, obtain image data of the LED display screen, and collect operating parameters of the LED display screen within a preset time period; the operating parameters include at least one of temperature, voltage and current; Using a pre-trained display screen abnormality recognition model based on the image data and the operating parameters, an abnormality recognition result of the LED display screen is obtained; the abnormality recognition result includes a failure prediction probability of the LED display screen; Output the abnormality recognition result of the LED display screen.
2. The method according to claim 1, characterized in that The display screen abnormality recognition model includes a convolutional neural network, a time recurrent neural network and an output module; The method of using a pre-trained LED display screen abnormality recognition model to obtain an abnormality recognition result of the LED display screen based on the image data and the operating parameters includes: Preprocessing the image data and the operating parameters within the preset time period respectively; Inputting the preprocessed image data into the convolutional neural network to obtain a first eigenvector output by the convolutional neural network; Inputting the preprocessed operating parameters within the preset time period into the time recurrent neural network to obtain a second feature vector output by the time recurrent neural network; Merge the first eigenvector and the second eigenvector to obtain a eigenvector corresponding to the LED display screen; The feature vector corresponding to the LED display screen is input into the output module to obtain an abnormality recognition result of the LED display screen.
3. The method according to claim 1, characterized in that The training process of the display screen abnormality recognition model includes: Obtain a training data set and an initial display screen abnormality recognition model to be trained; the training data set includes multiple training samples and a sample label corresponding to each training sample; each training sample includes historical image data of an LED display screen and historical operating parameters in a historical time period, and the sample label of each training sample is used to indicate whether the LED display screen has a fault; Selecting a target training sample currently used for training from the training data set; Inputting the target training sample into the initial display screen abnormality recognition model to obtain a recognition result corresponding to the target training sample; Based on a preset binary cross entropy loss function, the recognition result corresponding to the target training sample and the sample label, a loss function value of the initial display screen abnormality recognition model is calculated; the loss function value represents the degree of difference between the recognition result corresponding to the target training sample and the sample label; Updating the model parameters of the initial display screen abnormality recognition model according to the loss function value; When the initial display screen abnormality recognition model after updating the model parameters does not meet the preset training completion condition, returning to the step of selecting a target training sample currently used for training from the training data set; When the initial display screen abnormality recognition model after updating the model parameters satisfies the preset training completion condition, the initial display screen abnormality recognition model that satisfies the training completion condition is determined as the trained display screen abnormality recognition model.
4. The method according to claim 1, characterized in that After obtaining the abnormal identification result of the LED display screen, the method further includes: If the fault prediction probability is greater than a preset probability threshold, fault warning information of the LED display screen is generated; The fault warning information is output.
5. The method according to claim 4, characterized in that After generating the fault warning information of the LED display screen, the method further includes: Detecting whether the temperature of the LED display screen is greater than a preset temperature threshold; When the temperature of the LED display screen is greater than a preset temperature threshold, a preset protection operation is performed; the protection operation includes at least one of reducing the brightness of the LED display screen and closing a portion of the display area of the LED display screen.
6. An operation monitoring system for an LED display screen, characterized in that: include: An acquisition unit, used for acquiring an image to be displayed; A control unit, used to control the LED display screen to display the image; A first execution unit is used to use a preset image acquisition device to shoot the LED display screen, obtain image data of the LED display screen, and collect operating parameters of the LED display screen within a preset time period; the operating parameters include at least one of temperature, voltage and current; A second execution unit is used to obtain an abnormality recognition result of the LED display screen based on the image data and the operating parameters using a pre-trained display screen abnormality recognition model; the abnormality recognition result includes a failure prediction probability of the LED display screen; The first output unit is used to output the abnormality recognition result of the LED display screen.
7. The system according to claim 6, characterized in that The display screen abnormality recognition model includes a convolutional neural network, a time recurrent neural network and an output module; The second execution unit includes: A preprocessing subunit, used to preprocess the image data and the operating parameters within the preset time period respectively; A first execution subunit, used for inputting the preprocessed image data into the convolutional neural network to obtain a first feature vector output by the convolutional neural network; A second execution subunit is used to input the preprocessed operating parameters within the preset time period into the time recurrent neural network to obtain a second feature vector output by the time recurrent neural network; A third execution subunit is used to merge the first feature vector and the second feature vector to obtain a feature vector corresponding to the LED display screen; The fourth execution subunit is used to input the feature vector corresponding to the LED display screen into the output module to obtain the abnormality recognition result of the LED display screen.
8. The system according to claim 6, characterized in that The second execution unit includes: An acquisition subunit is used to acquire a training data set and an initial display screen abnormality recognition model to be trained; the training data set includes a plurality of training samples and a sample label corresponding to each training sample; each training sample includes historical image data of an LED display screen and historical operating parameters in a historical time period, and the sample label of each training sample is used to indicate whether a fault occurs in the LED display screen; A selection subunit is used to select a target training sample currently used for training from the training data set; The recognition subunit is used to input the target training sample into the initial display screen abnormality recognition model to obtain a recognition result corresponding to the target training sample; A calculation subunit, configured to calculate a loss function value of the initial display screen abnormality recognition model based on a preset binary cross entropy loss function, the recognition result corresponding to the target training sample, and the sample label; the loss function value represents the degree of difference between the recognition result corresponding to the target training sample and the sample label; An updating subunit, used for updating the model parameters of the initial display screen abnormality recognition model according to the loss function value; A fifth execution subunit, configured to trigger the selection subunit to reselect a target training sample currently used for training from the training data set when the initial display screen abnormality recognition model after updating the model parameters does not meet a preset training completion condition; The determination subunit is used to determine the initial display screen abnormality recognition model that meets the preset training completion condition as the trained display screen abnormality recognition model when the initial display screen abnormality recognition model after updating the model parameters meets the preset training completion condition.
9. The system according to claim 6, characterized in that Also includes: A generating unit, configured to generate fault warning information of the LED display screen if the fault prediction probability is greater than a preset probability threshold; The second output unit is used to output the fault warning information.
10. The system according to claim 9, characterized in that Also includes: A detection unit, used to detect whether the temperature of the LED display screen is greater than a preset temperature threshold; The third execution unit is used to execute a preset protection operation when the temperature of the LED display screen is greater than a preset temperature threshold; the protection operation includes at least one of reducing the brightness of the LED display screen and closing a portion of the display area of the LED display screen.
Citation Information
Patent Citations
Circuit working state detection method and system
CN107192939A
Display panel defect classification method and related device
CN112884712A
Display screen control method and device, electronic equipment and storage medium
CN115331605A
Performance detection method, device and equipment of liquid crystal display screen and storage medium
CN117423299A
Screen anomaly detection method, device and equipment and readable storage medium
CN117906919A
Cited By
Control method and system of outdoor LED display screen
CN119152803A
A control method and system for outdoor LED display screen
CN119152803B
Outdoor large screen monitoring method and device based on non-visual perception and medium
CN120223836A
