Method, device and equipment for repairing defective pixels of LED display screen

Abstract

The invention relates to the field of display screen pixel repairing, in particular to an LED display screen pixel defective pixel repairing method, device and equipment. The method comprises the following steps: acquiring pixel point multi-dimensional operation monitoring parameters of the LED display screen; carrying out pixel dead pixel identification according to the pixel multi-dimensional operation monitoring parameters, and marking all pixel dead pixels; carrying out dead pixel position positioning calculation on all the pixel dead pixels, and carrying out regional color temperature distribution feature mining to generate color temperature distribution features of each to-be-repaired region; according to the color temperature distribution characteristics of each to-be-repaired area, defective pixel repair value calculation is carried out one by one, adaptive color repair processing is carried out, and repaired display screen pixel point parameters are collected; and according to the repaired display screen pixel point parameters, performing repair area brightness non-uniform node identification, and then performing local brightness uniform compensation, thereby constructing a local brightness compensation strategy. According to the invention, efficient and accurate pixel dead pixel restoration is realized.

Description

Technical Field

[0001] The present invention relates to the field of display screen pixel repair, and in particular to a method, device and equipment for repairing bad pixels of an LED display screen. Background Art

[0002] With the continuous development of LED display technology, LED display screens are widely used in various fields such as advertising, display, and monitoring due to their advantages such as bright colors, high brightness, energy saving and environmental protection. The quality of the pixels of LED display screens directly affects the display effect and perception. Therefore, the health of the pixels plays a vital role in the stability of the display system and the visual experience. Especially during long-term use, due to factors such as voltage fluctuations, temperature changes, external vibrations, and aging, the pixels of LED display screens may fail, manifested as bad pixels, dead pixels, overly bright pixels, or color anomalies. These problems will not only affect the display effect, but may also seriously affect the overall service life and reliability of the screen.

[0003] Traditional methods of repairing bad pixels on LED display screens rely on human intervention and simple manual repair, usually by replacing bad pixels or adjusting the brightness locally. However, this method has the disadvantages of low repair accuracy, low efficiency and high labor costs. Especially in large-scale LED display systems, manual repair is not only difficult to meet the needs of real-time repair, but also prone to uneven repair due to human factors, affecting the overall display effect of the screen.

[0004] However, existing technologies mostly focus on single pixel repair or brightness adjustment, lacking a comprehensive, dynamically adjusted intelligent repair solution. Especially in the continuous operation of large-scale display screens, pixel failures often present complex and changeable characteristics. Therefore, there is an urgent need for a comprehensive LED display screen pixel repair method that integrates real-time monitoring, intelligent identification, automatic repair and predictive diagnosis to meet the needs of modern display systems for efficient, accurate, continuous and stable operation. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a method, device and equipment for repairing bad pixels of LED display screens to solve at least one of the above technical problems.

[0006] To achieve the above object, the present invention provides a method for repairing bad pixels of LED display screen, comprising the following steps: Step S1: obtaining multi-dimensional operation monitoring parameters of pixels of the LED display screen; identifying bad pixels according to the multi-dimensional operation monitoring parameters of the pixels, and marking all bad pixels; Step S2: Calculate the position of all bad pixels and mine the regional color temperature distribution features to generate the color temperature distribution features of each area to be repaired; Step S3: Calculate the bad pixel repair value one by one according to the color temperature distribution characteristics of each area to be repaired, perform adaptive color repair processing, and collect the pixel parameters of the repaired display screen; Step S4: Identify nodes with uneven brightness in the repair area according to the repaired display pixel parameters, and then perform local brightness uniformity compensation, thereby constructing a local brightness compensation strategy; Step S5: mining dynamic photoelectric change characteristics according to the multi-dimensional operation monitoring parameters of the pixel points, identifying photoelectric situation mutations, and locating the photoelectric situation mutation pixel points; Step S6: Pixel fault prediction is performed for pixels with sudden photoelectric state changes, and pixel bad point repair is iteratively optimized according to the local brightness compensation strategy to build an intelligent pixel prediction and repair engine.

[0007] In this specification, a device for repairing bad pixel points of an LED display screen is provided, which is used to perform the above-mentioned method for repairing bad pixel points of an LED display screen, comprising: A bad pixel identification module is used to obtain multi-dimensional operation monitoring parameters of pixels of the LED display screen; identify bad pixels according to the multi-dimensional operation monitoring parameters of the pixels, and mark all bad pixels; The bad pixel location module is used to calculate the bad pixel locations of all bad pixels and mine the regional color temperature distribution features to generate the color temperature distribution features of each area to be repaired; The adaptive repair module is used to calculate the repair value of each bad pixel one by one according to the color temperature distribution characteristics of each area to be repaired, and adaptively perform color repair processing, and collect the pixel parameters of the repaired display screen; A local brightness compensation module is used to identify nodes with uneven brightness in the repair area according to the repaired display pixel parameters, and then perform local brightness uniformity compensation, thereby constructing a local brightness compensation strategy; The photoelectric characteristic module is used to mine the dynamic photoelectric change characteristics according to the multi-dimensional operation monitoring parameters of the pixel points, identify the photoelectric situation mutation, and locate the photoelectric situation mutation pixel points; The iterative optimization module is used to predict pixel faults for pixels with sudden photoelectric state changes, and iteratively optimize pixel bad pixel repair based on the local brightness compensation strategy to build an intelligent pixel prediction and repair engine.

[0008] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the method for repairing bad pixels of LED display screens described in any one of the above are implemented.

[0009] The beneficial effects of the present invention are as follows: by acquiring the multi-dimensional operating parameters of the display screen (such as brightness, current, color deviation, temperature, etc.), the operating status of each pixel can be fully monitored. This provides an accurate data basis for the subsequent identification of bad pixels, avoiding the limitations of traditional methods that rely only on simple brightness or color analysis. The system realizes accurate bad pixel identification based on real-time monitoring parameters, and can mark all bad pixels, including dead pixels, abnormal color difference points, over-bright pixels, etc. This ensures that all pixels with problems are marked and processed in a timely manner, improving the comprehensiveness and accuracy of the repair. Each bad pixel is located to ensure the accuracy of bad pixel repair. Unlike the traditional method of relying on visual observation for repair, the algorithm-based precise positioning can greatly reduce erroneous repairs and avoid interference with fault-free pixels. By analyzing the color temperature distribution characteristics of the area where each bad pixel is located, the system can accurately identify the color temperature differences and lighting environment differences in the repair area. Through this process, the repair work is more in line with the overall color standard of the display screen, reducing the occurrence of unevenness or distortion after repair. Through the adaptive color repair algorithm, the system can dynamically adjust the repair value according to the specific situation of each bad pixel, ensuring that the repair process does not rely on fixed parameters, but responds to the needs of the display in real time. This dynamic repair method can effectively avoid repair errors caused by improper manual settings. During the repair process, the system will collect the data of the repaired display screen, evaluate the effect of the repair, ensure that the repair effect of each pixel meets the high quality standards, and improve the consistency and accuracy of the display effect. Through real-time analysis of the photoelectric change characteristics, the sudden change of the pixel point during the working process can be accurately identified. This process can detect potential faults or abnormal photoelectric fluctuations in advance, and buy time for timely repair. The system can quickly and accurately locate the pixel points where the photoelectric state suddenly changes, avoiding the expansion of display problems caused by the failure to find the fault point in time in the traditional method, and enhancing the predictability and response speed of fault identification. Through the intelligent pixel fault prediction algorithm, the system can predict the occurrence time of the fault before the bad pixel appears, which provides effective decision support for timely repair. The application of pre-repair measures reduces the impact caused by the failure to repair the pixel point in time, making the maintenance of the display screen more proactive rather than passive. Through the iteratively optimized intelligent pixel prediction repair engine, the system can continuously adjust the repair strategy based on historical repair data and real-time monitoring results to ensure that the repair method can maintain efficiency and accuracy in different environments and usage conditions. The feedback data after each repair helps the system continuously optimize the repair process, making the repair effect more and more ideal. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic diagram of the steps of a method for repairing bad pixels on an LED display screen according to the present invention; Figure 2 Detailed implementation flow chart of step S1; Figure 3 Detailed implementation flow chart of step S2; Figure 4 Detailed implementation flow chart of step S3. DETAILED DESCRIPTION

[0011] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0012] The present application example provides a method, device and equipment for repairing bad pixels on an LED display screen. The execution subjects of the method, device and equipment for repairing bad pixels on an LED display screen include but are not limited to: mechanical equipment, data processing platform, cloud server node, network upload equipment, etc. equipped with the system can be regarded as the general computing nodes of the present application, and the data processing platform includes but is not limited to: at least one of an audio and image management system, an information management system, and a cloud data management system.

[0013] See also Figures 1 to 4 The present invention provides a method for repairing bad pixels of LED display screens, and the method for repairing bad pixels of LED display screens comprises the following steps: Step S1: obtaining multi-dimensional operation monitoring parameters of pixels of the LED display screen; identifying bad pixels according to the multi-dimensional operation monitoring parameters of the pixels, and marking all bad pixels; Step S2: Calculate the position of all bad pixels and mine the regional color temperature distribution features to generate the color temperature distribution features of each area to be repaired; Step S3: Calculate the bad pixel repair value one by one according to the color temperature distribution characteristics of each area to be repaired, perform adaptive color repair processing, and collect the pixel parameters of the repaired display screen; Step S4: Identify nodes with uneven brightness in the repair area according to the repaired display pixel parameters, and then perform local brightness uniformity compensation, thereby constructing a local brightness compensation strategy; Step S5: mining dynamic photoelectric change characteristics according to the multi-dimensional operation monitoring parameters of the pixel points, identifying photoelectric situation mutations, and locating the photoelectric situation mutation pixel points; Step S6: Pixel fault prediction is performed for pixels with sudden photoelectric state changes, and pixel bad point repair is iteratively optimized according to the local brightness compensation strategy to build an intelligent pixel prediction and repair engine.

[0014] In the embodiment of the present invention, refer to Figure 1 , is a schematic flow chart of the steps of a method for repairing bad pixel points on an LED display screen according to the present invention. In this example, the steps of the method for repairing bad pixel points on an LED display screen include: Step S1: obtaining multi-dimensional operation monitoring parameters of pixels of the LED display screen; identifying bad pixels according to the multi-dimensional operation monitoring parameters of the pixels, and marking all bad pixels; In this embodiment, multi-dimensional information such as brightness, color, temperature, humidity, current and voltage of the pixel is obtained. In order to ensure the accuracy and real-time performance of the data, a reasonable parameter acquisition frequency is set. In general, the acquisition frequency can be set to once per second, so that the rapidly changing state can be captured. For some environmental parameters such as temperature and humidity, it can be set to be collected once every 5 seconds to reduce data redundancy. During the acquisition process, the unit of each parameter is recorded, the brightness is expressed in candela per square meter (cd / m²), the current is expressed in milliamperes (mA), and the temperature is expressed in degrees Celsius (°C). All collected parameters should be stored in a database or data storage system in real time for subsequent analysis. Each piece of data can be marked with a timestamp to track the time series of parameter changes. The collected real-time parameters are initially filtered and cleaned by data processing software to remove outliers and noise. This process can use simple statistical methods, such as calculating the mean and standard deviation, to judge the rationality of the data. Develop a real-time monitoring interface to display the collected real-time parameters in a graphical way. The interface should include real-time data charts, alarm systems, and historical data records to help technicians quickly identify the operating status of the display screen. The interface can use different colors and charts to identify normal and abnormal states. When the brightness value is low, it will be displayed in red, indicating that there may be a bad pixel. Establish the identification standard of the bad pixel and set the threshold according to the normal operating parameters of the display. If the brightness of a pixel is lower than the set lower limit (such as 50 cd / m²) or higher than the upper limit (such as 800 cd / m²), the pixel is considered a bad pixel. In addition, parameters such as color deviation and response time should also be considered and corresponding standards should be set. If the RGB value of a pixel deviates from the normal value by more than the set threshold (such as ±10%), it should also be marked as a bad pixel. Use data analysis tools to analyze the parameters collected in real time and check each pixel one by one. You can use the traversal algorithm to check the brightness and color value of each pixel row by row and column by column. For each pixel, calculate its difference with the surrounding pixels to assist in determining whether there is a bad pixel. If the brightness value of a pixel differs from the average value of the surrounding four pixels by more than the set threshold, the pixel is marked as a bad pixel. Once a bad pixel is identified, it is immediately marked in the system. For each bad pixel, its coordinates (such as (x, y)), brightness value, color value and its status (such as "normal" or "bad pixel") can be recorded. All marked bad pixel information should be stored in the database to form a detailed record of the bad pixels for subsequent analysis and repair. Generate a visual image of the display screen and mark the location of all bad pixels on the image. Use different colors or symbols to represent normal pixels and bad pixels so that technicians can quickly identify and locate problem areas. Generate a bad pixel distribution map through visualization tools to help technicians understand the spatial distribution characteristics of bad pixels and formulate corresponding maintenance plans.

[0015] Step S2: Calculate the position of all bad pixels and mine the regional color temperature distribution features to generate the color temperature distribution features of each area to be repaired; In this embodiment, the coordinates of all bad pixels (such as (x, y)) are extracted from the bad pixel information collected in the previous step. These coordinates are the specific locations of the bad pixels on the display screen, forming a bad pixel location list. For each bad pixel, its brightness, color and state are recorded to ensure the integrity of the data. According to the distribution of the bad pixels, the display screen is divided into multiple areas to be repaired. The size of each area can be set to divide the display screen into small areas of 10x10 pixels. By traversing all the bad pixels, each bad pixel is classified into its area for subsequent analysis. For each divided area, the number of bad pixels in the area is calculated. According to the density value, the repair priority of each area can be determined, and the area with a high bad pixel density needs to be repaired first. The bad pixel location and related information of each area to be repaired are recorded in the database, including the area coordinates, the number of bad pixels and the bad pixel density. A detailed record of the area to be repaired is formed to provide a basis for the subsequent mining of color temperature distribution features. For each area to be repaired, the color temperature characteristics of the area need to be calculated. Traverse each area to be repaired and calculate the average color temperature value of all pixels in the area. The RGB values ​​of all pixels in the area are added together to obtain the total RGB value in the area, and then the average value is calculated and converted to color temperature. The standard deviation of the color temperature can also be calculated to evaluate the uniformity of the color temperature in the area. Areas with smaller standard deviations indicate that the color temperature distribution is more uniform, and vice versa, it indicates that the color temperature distribution is poor. The color temperature characteristics (average color temperature and standard deviation) of each area to be repaired are recorded in the database. A color temperature distribution characteristic report is formed to help understand the color temperature distribution in different areas. These color temperature characteristics are visualized and the color temperature distribution of each area is displayed through a color temperature heat map. Dark colors indicate low color temperatures and light colors indicate high color temperatures. This visualization helps technicians quickly understand the color temperature distribution characteristics. According to the color temperature distribution characteristics, combined with the density of bad pixels, the repair priority of each area is determined. Areas with uneven color temperature distribution and dense bad pixels should be repaired first to ensure an overall improvement in the display effect.

[0016] Step S3: Calculate the bad pixel repair value one by one according to the color temperature distribution characteristics of each area to be repaired, perform adaptive color repair processing, and collect the pixel parameters of the repaired display screen; In this embodiment, the color temperature distribution characteristics of each area to be repaired are extracted from the previous step, including the average color temperature and the standard deviation. The clear goal is to adjust the color temperature of the bad pixel to the average color temperature value of the area. Assuming that the average color temperature of a certain area is 6500K and the standard deviation is 500K, it means that the color temperature of the area fluctuates between 6000K and 7000K. For each bad pixel, calculate the deviation between its current color temperature value and the target color temperature value (i.e., the average color temperature of the area). Assuming that the color temperature of a bad pixel is 5800K, the deviation is: deviation = target color temperature - current color temperature = 6500K - 5800K = 700K, and convert the deviation value into a specific repair value. Set a proportional coefficient (such as 0.5) to control the repair amplitude. The repair value calculation formula can be set as: repair value = deviation × proportional coefficient = 700K × 0.5 = 350K, and select a suitable color repair algorithm, such as a color adjustment algorithm based on a color wheel or an HSV (hue, saturation, brightness) model. The HSV model is suitable for color adjustment because it can more intuitively control hue and saturation. In HSV space, adjust the hue value to match the target color temperature to make the color look more natural. Traverse all marked bad pixels and perform adaptive color adjustment for each bad pixel according to its repair value. If the hue value of a bad pixel is 30° and the repair value is 350K, its hue value can be adjusted to: new hue = current hue + (repair value 1000×360) Through this method, ensure that the color of each bad pixel gradually approaches the color temperature target of the area. After completing the color adjustment, remeasure and record the color parameters of each pixel, including RGB value and brightness value. These parameters will be used for subsequent analysis to ensure that the repair effect meets the expectations. The RGB value of the repaired pixel is adjusted from (200, 150, 100) to (210, 160, 110), and these changes are recorded. Determine the criteria for evaluating the repair effect, including color temperature uniformity, brightness consistency, and visual effect. A threshold can be set, and the standard deviation of color temperature uniformity should be less than 100K. After the repair is completed, re-evaluate the color temperature distribution characteristics of each area to be repaired, calculate the new mean and standard deviation, and compare and analyze with the data before the repair. If the standard deviation after the repair is smaller than the previous standard deviation, the repair is considered successful. Record the pixel parameters of the repaired display (RGB value, brightness, color temperature, etc.) in the database, and generate a visualization chart to show the comparison before and after the repair. Show the improvement of color temperature distribution through a heat map or a bar chart. This visualization not only helps to evaluate the repair effect, but also provides data support for subsequent maintenance and optimization.

[0017] Step S4: Identify nodes with uneven brightness in the repair area according to the repaired display pixel parameters, and then perform local brightness uniformity compensation, thereby constructing a local brightness compensation strategy; In this embodiment, the pixel parameters of the repaired display screen are extracted from the previous step, including the brightness value and color information of each pixel. Ensure the integrity and accuracy of the data, and record the coordinates, brightness and color values ​​of each pixel for subsequent analysis. Determine the standard for brightness uniformity identification. Usually, the standard deviation is used to measure the uniformity of brightness. Set a threshold, and when the brightness standard deviation of a certain area exceeds 5%, it is marked as a brightness uneven node. Calculate the brightness uniformity of each area to be repaired, set the brightness uniformity evaluation index, traverse each pixel, and calculate the difference between its brightness and the average brightness of the surrounding pixels. For each pixel, calculate the difference between its brightness and the surrounding pixels (such as 4 adjacent pixels). If the difference exceeds the set threshold, it is marked as an uneven node. If the brightness of a pixel is 200 cd / m², and the average brightness of the surrounding pixels is 150 cd / m², the pixel is considered to have uneven brightness. The identified brightness uneven nodes are recorded in the database, including their coordinates, brightness values ​​and their deviations. Generate a visual chart to show the spatial distribution of nodes with uneven brightness. Use color coding to identify different degrees of uneven brightness to help technicians quickly locate the problem area. For each node with uneven brightness, calculate its compensation value. The compensation value can be determined based on the brightness difference between the pixel and the surrounding pixels. Set the compensation formula as follows: compensation value = target brightness − current brightness. The target brightness can be the average brightness of the surrounding normal pixels to ensure that the compensation effect is natural. For each marked node with uneven brightness, adjust its brightness according to the calculated compensation value. If the current brightness of a node is 100 cd / m² and the target brightness is 150 cd / m², the compensation value is 50 cd / m², and the brightness of the pixel needs to be increased by 50 cd / m². During the compensation process, ensure that the compensated brightness does not exceed the maximum brightness limit of the display (such as 800 cd / m²) to avoid visual discomfort caused by over-compensation. After completing the local brightness compensation, re-measure and record the brightness value of each pixel, and calculate the brightness standard deviation and uniformity index of the compensated area. Compare with the data before compensation to evaluate whether the compensation effect meets the standard. If the standard deviation of regional brightness before compensation is 10 cd / m² and it is reduced to 3 cd / m² after compensation, it indicates that the compensation effect is good. The compensated brightness parameters are recorded in the database, and a visual chart is generated to show the brightness changes before and after compensation. This process not only helps to evaluate the compensation effect, but also provides data support for subsequent maintenance. According to the evaluation results of the compensation effect, a local brightness compensation strategy is formulated. The strategy should clarify the priority of compensation and give priority to compensation for areas with more nodes of uneven brightness. Formulate the frequency and method of compensation, regularly evaluate the brightness uniformity of the display screen, and compensate in time when problems are found. In actual operation, apply the formulated compensation strategy to regularly check and compensate the brightness uniformity of the display screen. Record the time, area and effect of each compensation for subsequent analysis and optimization.The monitoring system tracks the brightness changes of the display in real time to ensure the effectiveness and adaptability of the compensation strategy. Based on the implementation results and actual feedback, the local brightness compensation strategy is continuously optimized to ensure that it adapts to the needs of different environments and usage conditions. Regular effect evaluation is carried out to ensure that the brightness uniformity of the display meets expectations and improve the user's visual experience.

[0018] Step S5: mining dynamic photoelectric change characteristics according to the multi-dimensional operation monitoring parameters of the pixel points, identifying photoelectric situation mutations, and locating the photoelectric situation mutation pixel points; In this embodiment, the multi-dimensional operation monitoring parameters of the pixel points are obtained from the previous steps, including data such as brightness, color, temperature, and humidity. These parameters should be recorded in a time series manner to ensure that the change of each parameter has a corresponding timestamp. The data acquisition frequency is set, usually once per second, in order to capture the rapidly changing photoelectric characteristics. During the data acquisition process, the brightness value (unit: cd / m²), color value (RGB) and temperature (unit: °C) of each pixel are recorded. A dynamic characteristic calculation model is established for the collected time series data. The sliding window method is usually used to calculate the changes of each pixel within a certain time range. Set the sliding window to 10 seconds, and calculate the average brightness, color change rate and temperature change rate of each pixel in this time period. Record the change characteristics of each pixel, and store the calculation results in the database to form a dynamic photoelectric characteristic data set. Analyze the extracted dynamic photoelectric change characteristics and calculate the change trend of each pixel. The linear regression slope of the brightness change can be calculated to determine whether the brightness is rising, falling or remaining stable. Generate a visual chart to show the changes in the photoelectric characteristics of each pixel at different time points. Use line graphs to show the changing trends of brightness, color and temperature, helping technicians to quickly understand the photoelectric characteristics. According to the dynamic photoelectric change characteristics, set the recognition standard of photoelectric situation mutation. The threshold method and change rate method are usually used to identify mutation phenomena. Set the change rate threshold. If the brightness change rate of a pixel exceeds the set value (such as ±20%), it is marked as a mutation point. In addition, the change amplitude of each pixel can be calculated. If the brightness change of a pixel exceeds a certain absolute value (such as 50 cd / m²) in a short time (such as 1 second), it is also marked as a mutation point. Traverse the dynamic photoelectric characteristic data of each pixel and calculate its change rate and change amplitude one by one. If the mutation standard is met, the pixel is marked as a photoelectric situation mutation pixel. If the brightness of a pixel at time point T1 is 200 cd / m², and the brightness changes to 150 cd / m² at T2 (1 second later), the change rate is: change rate = 150-200 / 1=-50cd / m² / s. If this change rate exceeds the set threshold, the pixel is marked as a mutation point. The identified photoelectric state mutation pixel points are recorded in the database, including the coordinates of the mutation pixel, the brightness value before and after the mutation, and the rate of change. A detailed mutation record is formed for subsequent analysis. Through visualization tools, the positions of all mutation pixels are marked on the image of the display screen, and different colors or symbols are used to represent different levels of mutation, so that technicians can quickly locate the problem area. All marked photoelectric state mutation pixel points are summarized and analyzed, and their distribution on the entire display screen is calculated. Analyze the frequency and distribution area of ​​the mutation phenomenon and its relationship with environmental factors (such as temperature and humidity). If the mutation frequency in a specific area is found to be significantly higher than that in other areas, it may indicate that there is a potential fault or environmental impact in that area.Generate a detailed report based on the analysis results, which should include statistical data on photoelectric situation mutations, distribution of mutation pixels and possible causes. Propose targeted optimization suggestions, such as focusing on monitoring areas with frequent mutations or improving the cooling system of the display to reduce the occurrence of mutations.

[0019] Step S6: Pixel fault prediction is performed for pixels with sudden photoelectric state changes, and pixel bad point repair is iteratively optimized according to the local brightness compensation strategy to build an intelligent pixel prediction and repair engine.

[0020] In this embodiment, the data of the photoelectric state mutation pixel point is extracted from the previous step, including brightness, color, change rate and other parameters. Ensure the integrity of the data and record the operation history of each pixel for fault prediction. Use statistical analysis methods to extract feature data related to the fault. The standard deviation of the average brightness, color deviation and change rate of each pixel can be calculated, and these features will be used as inputs for subsequent prediction models. Select a suitable machine learning model for fault prediction. Commonly used models include decision trees, random forests and support vector machines (SVM). When selecting a model, the feature dimensions and sample size of the data should be taken into account to ensure the accuracy and generalization ability of the model. When using a random forest model, the performance of the model can be evaluated by cross-validation methods (such as K-fold cross-validation) to ensure its stability on different data sets. Use historical data (including normal and fault data) for model training. Divide the data into training sets and test sets, usually setting 70% for training and 30% for testing. During the training process, adjust the model parameters to improve the prediction accuracy. When evaluating the performance of the model, indicators such as accuracy, recall and F1-score can be used to ensure that the model can accurately identify potential faulty pixels. Apply the trained model to the pixels with sudden photoelectric changes to predict faults. The model will output the fault probability of each pixel. If the probability exceeds the set threshold (such as 70%), it will be marked as a bad pixel that needs to be repaired. Record each prediction result, including the coordinates, fault probability and status information of the pixel, to provide a basis for the subsequent repair process. According to the local brightness compensation strategy formulated in the previous step, evaluate the brightness distribution of the current bad pixel. Ensure that the compensation strategy can effectively deal with the brightness differences of different bad pixels. Set the brightness compensation goal to make the brightness of each bad pixel reach the average brightness value of the normal pixels around it to ensure the uniformity of the visual effect. After the initial repair, re-measure the brightness and color value of each pixel to evaluate the repair effect. Calculate the standard deviation of the brightness after repair. If the standard deviation is still high (such as more than 3 cd / m²), a second repair is required. During the iterative optimization process, record the parameters and effects of each repair for subsequent analysis. Set the maximum number of iterations (such as 3 times) to avoid over-repair. Build an intelligent pixel prediction and repair engine based on the data and feedback during the repair process. The engine should have real-time monitoring, fault prediction and automatic repair functions, and be able to automatically adjust the repair strategy based on real-time data. The core of the engine is to integrate the fault prediction model and brightness compensation strategy, and make intelligent decisions in a data-driven way. When a sudden change in the photoelectric situation is identified, the engine can automatically start the prediction and repair process. During each repair process, all relevant data is recorded, including the brightness, color, fault probability and repair effect of each pixel. This data will be used for subsequent analysis and optimization to help evaluate the effectiveness of the repair strategy. Generate a detailed repair report, including comparison data before and after the repair, compensation value, and brightness uniformity after the repair, so that the technical team can make further decisions.

[0021] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Step S11: Obtaining multi-dimensional operation monitoring parameters of pixels of the LED display screen; Step S12: Calculate the brightness of each pixel point according to the multi-dimensional operation monitoring parameters of the pixel point, and extract the brightness value of each pixel point; Step S13: performing global color analysis on the multi-dimensional operation monitoring parameters of the pixel points to generate global color features; Step S14: Calculating the pixel display response time of the multi-dimensional operation monitoring parameters of the pixel point to generate the display response time of each pixel point; Step S15: bad pixels are identified according to the brightness value of each pixel, the global color characteristics and the display response time of each pixel, and all bad pixels are marked.

[0022] In this embodiment, a data acquisition system is built, which includes a variety of sensors for real-time monitoring of the operating parameters of the LED display. These sensors should be able to obtain multi-dimensional information such as brightness, color, temperature, humidity, current and voltage of the pixel. Use a light sensor to measure brightness, a color sensor (such as an RGB sensor) to obtain the color information of the pixel, and a temperature and humidity sensor to monitor environmental impact. The selection of these sensors should take into account their accuracy, response time and ability to adapt to the environment. Determine the frequency of data acquisition to ensure real-time and accuracy. Generally, the acquisition frequency is set to once per second, so that rapidly changing states can be captured. For some parameters, such as temperature and humidity, it can be set to be collected every 5 seconds. Through real-time data streams, the collected parameters are stored in a database to ensure that each parameter has a timestamp record for subsequent analysis. The collected data needs to be integrated and preprocessed to remove outliers and noise. Simple statistical methods, such as mean and standard deviation, can be used to judge the rationality of the data and eliminate unreasonable data. Ensure that the multi-dimensional parameters of each pixel are recorded at the same time point so that they can be compared and associated in subsequent analysis. According to the working principle of the LED display, a brightness calculation model is established. The brightness value of each pixel is usually related to its driving current and PWM (pulse width modulation) duty cycle. Determine the brightness calculation formula for each pixel, for example: brightness = current × PWM duty cycle. Set the current range (such as 0-20 mA) according to the actual application. Use the control system to read the current and PWM information of each pixel one by one. By writing a control program, the system can access the control register of each pixel in turn to obtain its current current value and PWM duty cycle. Set an appropriate sampling frequency (such as sampling each pixel once per second) to ensure the real-time nature of the brightness value. Through the calculation model, convert the acquired current and PWM information into the corresponding brightness value. Store the brightness value of each pixel in the database to form a complete brightness matrix. When recording data, ensure that the format is unified for subsequent processing and analysis. Determine the color space used, such as RGB, HSV, or CIE Lab. The RGB color space is suitable for extracting color information directly from the LED display, while the CIE Lab color space is more suitable for color difference analysis. Select the appropriate color space according to the target application. Use the brightness value and pixel color information in the real-time monitoring parameters to calculate the global color characteristics. Common features include average hue, saturation, and brightness. By traversing all pixels, the average value of these features is calculated. The average hue can be obtained by adding the hue values ​​of all pixels and dividing by the total number of pixels. The extracted global color features are visualized, and the distribution of different colors can be displayed using a histogram or pie chart. This process can intuitively reflect the color performance of the display and provide a basis for subsequent color adjustment and optimization. Determine the measurement method of pixel display response time.Generally, the response time can be calculated by comparing the time when the display command is sent and the time when the actual brightness change reaches a stable state. Set a suitable test pattern (such as alternating black and white) to facilitate the observation of brightness changes. During the test, record the brightness change data and the corresponding timestamp of each pixel. Use a high-speed data acquisition system (such as a 10 kHz sampling frequency) to ensure that rapid brightness changes can be captured. Based on the recorded data, calculate the display response time of each pixel. Response time is usually defined as the time required from signal input to output reaching a stable brightness. Record the response time of all pixels in a database to form a response time matrix. Establish identification criteria for bad pixels, which usually include uneven brightness, color deviation exceeding a threshold, and long response time. Set a brightness threshold (such as a brightness difference of more than 20% from surrounding pixels), a color deviation threshold (such as ΔE greater than 5), and a response time threshold (such as more than 50 ms). Traverse all pixels and analyze each pixel one by one according to the established criteria. By comparing the brightness value, global color characteristics, and response time of each pixel, determine whether it meets the bad pixel standard. For the identified bad pixels, mark them and record them in the database. Visual tools can be used to highlight the location of bad pixels on the display screen to facilitate subsequent inspection and maintenance. Make sure the recorded information includes the coordinates, brightness value, color characteristics and response time of the bad pixels to facilitate subsequent analysis and processing.

[0023] In this embodiment, the specific steps of step S15 are: Calculate the average brightness of the display screen according to the brightness value of each pixel to generate an average brightness value of the display screen; Based on the average brightness value of the display screen, outlier detection of brightness of each pixel is performed to extract multiple outlier pixels; The deviation types of multiple outlier deviation pixels are identified to obtain over-bright pixels and dark pixels; Detect color deviation points based on global color features and mark color deviation pixels; Define the preset response time length; Performing a slow response pixel analysis on the display response time of each pixel according to the preset response time length, and extracting a response delay dead point; Determine overbright pixels, dark pixels, color-shifted pixels, and response delay dead pixels as bad pixels, perform distribution statistics, and mark all bad pixels.

[0024] In this embodiment, the brightness value of each pixel is collected. Through the monitoring in the previous stage, it is ensured that the brightness value of each pixel is stored in a matrix form. Assume that the display screen has 640x480 pixels, a total of 307200 pixels, and the brightness value range is 0 to 255 (8-bit depth). The average brightness value of the display screen is calculated using the formula: Average brightness = , where N is the total number of pixels. Write a program to accumulate the brightness values ​​of each pixel and finally divide it by the total number of pixels to obtain the average brightness value. In the experiment, the data acquisition cycle can be set, for example, the average brightness is calculated every 30 seconds to monitor the changes in the display screen. The calculated average brightness value is recorded in the database, and a visual chart is generated to show the average brightness changes in different time periods. This process helps to analyze the brightness stability of the display screen and its response to changes in ambient light. The standard deviation (SD) method is usually used to formulate the standard for outlier deviation. If the difference between the brightness value of a pixel and the average brightness value exceeds the set threshold (average brightness ± 2 × SD), the pixel is considered an outlier. Traverse each pixel and calculate its deviation from the average brightness value. Record the deviation value and determine whether it exceeds the set threshold. If the absolute value of the deviation exceeds the threshold, the pixel is added to the list of outlier deviation pixels. All detected outlier deviation pixels are recorded in the database, and their coordinates and deviation values ​​are marked for subsequent analysis and processing. At the same time, a distribution map of outliers is generated to help visual analysis. Establish identification standards for overbright and dark points. Usually, the threshold of the bright spot is set to the average brightness + 2 × SD, and the threshold of the dark spot is set to the average brightness - 2 × SD. Traverse the previously extracted outlier deviation pixels and classify them according to their deviation values. For each outlier, if its brightness value exceeds the overbrightness threshold, it is marked as an overbright pixel; if its brightness value is lower than the dark spot threshold, it is marked as a dark spot. When the brightness>overbrightness threshold, it is marked as overbright. When the brightness<dark spot threshold, it is marked as a dark spot. First calculate the global color features of the display, such as average hue, saturation, and brightness. Use the RGB color value, convert it to the HSV color space, extract the hue (Hue) and saturation (Saturation), and calculate the global average. Formulate the detection standard of color deviation, usually set the color deviation threshold (such as ΔE>5), which is used to judge the difference between the color of a certain pixel and the global average color. ΔE is a measure of color difference. Traverse each pixel and calculate the ΔE value of its color and the global color feature. If ΔE exceeds the set threshold, the pixel is marked as a color deviation pixel. Record the coordinates and color information of the color deviation point for subsequent analysis. Determine the response time standard of the display. For LED displays, the response time threshold is usually set to 50 milliseconds. Response time refers to the time required from signal input to image output stabilization. Use a specific test pattern (such as alternating black and white) to record the response time of each pixel. Accurate measurement can be ensured by a high-frequency data acquisition system (such as a 10 kHz sampling frequency). Record the response time of each pixel in a database to form a response time matrix. According to the set response time standard, mark pixels with longer response times for subsequent analysis. Mark pixels with response times exceeding 50 milliseconds as slow-responding pixels. Use the response time data to make judgments for each pixel.Traverse all pixels and check whether the response time exceeds the set threshold. If the response time is greater than 50 milliseconds, add the pixel to the list of slow response points. Record all identified slow response pixels in the database and generate a distribution map of slow response points to facilitate subsequent inspection and maintenance. Define overbright pixels, dark pixels, color-shifted pixels, and response delay dead points as bad pixels. Set the classification standards for bad pixels and create a bad pixel record table. Count all identified bad pixels and record their number, type, and location. Generate a statistical report on bad pixels, including the number of each type of bad pixel and its distribution on the display. Use visualization tools to display the distribution of bad pixels and generate heat maps or scatter plots to help quickly locate problem areas. This process can provide important data support for subsequent maintenance and repair.

[0025] In this embodiment, refer to Figure 3 , is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Step S21: performing bad pixel location calculation on all bad pixels to generate the on-screen position coordinates of each bad pixel; Step S22: performing bad pixel spatial distribution analysis according to the position coordinates of each bad pixel within the screen to generate a bad pixel spatial distribution topology map; Step S23: dividing the bad pixel spatial distribution topology map into bad pixel cluster regions to generate multiple bad pixel cluster regions to be repaired; Step S24: mining the regional color temperature distribution features of the multiple bad pixel clustered areas to be repaired, so as to generate the color temperature distribution features of each area to be repaired.

[0026] In this embodiment, in the previous steps, all bad pixels have been identified and their coordinate information has been recorded. These coordinates are usually expressed in the form of rows and columns. For a 640x480 LED display screen, the coordinates of each pixel can be expressed as (x, y), where x is the number of columns and y is the number of rows. By traversing all the identified bad pixels, the corresponding screen position coordinates are extracted. For each bad pixel, its position in the matrix is ​​recorded and converted into actual physical coordinates. If the size of each pixel is 2mm, the actual position of the bad pixel can be calculated by the formula: actual position = (x×pixel width, y×pixel height). In this way, the screen position coordinates of each bad pixel are generated and stored in the database. The calculated bad pixel position coordinates are recorded in the form of a list or array for subsequent analysis. At the same time, a visual representation of the display screen can be generated to mark the positions of all bad pixels for subsequent maintenance and repair work. A spatial distribution model of bad pixels is established. All position coordinates are mapped into a two-dimensional coordinate system, and the position of the bad pixel is represented in the form of a scatter plot. At this time, the coordinates of each bad pixel will be used as a point in the coordinate system. Use spatial statistics to calculate the distribution density of bad pixels. Kernel Density Estimation (KDE) can be used to analyze the distribution of bad pixels in space and identify the clustered and sparse areas of bad pixels. Visualize the calculated bad pixel distribution information to generate a topological map of the bad pixel spatial distribution. The topological map can be displayed in the form of a heat map. The color depth indicates the density of bad pixels. Dark red indicates a dense area of ​​bad pixels, and light color indicates a sparse area. Make sure the topological map is clear and easy to read for subsequent analysis. Select a suitable clustering algorithm, such as DBSCAN (Density-Based Spatial Clustering of Applications with Noise) or K-means, to perform cluster analysis on bad pixels. DBSCAN is suitable for processing clusters of different densities and can effectively identify the clustered areas of bad pixels. Input the bad pixel location coordinates into the clustering algorithm for cluster analysis. Set appropriate parameters, such as the minimum number of samples and neighborhood radius, to ensure the accuracy of clustering. For DBSCAN, the minimum number of samples can be set to 3 and the neighborhood radius can be set to 5 pixels. Analyze the clustering results and record the coordinate range of each cluster area and the number of bad pixels it contains. Generate an identifier for each cluster area to facilitate subsequent repair plans. The clustering results can be displayed through visualization tools to help intuitively understand the clustering of bad pixels. For each area to be repaired, calculate its color temperature characteristics. The color temperature can be obtained by converting the RGB value. Color temperature (K) = T / (R+G+B), where R, G, and B are the average values ​​of all pixels in the area, and T is the preset reference temperature used to adjust the color temperature calculation. Count the color temperature values ​​in each cluster area, and calculate statistical features such as mean and variance to generate a color temperature distribution feature report.The color temperature distribution of each area can be displayed through a histogram to help understand the changes in color temperature within the area. The generated color temperature distribution characteristics are recorded in the database to ensure the integrity of the data. At the same time, a visual chart is generated for each area to be repaired to show the trend of color temperature distribution changes, so as to facilitate subsequent repair and maintenance decisions.

[0027] In this embodiment, refer to Figure 4 , is a flowchart of detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include: Step S31: comparing the color differences between the multiple bad pixel clustering areas to be repaired and the adjacent pixels in the areas, and generating a bad pixel color difference value for each area to be repaired; Step S32: Calculate the bad pixel repair value one by one according to the color temperature distribution characteristics of each area to be repaired and the bad pixel color difference value of each area to be repaired, so as to obtain the color repair value of each bad pixel; Step S33: performing iterative color difference compensation calculation on the color restoration value of each defective pixel, so as to generate the optimal color restoration value of each defective pixel; Step S34: performing adaptive color restoration processing on all bad pixels one by one based on the optimal color restoration value of each bad pixel, and collecting the restored display pixel parameters.

[0028] In this embodiment, the boundary of each area to be repaired is determined so that adjacent pixels within the area are selected. In each area to be repaired on the display screen, pixels adjacent to the bad pixel are selected, and four adjacent pixels above, below, left and right are usually selected. For each bad pixel, ensure that at least four neighboring pixels can be obtained. Calculate the color difference between each bad pixel and its adjacent pixels. The bad pixel color difference value of each area to be repaired is recorded in the database to form a color difference analysis report. By generating a visual chart to show the color difference of each area, it helps to identify which areas have more urgent repair needs. This process can set a threshold for each area. If the color difference exceeds 15, it is marked as an area that needs to be repaired first. The color temperature distribution characteristics of each area to be repaired extracted in the previous step are used to determine the average color temperature value in the area. The color temperature value is usually in Kelvin (K), and the color temperature mean and its variance of each area are recorded to understand the color temperature changes in the area. Establish a color repair value calculation model. According to the color difference value of the bad pixel and the color temperature distribution characteristics of the area, set the calculation formula of the repair value. Color restoration value = color temperature average − color difference of bad pixel. For each bad pixel in the area to be restored, calculate its color restoration value one by one and record it in the database. This process can be achieved by writing an automated script to ensure the accuracy and efficiency of the calculation. The restoration value of each bad pixel should be within a reasonable range (RGB value from 0 to 255). Select a suitable iterative algorithm, such as Newton's method or gradient descent method, to optimize the color restoration value of each bad pixel. The iterative method should gradually adjust the restoration value to reduce the color difference based on the preliminary calculation results of the color restoration value. Set the optimization goal during the iteration process, usually to minimize the color difference after restoration. Record the change of the restoration value in each iteration and monitor its convergence speed and stability. Set a convergence threshold. When the color difference change is less than 1, stop the iteration and record the final optimal color restoration value. According to the optimal color restoration value of each bad pixel, perform adaptive color restoration on all bad pixels. The restoration algorithm should take into account the color information of the surrounding pixels to ensure that the restoration effect is natural and harmonious. Traverse each bad pixel and adjust the color according to its optimal restoration value. Adjust the RGB value of the bad pixel to make it close to the optimal color restoration value. Set the repair amplitude (such as ±10) to avoid visual discomfort caused by over-repair. After the repair is completed, re-collect the pixel parameters of the display, including brightness, color, and response time, to form a repaired data set. Record the changes in the repair effect and conduct comparative analysis to evaluate the success rate and improvement effect of the repair. Generate a status report after the repair to provide data support for subsequent maintenance.

[0029] In this embodiment, step S4 includes the following steps: Step S41: performing pixel uniformity analysis of the repaired area according to the repaired display pixel parameters, and extracting pixel uniformity of the repaired area; Step S42: performing regional brightness histogram distribution statistics on the restored display screen pixel parameters to generate brightness histogram distribution features; Step S43: Identify nodes with uneven brightness according to the brightness histogram distribution characteristics and the pixel uniformity of the repaired area, and mark the points with unbalanced brightness distribution; Step S44: performing local brightness deviation compensation calculation on the brightness distribution imbalance point, thereby obtaining a local brightness deviation compensation value; Step S45: performing local brightness uniformity compensation on the LED display screen based on the local brightness deviation compensation value, thereby constructing a local brightness compensation strategy.

[0030] In this embodiment, the brightness values ​​of all pixels in the repair area are extracted to ensure the integrity and accuracy of the data. These data should be stored in a matrix form for subsequent analysis. To determine the quantitative standard of uniformity, the standard deviation (SD) is usually used to measure the brightness uniformity of the repair area. A uniformity threshold is set. If the standard deviation is less than 5, the brightness uniformity of the area is considered to be good. The average brightness and standard deviation of all pixels in the repair area are calculated. Uniformity = Standard Deviation = ( ), where n is the total number of pixels in the repaired area. Through this calculation, the pixel uniformity index of the repaired area can be extracted to determine whether the brightness of the area meets the expected standard. The brightness values ​​of the pixels of the repaired display screen are extracted to form a list of brightness values. Assume that the brightness value range of the display screen is 0 to 255 to ensure the integrity and accuracy of the data. Use the histogram statistical method to divide the brightness value into several intervals (bins), and divide the brightness value into 10 intervals, each with a width of 25. Count the number of pixels in each interval to form the statistical data of the brightness distribution. Draw the statistical results into a histogram, with the horizontal axis being the brightness value interval and the vertical axis being the number of pixels in the corresponding interval. Through this histogram, the pixel distribution in each brightness interval can be clearly observed, and the concentrated and sparse areas of the brightness distribution can be identified. According to the brightness histogram distribution characteristics and the pixel uniformity index, the recognition standard of the brightness uneven node is set. Set a threshold value. When the number of pixels in a brightness interval is significantly less than that in other intervals (less than 10% of the total number of pixels), it is marked as a brightness uneven node. Traverse each interval of the brightness histogram and calculate the number of pixels in the interval compared with other intervals. If it is found that the number of pixels in a certain interval is less than the set threshold and is consistent with the uniformity analysis result, the interval is marked as a brightness imbalance point. The identified brightness uneven nodes are recorded in the database and marked on the brightness histogram, using different colors or symbols to represent the imbalance points. This process can intuitively display the brightness imbalance and provide a basis for subsequent compensation. For the identified brightness imbalance points, calculate their local brightness deviation. The local brightness deviation can be determined by comparing the average brightness of the imbalance point area with the average brightness of the surrounding normal area. Set the compensation formula: local deviation = target brightness − current brightness. For each imbalance point, calculate its compensation value and ensure that the value is within a reasonable range (the compensation value does not exceed ±20). This process can be achieved by traversing each pixel one by one to ensure that the deviation of each imbalance point is recorded. The calculated local brightness deviation compensation value is recorded in the database and a compensation analysis report is generated. By analyzing these compensation values, data support can be provided for the subsequent brightness uniformity compensation strategy. According to the calculated local brightness deviation compensation value, a corresponding compensation strategy is formulated. The compensation strategy should take into account the compensation value of each imbalance point and its impact on surrounding pixels to ensure a natural repair effect. The brightness value of each brightness imbalance point is adjusted one by one through the control system, and adjusted according to the compensation value. If the compensation value of an imbalance point is +15, the brightness of the pixel is increased by 15 units to ensure that the adjusted brightness meets the set standard. After completing the local brightness uniformity compensation, re-measure and evaluate the brightness of the display to ensure that the compensation strategy is effective. If it is found that the brightness of some areas is still uneven, the compensation strategy can be further refined and adjusted again. The brightness parameters of the compensated display are recorded in the database, and a compensation effect report is generated to show the brightness changes before and after compensation.This process provides data support for subsequent maintenance and optimization, ensuring that the brightness uniformity of the display reaches the expected target.

[0031] In this embodiment, the specific steps of step S5 are: Step S51: Calculate the current and voltage values ​​of each pixel point according to the multi-dimensional operation monitoring parameters of the pixel point; Step S52: mining the dynamic photoelectric change characteristics of the current and voltage values ​​of each pixel point to generate the photoelectric change characteristics of each pixel point; Step S53: performing multi-period photoelectric characteristic fitting on the photoelectric change characteristics of each pixel point to construct a photoelectric characteristic curve for each pixel point; Step S54: predicting the photoelectric trend evolution based on the photoelectric characteristic curve of each pixel point to generate a photoelectric characteristic situation map of each pixel point; Step S55: performing photoelectric situation mutation identification on the photoelectric characteristic situation map of each pixel point, and locating the pixel point where the photoelectric situation mutation occurs.

[0032] In this embodiment, the integrity of the multi-dimensional operation monitoring parameters of the pixel point is ensured. These parameters include brightness, color information, and environmental factors such as temperature and humidity that have been recorded before. These parameters are used to derive the current and voltage values ​​of each pixel. According to the working principle of the LED display screen, a calculation model for current and voltage is established. Generally, the current (I) is proportional to the brightness (L), and the voltage (V) is related to the driving current and the characteristics of the LED. I=k×L, where k is a proportional constant, usually provided by the manufacturer. The voltage can be calculated using the following formula: V=I×R, where R is the internal resistance of the LED, which needs to be set according to the specific LED model. Using the above model, the current and voltage values ​​of each pixel are calculated one by one. By writing an automated script, the efficiency and accuracy of the calculation are ensured. The current and voltage values ​​of each pixel are recorded and stored in a database to provide data support for subsequent analysis. The calculation results are generated into a table or a visual chart to show the current and voltage distribution of each pixel. This process helps to identify the working status of the display screen and lays the foundation for the subsequent mining of dynamic photoelectric change characteristics. The changes of these values ​​at different time points are recorded to form a dynamic data set. You can set a time interval, such as recording data once per second. Select an appropriate analysis method, such as time series analysis or dynamic system modeling, to mine the photoelectric change characteristics of each pixel. These characteristics include response speed, stability, and dynamic range. Extract dynamic photoelectric characteristics by analyzing the current and voltage changes of each pixel at different time points. Calculate the rate of change of current and voltage, peak response time, etc. ΔI=I(t)−I(t−1), ΔV=V(t)−V(t−1), record the changes in these characteristics, and generate the photoelectric change characteristic data of each pixel. Select a suitable fitting model, such as linear regression, polynomial fitting, or nonlinear regression. You can select a suitable model to improve the fitting accuracy based on the photoelectric change characteristics of the pixel. Divide the photoelectric change characteristic data of each pixel into multiple time periods, such as by hour or minute, to ensure that the data volume of each time period is sufficient for accurate fitting. Apply the fitting model to the data of each time period, calculate the fitting parameters and generate the photoelectric characteristic curve. Record the R² value of the fit to evaluate the goodness of fit. Record the photoelectric characteristic curve of each pixel and its fitting parameters in the database, and generate a visual chart to show the photoelectric characteristic changes of different pixels. This process helps to predict the subsequent trend evolution. Select a suitable time series prediction model, such as ARIMA (autoregressive integrated moving average model) or LSTM (long short-term memory network), for trend prediction of the photoelectric characteristics of each pixel. Prepare training and test sets using the photoelectric characteristic curve data generated in the previous step. Divide the data into a training set (e.g. 70%) and a test set (e.g. 30%) for model training and verification. Apply the selected prediction model to predict the future photoelectric characteristics of each pixel. Record the prediction results and calculate the confidence interval of the prediction for subsequent analysis.The predicted photoelectric characteristic situation map is recorded in the database, and a visual chart is generated to show the photoelectric characteristic trend of each pixel. This process can help identify potential abnormal changes and provide data support for subsequent mutation identification. The identification criteria for photoelectric situation mutation are usually determined by using the threshold method or the change rate method. If the photoelectric characteristic of a pixel changes by more than the set threshold (the change rate exceeds 30%) in a short period of time, it is marked as a mutation point. The photoelectric characteristic situation map of each pixel is traversed to calculate its change rate and change amplitude. When it is found that the change of a pixel exceeds the set threshold, it is marked as a photoelectric situation mutation pixel. The identified mutation pixels are recorded in the database, and a visual chart of the mutation pixels is generated at the same time, marking the location of the mutation point. Through visualization tools, these mutation points can be quickly located and analyzed. Further analysis is performed on the identified mutation pixels, such as checking their current, voltage and environmental parameters to understand the cause of the mutation. This process can provide important data support for subsequent maintenance and repair to ensure the normal operation of the display.

[0033] In this embodiment, the specific steps of step S6 are: Step S61: performing pixel point fault prediction on the pixel point where the photoelectric state suddenly changes, so as to obtain pixel point fault prediction data; Step S62: Calculate the fault time of the pixel point fault prediction data to generate a fault prediction time point; Step S63: quantifying the fault repair time of the pixel point fault prediction data to generate the pixel point fault repair time; Step S64: Estimating the optimal pre-repair time point according to the fault prediction time point and the pixel fault repair time, thereby generating the optimal pre-repair time point; Step S65: performing pre-repair diagnosis on the optimal pre-repair time point to generate a pixel point fault prediction and repair strategy; Step S66: Perform iterative optimization on the pixel fault prediction and repair strategy and the local brightness compensation strategy to repair bad pixels and build an intelligent pixel prediction and repair engine.

[0034] In this embodiment, the relevant data of all the pixels with sudden photoelectric situation changes are extracted from the previous steps, including current, brightness, color and historical fault records. These data will serve as the basis for fault prediction. Select a suitable fault prediction model, such as a decision tree, random forest or support vector machine (SVM) in machine learning. By analyzing historical data, determine which features have a significant impact on the occurrence of faults. Use the collected photoelectric situation data to train the prediction model. During the model training process, set the ratio of the training set to the test set, for example, 70% for training and 30% for testing. Adjust the model parameters to improve the prediction accuracy. Use the trained model to predict the fault of the pixel with sudden photoelectric situation changes. The model outputs the fault probability and prediction results of each pixel, and records these results as pixel fault prediction data. The fault prediction data is stored in a database, and a visual chart is generated to show the fault prediction probability of different pixels. This process can help identify which pixels are most likely to fail and provide data support for subsequent time calculations. According to historical fault data, set the time window for the occurrence of faults. By analyzing past fault records, it is determined that the fault generally occurs within a specific time after the prediction, such as 1 hour, 3 hours or 24 hours. Establish a time model based on the probability of fault prediction. Exponential distribution or normal distribution can usually be used to simulate the time when a fault occurs. Set the expected value of the time of fault occurrence to the inverse of the predicted probability. For each pixel, calculate the predicted time point of the fault based on its predicted probability of fault and historical data. Record the predicted time points of faults for all pixels to ensure the accuracy of the data. Analyze the repair time of different types of faults based on the repair records of historical faults. Determine that the repair time of simple faults is about 10 minutes, while complex faults may take 1 hour. Establish a repair time model, which can usually be combined with the severity and type of fault. Linear regression or other statistical models can be used to predict the repair time by analyzing historical data. For each pixel, use the repair time model to calculate its fault repair time. Record the repair time of each pixel to ensure the accuracy and completeness of the data. Record the fault repair time of each pixel in the database and generate a visual chart to show the distribution of repair time for different pixels. This process provides data support for the subsequent pre-repair time point estimation. Set a pre-repair time window, such as repairing 10 minutes before the predicted time point of the fault to ensure that the problem does not affect the normal operation of the display. For each pixel, calculate the optimal pre-repair time point based on its fault prediction time point and the corresponding fault repair time. Establish pre-repair diagnostic criteria to determine which pixels need to be repaired first. If the fault prediction probability exceeds 70%, repair is given priority. Perform pre-repair diagnosis on each pixel based on the pre-repair time point and the set diagnostic criteria. Record the pixels that need to be repaired and generate a repair strategy. Record the diagnostic results in the database to form a pixel fault prediction repair strategy report.By analyzing these strategies, optimization can be performed to improve repair efficiency. Generate a visual chart to display the pre-repair strategies and their priorities to help the maintenance team quickly locate the pixels that need to be repaired. Design an intelligent repair engine framework to integrate fault prediction repair strategies and local brightness compensation strategies. Ensure that the engine can receive pixel status information in real time and respond quickly. Select a suitable optimization algorithm, such as genetic algorithm, particle swarm optimization, or reinforcement learning, to achieve iterative optimization of the repair strategy. Set appropriate parameters to ensure the effectiveness of the algorithm. Perform iterative optimization of pixel bad pixel repair based on fault prediction repair strategy and local brightness compensation strategy. Record the effect of each optimization and evaluate its impact on the display effect. Record the optimization results and feedback information in the database for subsequent analysis and improvement. Based on the feedback data, continuously adjust the repair strategy to improve the intelligence level of the repair engine. Generate a visual report to show the performance and effect of the intelligent repair engine. Through comparative analysis, show the effect changes before and after the repair, and provide data support for subsequent maintenance and optimization.

[0035] In the present invention, a device for repairing bad pixels of LED display screen is provided, comprising: A bad pixel identification module is used to obtain multi-dimensional operation monitoring parameters of pixels of the LED display screen; identify bad pixels according to the multi-dimensional operation monitoring parameters of the pixels, and mark all bad pixels; The bad pixel location module is used to calculate the bad pixel locations of all bad pixels and mine the regional color temperature distribution features to generate the color temperature distribution features of each area to be repaired; The adaptive repair module is used to calculate the repair value of each bad pixel one by one according to the color temperature distribution characteristics of each area to be repaired, and adaptively perform color repair processing, and collect the pixel parameters of the repaired display screen; A local brightness compensation module is used to identify nodes with uneven brightness in the repair area according to the repaired display pixel parameters, and then perform local brightness uniformity compensation, thereby constructing a local brightness compensation strategy; The photoelectric characteristic module is used to mine the dynamic photoelectric change characteristics according to the multi-dimensional operation monitoring parameters of the pixel points, identify the photoelectric situation mutation, and locate the photoelectric situation mutation pixel points; The iterative optimization module is used to predict pixel faults for pixels with sudden photoelectric state changes, and iteratively optimize pixel bad pixel repair based on the local brightness compensation strategy to build an intelligent pixel prediction and repair engine.

[0036] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned methods for repairing bad pixels of LED display screens are implemented.

[0037] Those skilled in the art clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0038] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it is stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or partly or all or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media for storing program codes such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks or optical disks.

[0039] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0040] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for repairing bad pixels on an LED display screen, characterized in that: The following steps are involved: Step S1: Obtaining multi-dimensional operation monitoring parameters of pixels of the LED display screen; Identify bad pixels according to the multi-dimensional operation monitoring parameters of the pixels, and mark all bad pixels; Step S2: Calculate the position of all bad pixels and mine the regional color temperature distribution features to generate the color temperature distribution features of each area to be repaired; Step S3: Calculate the bad pixel repair value one by one according to the color temperature distribution characteristics of each area to be repaired, perform adaptive color repair processing, and collect the pixel parameters of the repaired display screen; Step S4: Identify nodes with uneven brightness in the repair area according to the repaired display pixel parameters, and then perform local brightness uniformity compensation, thereby constructing a local brightness compensation strategy; Step S5: mining dynamic photoelectric change characteristics according to the multi-dimensional operation monitoring parameters of the pixel points, identifying photoelectric situation mutations, and locating the photoelectric situation mutation pixel points; Step S6: Pixel fault prediction is performed for pixels with sudden photoelectric state changes, and pixel bad point repair is iteratively optimized according to the local brightness compensation strategy to build an intelligent pixel prediction and repair engine.

2. The method for repairing bad pixels of LED display screen according to claim 1, characterized in that: The specific steps of step S1 are: Step S11: Obtaining multi-dimensional operation monitoring parameters of pixels of the LED display screen; Step S12: Calculate the brightness of each pixel point according to the multi-dimensional operation monitoring parameters of the pixel point, and extract the brightness value of each pixel point; Step S13: performing global color analysis on the multi-dimensional operation monitoring parameters of the pixel points to generate global color features; Step S14: Calculating the pixel display response time of the multi-dimensional operation monitoring parameters of the pixel point to generate the display response time of each pixel point; Step S15: bad pixels are identified according to the brightness value of each pixel, the global color characteristics and the display response time of each pixel, and all bad pixels are marked.

3. The method for repairing bad pixels of LED display screen according to claim 2, characterized in that: The specific steps of step S15 are: Calculate the average brightness of the display screen according to the brightness value of each pixel to generate an average brightness value of the display screen; Based on the average brightness value of the display screen, outlier detection of brightness of each pixel is performed to extract multiple outlier pixels; The deviation types of multiple outlier deviation pixels are identified to obtain over-bright pixels and dark pixels; Detect color deviation points based on global color features and mark color deviation pixels; Define the preset response time length; Performing a slow response pixel analysis on the display response time of each pixel according to the preset response time length, and extracting a response delay dead point; Determine overbright pixels, dark pixels, color-shifted pixels, and response delay dead pixels as bad pixels, perform distribution statistics, and mark all bad pixels.

4. The method for repairing bad pixels of LED display screen according to claim 1, characterized in that: The specific steps of step S2 are: Step S21: performing bad pixel location calculation on all bad pixels to generate the on-screen position coordinates of each bad pixel; Step S22: performing bad pixel spatial distribution analysis according to the position coordinates of each bad pixel within the screen to generate a bad pixel spatial distribution topology map; Step S23: dividing the bad pixel spatial distribution topology map into bad pixel cluster regions to generate multiple bad pixel cluster regions to be repaired; Step S24: mining the regional color temperature distribution features of the multiple bad pixel clustered areas to be repaired, so as to generate the color temperature distribution features of each area to be repaired.

5. The method for repairing bad pixels of LED display screen according to claim 1, characterized in that: The specific steps of step S3 are: Step S31: comparing the color differences between the multiple bad pixel clustering areas to be repaired and the adjacent pixels in the areas, and generating a bad pixel color difference value for each area to be repaired; Step S32: Calculate the bad pixel repair value one by one according to the color temperature distribution characteristics of each area to be repaired and the bad pixel color difference value of each area to be repaired, so as to obtain the color repair value of each bad pixel; Step S33: performing iterative color difference compensation calculation on the color restoration value of each defective pixel, so as to generate the optimal color restoration value of each defective pixel; Step S34: performing adaptive color restoration processing on all bad pixels one by one based on the optimal color restoration value of each bad pixel, and collecting the restored display pixel parameters.

6. The method for repairing bad pixels of LED display screen according to claim 1, characterized in that: The specific steps of step S4 are: Step S41: performing pixel uniformity analysis of the repaired area according to the repaired display pixel parameters, and extracting pixel uniformity of the repaired area; Step S42: performing regional brightness histogram distribution statistics on the restored display screen pixel parameters to generate brightness histogram distribution features; Step S43: Identify nodes with uneven brightness according to the brightness histogram distribution characteristics and the pixel uniformity of the repaired area, and mark the points with unbalanced brightness distribution; Step S44: performing local brightness deviation compensation calculation on the brightness distribution imbalance point, thereby obtaining a local brightness deviation compensation value; Step S45: performing local brightness uniformity compensation on the LED display screen based on the local brightness deviation compensation value, thereby constructing a local brightness compensation strategy.

7. The method for repairing bad pixels of LED display screen according to claim 1, characterized in that: The specific steps of step S5 are: Step S51: Calculate the current and voltage values ​​of each pixel point according to the multi-dimensional operation monitoring parameters of the pixel point; Step S52: mining the dynamic photoelectric change characteristics of the current and voltage values ​​of each pixel point to generate the photoelectric change characteristics of each pixel point; Step S53: performing multi-period photoelectric characteristic fitting on the photoelectric change characteristics of each pixel point to construct a photoelectric characteristic curve for each pixel point; Step S54: predicting the photoelectric trend evolution based on the photoelectric characteristic curve of each pixel point to generate a photoelectric characteristic situation map of each pixel point; Step S55: performing photoelectric situation mutation identification on the photoelectric characteristic situation map of each pixel point, and locating the pixel point where the photoelectric situation mutation occurs.

8. The method for repairing bad pixels of LED display screen according to claim 1, characterized in that: The specific steps of step S6 are: Step S61: performing pixel point fault prediction on the pixel point where the photoelectric state suddenly changes, so as to obtain pixel point fault prediction data; Step S62: Calculate the fault time of the pixel point fault prediction data to generate a fault prediction time point; Step S63: quantifying the fault repair time of the pixel point fault prediction data to generate the pixel point fault repair time; Step S64: Estimating the optimal pre-repair time point according to the fault prediction time point and the pixel fault repair time, thereby generating the optimal pre-repair time point; Step S65: performing pre-repair diagnosis on the optimal pre-repair time point to generate a pixel point fault prediction and repair strategy; Step S66: Perform iterative optimization on the pixel fault prediction and repair strategy and the local brightness compensation strategy to repair bad pixels and build an intelligent pixel prediction and repair engine.

9. A device for repairing bad pixels on an LED display screen, characterized in that: The method for repairing bad pixel points of an LED display screen according to claim 1 comprises: A bad pixel identification module is used to obtain multi-dimensional operation monitoring parameters of pixels of the LED display screen; identify bad pixels according to the multi-dimensional operation monitoring parameters of the pixels, and mark all bad pixels; The bad pixel location module is used to calculate the bad pixel locations of all bad pixels and mine the regional color temperature distribution features to generate the color temperature distribution features of each area to be repaired; The adaptive repair module is used to calculate the repair value of each bad pixel one by one according to the color temperature distribution characteristics of each area to be repaired, and adaptively perform color repair processing, and collect the pixel parameters of the repaired display screen; A local brightness compensation module is used to identify nodes with uneven brightness in the repair area according to the repaired display pixel parameters, and then perform local brightness uniformity compensation, thereby constructing a local brightness compensation strategy; The photoelectric characteristic module is used to mine the dynamic photoelectric change characteristics according to the multi-dimensional operation monitoring parameters of the pixel points, identify the photoelectric situation mutation, and locate the photoelectric situation mutation pixel points; The iterative optimization module is used to predict pixel faults for pixels with sudden photoelectric state changes, and iteratively optimize pixel bad pixel repair based on the local brightness compensation strategy to build an intelligent pixel prediction and repair engine.

10. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method for repairing bad pixels of an LED display screen according to any one of claims 1 to 8 are implemented.

Images