A method for compensating for edge brightness attenuation of an LCD panel

Abstract

This invention belongs to the field of display technology and discloses a method for compensating for brightness attenuation at the edge of an LCD panel. The method includes the following steps: S1: Acquiring attenuation parameters of the edge region using a data acquisition device. This invention employs a three-level response mechanism with an attenuation score threshold. At the compensation execution level, it dynamically calculates the compensation intensity coefficient based on the attenuation score and synchronously adjusts the backlight drive current and pixel data voltage. PID closed-loop control is formed through feedback from an optical sensor. Based on a predictive model-driven preventative compensation mode, when it is predicted that an edge region will reach a severe attenuation standard within a period below a time threshold, a series of preventative measures are initiated in advance. For the special application scenario of splicing screens, a collaborative compensation strategy for adjacent panels is proposed. When the brightness of the edge of a panel changes due to compensation, the brightness output of the corresponding edge of the adjacent panel is synchronously fine-tuned through the splicing controller, utilizing the visual characteristics of the human eye to achieve a smooth transition of the brightness field.

Description

Technical Field

[0001] This invention relates to the field of display technology, and more specifically, to a method for compensating for edge brightness attenuation in an LCD panel. Background Technology

[0002] With the rapid development of ultra-high-definition display technology and large-size visualization application scenarios, large splicing displays, which are physically spliced ​​together by multiple LCD panels, have become core display equipment in fields such as command and dispatch centers, commercial exhibitions, and conference systems. During long-term continuous operation, LCD panels generally have the inherent defect that the brightness decays faster in the edge area than in the center area, i.e., dark corners or edge brightness decay.

[0003] Existing technologies for compensating for brightness decay at the edges of LCD panels, at the single-panel level, mostly employ closed-loop dimming technology based on optical sensor feedback. This involves dynamically adjusting the backlight drive current or pulse width by measuring the overall or zoned brightness of the panel to maintain stable brightness. Some advanced solutions introduce simple brightness uniformity correction, using pre-stored or real-time measured brightness distribution maps to fine-tune the gamma curve of the drive signal or pixel voltage. These methods primarily focus on post-event compensation for already occurring decay, lacking the ability to predict decay trends and failing to deeply distinguish the specific physical causes of decay. This results in insufficient targeting of compensation strategies, limited long-term compensation effects, and may even accelerate local aging due to overcompensation. At the splicing screen system level, existing technologies mainly focus on the initial calibration of color and brightness consistency between panels. They lack an effective dynamic collaborative compensation mechanism for the problem of brightness and color mismatch on both sides of the splicing seam caused by different decay rates at the edges of each panel during operation. Summary of the Invention

[0004] To address the problems in the background art, this invention proposes a method for compensating for brightness attenuation at the edge of an LCD panel.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for compensating for edge brightness attenuation in an LCD panel, comprising the following steps: S1: Collect attenuation parameters of the edge region through the acquisition device, and normalize the attenuation parameters; S2: Generate an attenuation score for each edge region based on the processed attenuation parameters, and calculate the rate of change of the attenuation score based on the attenuation score. S3: Classify each edge region according to the attenuation score, and implement an emergency compensation mode for severely attenuated regions; S4: Obtain historical data on the total number of attenuation cells and construct a prediction model for the total number of attenuation cells. Use the prediction model to predict the total number of attenuation cells in the edge region. S5: Calculate the decay time based on the number of predicted decay unit blocks, set a time threshold, determine the risk of rapid decay in the edge region, and trigger the preventive compensation mode.

[0006] Furthermore, large-scale splicing screens are composed of multiple LCD panels. LCD panels are typically rectangular, with two short sides and two long sides. The area closer to the short side is called the short edge area, and the area closer to the long side is called the long edge area. Both the short edge area and the long edge area are edge areas. The LCD panel is the central area except for the edge areas. The two long edge areas are located between the two short edge areas. The length of the short edge area is equal to the length of the long edge area, and the width of the short edge area is equal to the width of the long edge area. Each unit area can be divided into multiple unit blocks, with each unit block being 10x10 pixels in size. The width of the edge area accounts for 20% of the width of the LCD panel. The attenuation parameters refer to the brightness attenuation rate, brightness uniformity coefficient, response time attenuation rate, voltage-brightness conversion efficiency, and temperature gradient index within the edge region. The luminance of each edge region and the reference luminance of the central region are measured using a spectrocolorimeter. The luminance of each edge region is divided by the reference luminance of the central region to obtain the luminance attenuation rate L. The luminance attenuation rate is then normalized.

[0007] In the formula, To normalize the luminance decay rate, This represents the actual brightness decay rate. and These represent the lower and upper limits of the brightness decay rate; Nine points are uniformly selected within each edge region to measure brightness. The standard deviation σ and mean μ are calculated. The brightness uniformity coefficient U = 1 - (σ / μ) / 0.2. The brightness uniformity coefficient is then normalized.

[0008] In the formula, The normalized brightness uniformity coefficient, This represents the actual brightness uniformity coefficient; The response time of the edge region from 10% to 90% brightness is measured and compared with the standard value. That is, the response time decay rate R = standard response time / measured response time. The response time decay rate is then normalized.

[0009] In the formula, To normalize the response time decay rate, This represents the actual response time decay rate. and These are the lower and upper limits of the response time decay rate; The brightness of the edge region is measured under a fixed voltage, and the ratio to the initial efficiency is calculated, i.e., voltage-to-brightness conversion efficiency V = current brightness / initial brightness. The voltage-to-brightness conversion efficiency is then normalized.

[0010] In the formula, To normalize the voltage-to-brightness conversion efficiency, This represents the actual voltage-to-brightness conversion efficiency. and These represent the lower and upper limits of voltage-to-brightness conversion efficiency; The temperature difference ΔT between the edge region and the center region is measured, and the temperature gradient exponent T = 1 - (ΔT / 15). The temperature gradient exponent is then normalized.

[0011] In the formula, The normalized temperature gradient exponent, This is the actual temperature gradient exponent.

[0012] Furthermore, the process of generating an attenuation score for each edge region based on the processed attenuation parameters includes: Attenuation score S:

[0013] In the formula, , , , and These are weighting coefficients, obtained through training based on historical data.

[0014] Furthermore, the process of calculating the rate of change of the attenuation score based on the attenuation score includes: Set an evaluation period. Each evaluation period is used to obtain a decay score for the edge region. The latest decay score value of each edge region is continuously stored to form a historical decay score sequence for that edge region. When it is necessary to calculate the current rate of change of decay score in the edge region, the decay score data of the most recent four evaluation periods in the historical decay score sequence are extracted, and the least squares method is used to perform linear fitting on the time series data to obtain a trend line representing the change of decay score over time. The slope of the trend line is the recent rate of change of decay score.

[0015] Furthermore, the process of classifying each edge region based on the attenuation score includes: Set appropriate first and second attenuation score thresholds based on historical attenuation score data. The historical attenuation score data refers to the data set of previous edge region attenuation scores. Compare the attenuation scores of the edge region with the two attenuation score thresholds. When the attenuation score of the edge region is greater than or equal to the first attenuation score threshold, the edge region is marked as a healthy region without any compensation operation. When the second attenuation score threshold is less than or equal to the attenuation score of the edge region, which is less than the first attenuation score threshold, the edge region is marked as a warning region, the attenuation trend is recorded but not compensated immediately. When the attenuation score of the edge region is less than the second attenuation score threshold, the edge region is marked as a severely attenuated region, and the emergency compensation mode is activated.

[0016] Furthermore, emergency compensation model: The compensation controller reads the real-time brightness data of all unit blocks in the severely attenuated area, generates a brightness distribution heat map through an interpolation algorithm, identifies the sub-region with the most severe brightness attenuation, and calculates the compensation intensity coefficient K=1+α×(0.6-S) based on the attenuation score, where α is the compensation gain coefficient. Through the backlight driving circuit The interface adjusts the constant current drive value of the LED string in the corresponding area, changing the current from the standard value. Upgrade to I= ×K, the source driver IC applies voltage compensation ΔV=β×(0.6-S)× to severely attenuated regions by adjusting the mapping relationship between VCOM voltage and pixel data voltage according to the compensation intensity. Where β is the voltage compensation coefficient. This is the maximum voltage corresponding to this gray level; During the compensation process, the optical sensor provides real-time feedback on the compensated brightness value. The compensation parameters are dynamically adjusted through a PID control algorithm to ensure that the difference between the compensated brightness and the central area does not exceed 5%. All compensation parameters are stored in non-volatile memory, which supports power-off saving and power-on recovery. Each compensation operation is logged for subsequent model optimization.

[0017] Furthermore, the process of obtaining historical data on the total number of attenuation cells and constructing a prediction model for the total number of attenuation cells, and then using this prediction model to predict the total number of attenuation cells in the edge region, includes: The attenuation unit block refers to a unit block whose average brightness in the edge region is lower than that in the center region. The total number of attenuation unit blocks refers to the total number of attenuation unit blocks in the edge region after Z evaluation cycles. Let Z evaluation cycles be one monitoring cycle. Factors affecting the total number of attenuation unit blocks include: the current number of attenuation unit blocks in the edge region, the current attenuation score, the rate of change of the current attenuation score, the panel operating temperature, the backlight current density, the pixel driving voltage, and the polarizer transmittance attenuation rate. The panel's operating temperature is obtained by measuring the surface temperature distribution of the panel using an infrared thermal imager. The backlight current density is obtained by calculating the ratio of LED current to light-emitting area. The pixel driving voltage is obtained by reading the value of the driving voltage register; The transmittance attenuation rate of the polarizer is obtained by comparing the current transmittance with the initial transmittance. The total number of historical attenuation unit blocks in a single edge region during different monitoring periods is obtained. The total number of historical attenuation unit blocks includes the current number of attenuation unit blocks in the single edge region before different monitoring periods, the current attenuation score, the change rate of the current attenuation score, the panel operating temperature, the backlight current density, the pixel driving voltage, the polarizer transmittance attenuation rate, and the total number of historical attenuation unit blocks in the single edge region after the monitoring period. Based on the current number of attenuation units, current attenuation score, current attenuation score change rate, panel operating temperature, backlight current density, pixel driving voltage, polarizer transmittance attenuation rate, and corresponding historical total number of attenuation units in the corresponding edge region of different historical total number of attenuation units, a prediction set of the total number of attenuation units is generated, and it is divided into training set and test set. Convolutional neural networks are constructed, with the current number of attenuation units, current attenuation score, current attenuation score change rate, panel operating temperature, backlight current density, pixel driving voltage, and polarizer transmittance attenuation rate from the total number of historical attenuation units in the training set as input data, and the corresponding total number of historical attenuation units in the training set as output data. The convolutional neural network is trained to obtain an initial convolutional neural network. The initial convolutional neural network is validated using a test set. The initial convolutional neural network whose output is less than or equal to the preset test error threshold is used as the prediction model for the total number of decaying unit blocks. At each monitoring cycle, the current number of attenuation unit blocks, current attenuation score, current attenuation score change rate, panel operating temperature, backlight current density, pixel driving voltage, and polarizer transmittance attenuation rate of each edge region are input into the attenuation unit block total number prediction model to obtain the predicted total number of attenuation unit blocks for each edge region after the monitoring cycle.

[0018] Furthermore, after the model predicts the total number of decaying unit blocks, u, in the edge region after Z evaluation periods, let z be the time of Z periods. Calculate the number of decaying unit blocks increasing by Δ=uv, and the growth rate of decaying unit blocks by r=Δ / z. Through extensive sample testing, establish a statistical relationship curve between the decay score and the proportion of decaying unit blocks. Take the number of decaying unit blocks corresponding to the decay score being equal to the second decay score threshold as the threshold a. Calculate the difference d=au between the predicted total number of decaying unit blocks and the decay unit block threshold. Calculate the decay time t=d / r (r not equal to 0) required to reach the decay unit block threshold at the current growth rate. Set a time threshold J. When the decay time t is less than the preset time threshold J, it is determined that there is a risk of rapid decay in the edge area, and the preventive compensation mode is automatically triggered. At the same time, an early warning report is generated and sent to the maintenance system.

[0019] Furthermore, the preventative compensation model: Analyze the characteristics of the current displayed content, distinguish between static and dynamic areas through image recognition algorithms, implement a dynamic load balancing strategy for areas that display static images for a long time, reduce the backlight current of the edge area by 10%-20%, and at the same time improve the grayscale value of the pixels in the area through image processing algorithms to maintain consistent visual brightness but reduce actual light output. The active cooling system in the edge area is activated. Based on the temperature gradient index (TGI) value, the speed of the micro fan is adjusted by a PID controller to keep the temperature in the edge area within 3°C of the center area. The voltage optimization algorithm is executed to find the optimal combination of backlight current and pixel voltage while ensuring brightness consistency. The genetic algorithm is used to iteratively calculate and minimize power consumption while meeting brightness requirements. The panel refresh sequence is executed automatically every 12 hours, displaying a full white screen for 30 seconds, followed by a full black screen for 30 seconds to restore the uniformity of the liquid crystal molecule arrangement; Implementing collaborative compensation between adjacent panels: when the brightness attenuation accelerates in the edge area of ​​a certain panel, the brightness of the corresponding edge of the adjacent panel is synchronously adjusted by the splicing controller to form a smooth transition and avoid visible brightness boundaries; All preventative compensation operations are performed during panel standby or low-load periods and managed through a background task scheduler to ensure that normal display functions are not affected. Compensation parameters are dynamically adjusted based on real-time monitoring data and recorded in the maintenance database for model optimization.

[0020] The technical effects and advantages of the LCD panel edge brightness attenuation compensation method of the present invention are as follows: (1) By setting a three-level response mechanism with a decay score threshold, this graded processing optimizes the allocation of system resources and avoids ineffective intervention in normal areas. At the compensation execution level, the compensation intensity coefficient is dynamically calculated based on the decay score, and the backlight driving current and pixel data voltage are adjusted synchronously. PID closed-loop control is formed through optical sensor feedback to ensure compensation accuracy. Based on the predictive model, when the system predicts that a certain edge area will reach a severe decay standard within a period below the time threshold in the future, a series of preventive measures are initiated in advance. For the special application scenario of splicing screen, a collaborative compensation strategy for adjacent panels is proposed. When the brightness of the edge of a certain panel changes due to compensation, the brightness output of the corresponding edge of the adjacent panel is adjusted synchronously through the splicing controller. The smooth transition of the brightness field is achieved by utilizing the visual characteristics of the human eye, effectively eliminating the visual abruptness of the splicing seam that may be caused by single panel compensation.

[0021] (2) By setting an attenuation score, the score can more robustly and comprehensively reflect the overall health status of the edge area. By analyzing the time series data of the attenuation score, the rate of change is calculated using the linear fitting method, the dynamic trend of attenuation is quantified, and a prediction model for the total number of attenuation unit blocks is established. This model takes the current number of attenuation unit blocks, attenuation score, rate of change, panel operating temperature, backlight current density and other multi-dimensional state parameters as inputs. It learns the complex nonlinear relationship between the attenuation score and the future attenuation development through a convolutional neural network, and outputs the predicted value of the attenuation scale after a specific period in the future. The basis for compensation decision-making is extended from the current static situation to the prediction of future evolution trends, so that the system can provide early warning and plan intervention measures before the attenuation problem becomes serious. It truly realizes predictive maintenance and greatly improves the initiative and scientific nature of maintenance work. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0024] Reference Figure 1 A method for compensating for edge brightness attenuation in an LCD panel includes the following steps: S1: Collect attenuation parameters of the edge region through the acquisition device, and normalize the attenuation parameters; S2: Generate an attenuation score for each edge region based on the processed attenuation parameters, and calculate the rate of change of the attenuation score based on the attenuation score. S3: Classify each edge region according to the attenuation score, and implement an emergency compensation mode for severely attenuated regions; S4: Obtain historical data on the total number of attenuation cells and construct a prediction model for the total number of attenuation cells. Use the prediction model to predict the total number of attenuation cells in the edge region. S5: Calculate the decay time based on the number of predicted decay unit blocks, set a time threshold, determine the risk of rapid decay in the edge region, and trigger the preventive compensation mode.

[0025] It should be further explained that in the specific implementation process, large splicing screens are made up of multiple LCD panels. The LCD panels are usually rectangular and have two short sides and two long sides. Let the area near the short side be the short edge area and the area near the long side be the long edge area. Both the short edge area and the long edge area are edge areas. The LCD panel is the center area except for the edge areas. The two long edge areas are located between the two short edge areas. The length of the short edge area is equal to the length of the long edge area, and the width of the short edge area is equal to the width of the long edge area. Each unit area can be divided into multiple unit blocks. The size of the unit block is 10x10 pixels. The width of the edge area accounts for 20% of the width of the LCD panel. The attenuation parameters refer to the brightness attenuation rate, brightness uniformity coefficient, response time attenuation rate, voltage-brightness conversion efficiency, and temperature gradient index within the edge region. The luminance of each edge region and the reference luminance of the central region are measured using a spectrocolorimeter. The luminance of each edge region is divided by the reference luminance of the central region to obtain the luminance attenuation rate L. The luminance attenuation rate is then normalized.

[0026] In the formula, To normalize the luminance decay rate, This represents the actual brightness decay rate. and These represent the lower and upper limits of the brightness decay rate, specifically 0.5 and 1, respectively. Nine points are uniformly selected within each edge region to measure brightness. The standard deviation σ and mean μ are calculated. The brightness uniformity coefficient U = 1 - (σ / μ) / 0.2 (assuming a maximum allowable non-uniformity of 20%) is then normalized.

[0027] In the formula, The normalized brightness uniformity coefficient, This represents the actual brightness uniformity coefficient; The response time of the edge region from 10% to 90% brightness is measured and compared with the standard value. That is, the response time decay rate R = standard response time / measured response time. The response time decay rate is then normalized.

[0028] In the formula, To normalize the response time decay rate, This represents the actual response time decay rate. and The lower and upper limits of the response time decay rate are 0.5 and 1.2, respectively. The brightness of the edge region is measured under a fixed voltage, and the ratio to the initial efficiency is calculated, i.e., voltage-to-brightness conversion efficiency V = current brightness / initial brightness. The voltage-to-brightness conversion efficiency is then normalized.

[0029] In the formula, To normalize the voltage-to-brightness conversion efficiency, This represents the actual voltage-to-brightness conversion efficiency. and These represent the lower and upper limits of the voltage-to-brightness conversion efficiency, specifically 0.6 and 1, respectively. Measure the temperature difference ΔT between the edge and center regions, and the temperature gradient exponent T = 1 - (ΔT / 15) (assuming a maximum allowable temperature difference of 15℃). Normalize the temperature gradient exponent.

[0030] In the formula, The normalized temperature gradient exponent, This is the actual temperature gradient exponent.

[0031] It should be further explained that, in the specific implementation process, the process of generating an attenuation score for each edge region based on the processed attenuation parameters includes: Attenuation score S:

[0032] In the formula, , , , and These are weighting coefficients, obtained from training on historical data, and set to 0.3, 0.15, 0.15, 0.25, and 0.15 respectively. If the attenuation parameters of a certain edge region are: L=0.75, U=0.6, R=0.667, V=0.8, T=0.6, then the attenuation score S of the edge region is 0.705.

[0033] It should be further explained that, in the specific implementation process, the process of calculating the rate of change of the attenuation score based on the attenuation score includes: Set an evaluation period. Each evaluation period is used to obtain a decay score for the edge region. The latest decay score value of each edge region is continuously stored to form a historical decay score sequence for that edge region. When it is necessary to calculate the current rate of change of decay score in the edge region, the decay score data of the most recent four evaluation periods in the historical decay score sequence are extracted, and the least squares method is used to perform linear fitting on the time series data to obtain a trend line representing the change of decay score over time. The slope of the trend line is the recent rate of change of decay score.

[0034] It should be further explained that, in the specific implementation process, the process of classifying each edge region based on the attenuation score includes: Set appropriate first and second attenuation score thresholds based on historical attenuation score data. The historical attenuation score data refers to the data set of previous edge region attenuation scores. Compare the attenuation scores of the edge region with the two attenuation score thresholds. When the attenuation score of the edge region is greater than or equal to the first attenuation score threshold, the edge region is marked as a healthy region without any compensation operation. When the second attenuation score threshold is less than or equal to the attenuation score of the edge region, which is less than the first attenuation score threshold, the edge region is marked as a warning region, the attenuation trend is recorded but not compensated immediately. When the attenuation score of the edge region is less than the second attenuation score threshold, the edge region is marked as a severely attenuated region, and the emergency compensation mode is activated.

[0035] It should be further explained that, in the specific implementation process, the emergency compensation model is as follows: The compensation controller reads the real-time brightness data of all unit blocks in the severely attenuated area, generates a brightness distribution heat map through an interpolation algorithm, identifies the sub-region with the most severe brightness attenuation, and calculates the compensation intensity coefficient K=1+α×(0.6-S) based on the attenuation score, where α is the compensation gain coefficient (default value 1.5). Through the backlight driving circuit The interface adjusts the constant current drive value of the LED string in the corresponding area, changing the current from the standard value. Upgrade to I= ×K, the source driver IC applies voltage compensation ΔV=β×(0.6-S)× to severely attenuated regions by adjusting the mapping relationship between VCOM voltage and pixel data voltage according to the compensation intensity. , where β is the voltage compensation coefficient (default value 0.2). This is the maximum voltage corresponding to this gray level; During the compensation process, the optical sensor provides real-time feedback on the compensated brightness value. The compensation parameters are dynamically adjusted through a PID control algorithm to ensure that the difference between the compensated brightness and the central area does not exceed 5%. All compensation parameters are stored in non-volatile memory, which supports power-off saving and power-on recovery. Each compensation operation is logged for subsequent model optimization.

[0036] It should be further explained that, in the specific implementation process, the process of obtaining historical data on the total number of attenuation cells and constructing a prediction model for the total number of attenuation cells, and then using this prediction model to predict the total number of attenuation cells in the edge region, includes: The attenuation unit block refers to a unit block whose average brightness in the edge region is lower than that in the center region. The total number of attenuation unit blocks refers to the total number of attenuation unit blocks in the edge region after Z evaluation cycles. Let Z evaluation cycles be one monitoring cycle. Factors affecting the total number of attenuation unit blocks include: the current number of attenuation unit blocks in the edge region, the current attenuation score, the rate of change of the current attenuation score, the panel operating temperature, the backlight current density, the pixel driving voltage, and the polarizer transmittance attenuation rate. The panel's operating temperature is obtained by measuring the surface temperature distribution of the panel using an infrared thermal imager. The backlight current density is obtained by calculating the ratio of LED current to light-emitting area. The pixel driving voltage is obtained by reading the value of the driving voltage register; The transmittance attenuation rate of the polarizer is obtained by comparing the current transmittance with the initial transmittance. The total number of historical attenuation unit blocks in a single edge region during different monitoring periods is obtained. The total number of historical attenuation unit blocks includes the current number of attenuation unit blocks in the single edge region before different monitoring periods, the current attenuation score, the change rate of the current attenuation score, the panel operating temperature, the backlight current density, the pixel driving voltage, the polarizer transmittance attenuation rate, and the total number of historical attenuation unit blocks in the single edge region after the monitoring period. Based on the current number of attenuation units, current attenuation score, current attenuation score change rate, panel operating temperature, backlight current density, pixel driving voltage, polarizer transmittance attenuation rate, and corresponding historical total number of attenuation units in the corresponding edge region of different historical total number of attenuation units, a prediction set of the total number of attenuation units is generated, and it is divided into training set and test set. Convolutional neural networks are constructed, with the current number of attenuation units, current attenuation score, current attenuation score change rate, panel operating temperature, backlight current density, pixel driving voltage, and polarizer transmittance attenuation rate from the total number of historical attenuation units in the training set as input data, and the corresponding total number of historical attenuation units in the training set as output data. This convolutional neural network is specifically a temporal feature extraction network. Its input data is constructed as a two-dimensional matrix of shape [T, 7], where T represents the continuous time step (value 4, representing the most recent four evaluation periods), and 7 represents the seven features at each time step: current number of decay units, current decay score, current decay score change rate, panel operating temperature, backlight current density, pixel driving voltage, and polarizer transmittance decay rate. The network contains three one-dimensional convolutional layers and two fully connected layers: the first convolutional layer uses 32 kernels of size 2 with a stride of 1 to extract short-term dependencies between adjacent time steps; the second convolutional layer uses 64 kernels of size 3 with a stride of 1 to capture medium- to long-term trends across multiple time steps; the third... The convolutional layers use 128 kernels of size 2 with a stride of 1 to further fuse multi-scale temporal features. Each convolutional layer is batch-normalized and activated using a linear rectified function. The convolutional features are transformed by a flattening layer and then input into the fully connected part: the first fully connected layer contains 256 neurons, followed by a deactivation layer with a random deactivation rate of 0.3; the second fully connected layer is the output layer, containing 1 neuron, which outputs the predicted value as a linear function. The network training uses an adaptive moment estimation algorithm optimizer with an initial learning rate of 0.001 and a loss function of mean absolute error. The training batch size is 32, and an early stopping method is used to monitor the validation set loss. Training is terminated early when the validation loss does not decrease for 15 consecutive rounds, and the model weights with the lowest validation loss are saved as the final prediction model. The convolutional neural network is trained to obtain an initial convolutional neural network. The initial convolutional neural network is validated using a test set. The initial convolutional neural network whose output is less than or equal to the preset test error threshold is used as the prediction model for the total number of decaying unit blocks. At each monitoring cycle, the current number of attenuation unit blocks, current attenuation score, current attenuation score change rate, panel operating temperature, backlight current density, pixel driving voltage and polarizer transmittance attenuation rate of each edge region are input into the attenuation unit block total number prediction model to obtain the predicted total number of attenuation unit blocks for each edge region after the monitoring cycle. In an embodiment of the present invention, the predicted total number of attenuation unit blocks in all edge regions after the monitoring period is obtained by the attenuation unit block total number prediction model. The predicted total number of attenuation unit blocks is related to the current number of attenuation unit blocks, the current attenuation score, the change rate of the current attenuation score, the panel operating temperature, the backlight current density, the pixel driving voltage, and the polarizer transmittance attenuation rate. The number of current attenuation units directly affects the total number of predicted attenuation units. The more current attenuation units there are, the more attenuation will occur in the future, and the more predicted attenuation units there will be. Therefore, the number of current attenuation units is positively correlated with the total number of attenuation units. The size of the current decay score directly affects the total number of predicted decay units. The smaller the current decay score, the faster the future decay, and the more predicted decay units there are. Therefore, the current decay score and the total number of decay units are negatively correlated. The magnitude of the current rate of change of the attenuation score directly affects the total number of predicted attenuation units. The smaller the current rate of change of the attenuation score and the larger the negative value of the rate of change, the faster the attenuation and the more predicted attenuation units there are. Therefore, the current rate of change of the attenuation score is negatively correlated with the total number of attenuation units. The panel's operating temperature directly affects the total number of predicted attenuation units. The higher the panel's operating temperature, the faster the attenuation, and the more predicted attenuation units there are. Therefore, the panel's operating temperature and the total number of attenuation units are positively correlated. The size of the backlight current density directly affects the total number of predicted attenuation units. The larger the backlight current density, the faster the attenuation and the more predicted attenuation units there are. Therefore, the backlight current density and the total number of attenuation units are positively correlated. The size of the pixel driving voltage directly affects the total number of predictive attenuation unit blocks. The larger the pixel driving voltage, the faster the attenuation and the more predictive attenuation unit blocks there are. Therefore, the pixel driving voltage and the total number of attenuation unit blocks are positively correlated. The magnitude of the polarizer's transmittance attenuation rate directly affects the total number of predicted attenuation units. The smaller the polarizer's transmittance attenuation rate, the faster the attenuation, and the more predicted attenuation units there are. Therefore, the polarizer's transmittance attenuation rate is negatively correlated with the total number of attenuation units.

[0037] It should be further explained that, in the specific implementation process, the process of calculating the decay time based on the predicted number of decay unit blocks includes: After the model predicts the total number of decaying unit blocks, outputting the predicted total number u of decaying unit blocks in the edge region after Z evaluation periods, let the time of Z periods be z. Calculate the number of decaying unit blocks increasing by Δ=uv, and the growth rate of decaying unit blocks r=Δ / z. Through extensive sample testing, establish a statistical relationship curve between the decay score and the proportion of decaying unit blocks. Take the number of decaying unit blocks corresponding to the decay score being equal to the second decay score threshold as the threshold a. Calculate the difference d=au between the predicted total number of decaying unit blocks and the decay unit block threshold. Calculate the decay time t=d / r required to reach the decay unit block threshold at the current growth rate (r is not equal to 0, i.e., the number of decaying unit blocks increasing is not equal to 0). Set a time threshold J. When the decay time t is less than the preset time threshold J, it is determined that there is a risk of rapid decay in the edge area, and the preventive compensation mode is automatically triggered. At the same time, an early warning report is generated and sent to the maintenance system.

[0038] It should be further explained that, in the specific implementation process, the preventative compensation model includes: Analyze the characteristics of the current displayed content, distinguish between static and dynamic areas through image recognition algorithms, implement a dynamic load balancing strategy for areas that display static images for a long time, reduce the backlight current of the edge area by 10%-20%, and at the same time improve the grayscale value of the pixels in the area through image processing algorithms to maintain consistent visual brightness but reduce actual light output. The active cooling system in the edge area is activated. Based on the temperature gradient index (TGI) value, the speed of the micro fan is adjusted by a PID controller to keep the temperature in the edge area within 3°C of the center area. The voltage optimization algorithm is executed to find the optimal combination of backlight current and pixel voltage while ensuring brightness consistency. The genetic algorithm is used to iteratively calculate and minimize power consumption while meeting brightness requirements. The panel refresh sequence is executed automatically every 12 hours, displaying a full white screen for 30 seconds, followed by a full black screen for 30 seconds to restore the uniformity of the liquid crystal molecule arrangement; Implementing collaborative compensation between adjacent panels: when the brightness attenuation accelerates in the edge area of ​​a certain panel, the brightness of the corresponding edge of the adjacent panel is synchronously adjusted by the splicing controller to form a smooth transition and avoid visible brightness boundaries; All preventative compensation operations are performed during panel standby or low-load periods and managed through a background task scheduler to ensure that normal display functions are not affected. Compensation parameters are dynamically adjusted based on real-time monitoring data and recorded in the maintenance database for model optimization.

[0039] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0040] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for compensating for brightness attenuation at the edge of an LCD panel, characterized in that, Includes the following steps: S1: Collect attenuation parameters of the edge region through the acquisition device, and normalize the attenuation parameters; S2: Generate an attenuation score for each edge region based on the processed attenuation parameters, and calculate the rate of change of the attenuation score based on the attenuation score. S3: Classify each edge region according to the attenuation score, and implement an emergency compensation mode for severely attenuated regions; S4: Obtain historical data on the total number of attenuation cells and construct a prediction model for the total number of attenuation cells. Use the prediction model to predict the total number of attenuation cells in the edge region. S5: Calculate the decay time based on the number of predicted decay unit blocks, set a time threshold, determine the risk of rapid decay in the edge region, and trigger the preventive compensation mode.

2. The LCD panel edge brightness attenuation compensation method according to claim 1, characterized in that, Large video walls are made up of multiple LCD panels. LCD panels are usually rectangular and have two short sides and two long sides. Let the area closer to the short side be the short edge area and the area closer to the long side be the long edge area. Both the short edge area and the long edge area are edge areas. The LCD panel is the center area except for the edge areas. The two long edge areas are located between the two short edge areas. The length of the short edge area is equal to the length of the long edge area, and the width of the short edge area is equal to the width of the long edge area. Each unit area can be divided into multiple unit blocks. The size of the unit block is 10x10 pixels. The width of the edge area accounts for 20% of the width of the LCD panel. The attenuation parameters refer to the brightness attenuation rate, brightness uniformity coefficient, response time attenuation rate, voltage-brightness conversion efficiency, and temperature gradient index within the edge region. The luminance of each edge region and the reference luminance of the central region are measured using a spectrocolorimeter. The luminance of each edge region is divided by the reference luminance of the central region to obtain the luminance attenuation rate L. The luminance attenuation rate is then normalized. In the formula, To normalize the luminance decay rate, This represents the actual brightness decay rate. and These represent the lower and upper limits of the brightness decay rate; Nine points are uniformly selected within each edge region to measure brightness. The standard deviation σ and mean μ are calculated. The brightness uniformity coefficient U = 1 - (σ / μ) / 0.

2. The brightness uniformity coefficient is then normalized. In the formula, The normalized brightness uniformity coefficient, This represents the actual brightness uniformity coefficient; The response time of the edge region from 10% to 90% brightness is measured and compared with the standard value. That is, the response time decay rate R = standard response time / measured response time. The response time decay rate is then normalized. In the formula, To normalize the response time decay rate, This represents the actual response time decay rate. and These are the lower and upper limits of the response time decay rate; The brightness of the edge region is measured under a fixed voltage, and the ratio to the initial efficiency is calculated, i.e., voltage-to-brightness conversion efficiency V = current brightness / initial brightness. The voltage-to-brightness conversion efficiency is then normalized. In the formula, To normalize the voltage-to-brightness conversion efficiency, This represents the actual voltage-to-brightness conversion efficiency. and These represent the lower and upper limits of voltage-to-brightness conversion efficiency; The temperature difference ΔT between the edge region and the center region is measured, and the temperature gradient exponent T = 1 - (ΔT / 15). The temperature gradient exponent is then normalized. In the formula, The normalized temperature gradient exponent, This is the actual temperature gradient exponent.

3. The LCD panel edge brightness attenuation compensation method according to claim 2, characterized in that, The process of generating an attenuation score for each edge region based on the processed attenuation parameters includes: Attenuation score S: In the formula, , , , and These are weighting coefficients, obtained through training based on historical data.

4. The LCD panel edge brightness attenuation compensation method according to claim 3, characterized in that, The process of calculating the rate of change of the attenuation score based on the attenuation score includes: Set an evaluation period. Each evaluation period is used to obtain a decay score for the edge region. The latest decay score value of each edge region is continuously stored to form a historical decay score sequence for that edge region. When it is necessary to calculate the current rate of change of decay score in the edge region, the decay score data of the most recent four evaluation periods in the historical decay score sequence are extracted, and the least squares method is used to perform linear fitting on the time series data to obtain a trend line representing the change of decay score over time. The slope of the trend line is the recent rate of change of decay score.

5. The LCD panel edge brightness attenuation compensation method according to claim 4, characterized in that, The process of classifying each edge region based on the attenuation score includes: Set appropriate first and second attenuation score thresholds based on historical attenuation score data. The historical attenuation score data refers to the data set of previous edge region attenuation scores. Compare the attenuation scores of the edge region with the two attenuation score thresholds. When the attenuation score of the edge region is greater than or equal to the first attenuation score threshold, the edge region is marked as a healthy region without any compensation operation. When the second attenuation score threshold is less than or equal to the attenuation score of the edge region, which is less than the first attenuation score threshold, the edge region is marked as a warning region, the attenuation trend is recorded but not compensated immediately. When the attenuation score of the edge region is less than the second attenuation score threshold, the edge region is marked as a severely attenuated region, and the emergency compensation mode is activated.

6. The LCD panel edge brightness attenuation compensation method according to claim 5, characterized in that, Emergency Compensation Mode: The compensation controller reads the real-time brightness data of all unit blocks in the severely attenuated area, generates a brightness distribution heat map through an interpolation algorithm, identifies the sub-region with the most severe brightness attenuation, and calculates the compensation intensity coefficient K=1+α×(0.6-S) based on the attenuation score, where α is the compensation gain coefficient. Through the backlight driving circuit The interface adjusts the constant current drive value of the LED string in the corresponding area, changing the current from the standard value. Upgrade to I= ×K, the source driver IC applies voltage compensation ΔV=β×(0.6-S)× to severely attenuated regions by adjusting the mapping relationship between VCOM voltage and pixel data voltage according to the compensation intensity. Where β is the voltage compensation coefficient. This is the maximum voltage corresponding to this gray level; During the compensation process, the optical sensor provides real-time feedback on the compensated brightness value. The compensation parameters are dynamically adjusted through a PID control algorithm to ensure that the difference between the compensated brightness and the central area does not exceed 5%. All compensation parameters are stored in non-volatile memory, which supports power-off saving and power-on recovery. Each compensation operation is logged for subsequent model optimization.

7. The LCD panel edge brightness attenuation compensation method according to claim 6, characterized in that, The process of obtaining historical data on the total number of attenuation cells and constructing a prediction model for the total number of attenuation cells, and then using this prediction model to predict the total number of attenuation cells in the edge region, includes: The attenuation unit block refers to a unit block whose average brightness in the edge region is lower than that in the center region. The total number of attenuation unit blocks refers to the total number of attenuation unit blocks in the edge region after Z evaluation cycles. Let Z evaluation cycles be one monitoring cycle. Factors affecting the total number of attenuation unit blocks include: the current number of attenuation unit blocks in the edge region, the current attenuation score, the rate of change of the current attenuation score, the panel operating temperature, the backlight current density, the pixel driving voltage, and the polarizer transmittance attenuation rate. The panel's operating temperature is obtained by measuring the surface temperature distribution of the panel using an infrared thermal imager. The backlight current density is obtained by calculating the ratio of LED current to light-emitting area. The pixel driving voltage is obtained by reading the value of the driving voltage register; The transmittance attenuation rate of the polarizer is obtained by comparing the current transmittance with the initial transmittance. The total number of historical attenuation unit blocks in a single edge region during different monitoring periods is obtained. The total number of historical attenuation unit blocks includes the current number of attenuation unit blocks in the single edge region before different monitoring periods, the current attenuation score, the change rate of the current attenuation score, the panel operating temperature, the backlight current density, the pixel driving voltage, the polarizer transmittance attenuation rate, and the total number of historical attenuation unit blocks in the single edge region after the monitoring period. Based on the current number of attenuation units, current attenuation score, current attenuation score change rate, panel operating temperature, backlight current density, pixel driving voltage, polarizer transmittance attenuation rate, and corresponding historical total number of attenuation units in the corresponding edge region of different historical total number of attenuation units, a prediction set of the total number of attenuation units is generated, and it is divided into training set and test set. Convolutional neural networks are constructed, with the current number of attenuation units, current attenuation score, current attenuation score change rate, panel operating temperature, backlight current density, pixel driving voltage, and polarizer transmittance attenuation rate from the total number of historical attenuation units in the training set as input data, and the corresponding total number of historical attenuation units in the training set as output data. The convolutional neural network is trained to obtain an initial convolutional neural network. The initial convolutional neural network is validated using a test set. The initial convolutional neural network whose output is less than or equal to the preset test error threshold is used as the prediction model for the total number of decaying unit blocks. The convolutional neural network is a deep neural network comprising an input layer, at least two one-dimensional convolutional layers, at least one fully connected layer, and an output layer. The input layer receives a time series input matrix consisting of seven features: the current number of attenuation unit blocks, the current attenuation score, the rate of change of the current attenuation score, the panel operating temperature, the backlight current density, the pixel driving voltage, and the polarizer transmittance attenuation rate. The first dimension of the matrix is ​​the time step, and the second dimension is the number of features. The at least two one-dimensional convolutional layers sequentially extract temporal features from the input matrix, with the first convolutional layer extracting short-period variation features and the second convolutional layer extracting long-period variation features. The at least one fully connected layer maps the features extracted by the convolutional layer to a high-dimensional space; The output layer outputs the total number of predicted decay unit blocks in the edge region after the next Z evaluation cycles; The convolutional neural network uses a linear rectified function as the activation function and an adaptive moment estimation algorithm as the optimizer for training. The training batch size is 32 or 64, the initial learning rate is 0.001, and an exponential decay strategy is adopted. The convolutional neural network is trained using the training set and the trained network is validated using the test set. When the average absolute percentage error of the prediction is less than or equal to a preset test error threshold, the prediction model for the total number of decay unit blocks is obtained. At each monitoring cycle, the current number of attenuation unit blocks, current attenuation score, current attenuation score change rate, panel operating temperature, backlight current density, pixel driving voltage, and polarizer transmittance attenuation rate of each edge region are input into the attenuation unit block total number prediction model to obtain the predicted total number of attenuation unit blocks for each edge region after the monitoring cycle.

8. The LCD panel edge brightness attenuation compensation method according to claim 7, characterized in that, The process of calculating the decay time based on the predicted number of decay unit blocks includes: After the model predicts the total number of decaying unit blocks, u, in the edge region after Z evaluation periods, let z be the time of Z periods. Calculate the number of decaying unit blocks increasing by Δ=uv and the growth rate of decaying unit blocks r=Δ / z. Through extensive sample testing, establish a statistical relationship curve between the decay score and the proportion of decaying unit blocks. Take the number of decaying unit blocks corresponding to the decay score being equal to the second decay score threshold as the threshold a. Calculate the difference d=au between the predicted total number of decaying unit blocks and the decay unit block threshold. Calculate the decay time t=d / r (r not equal to 0) required to reach the decay unit block threshold at the current growth rate. Set a time threshold J. When the decay time t is less than the preset time threshold J, it is determined that there is a risk of rapid decay in the edge area, and the preventive compensation mode is automatically triggered. At the same time, an early warning report is generated and sent to the maintenance system.

9. The LCD panel edge brightness attenuation compensation method according to claim 8, characterized in that, Preventive compensation model: Analyze the characteristics of the current displayed content, distinguish between static and dynamic areas through image recognition algorithms, implement a dynamic load balancing strategy for areas that display static images for a long time, reduce the backlight current of the edge area by 10%-20%, and at the same time improve the grayscale value of the pixels in the area through image processing algorithms to maintain consistent visual brightness but reduce actual light output. The active cooling system in the edge area is activated. Based on the temperature gradient index (TGI) value, the speed of the micro fan is adjusted by a PID controller to keep the temperature in the edge area within 3°C of the center area. The voltage optimization algorithm is executed to find the optimal combination of backlight current and pixel voltage while ensuring brightness consistency. The genetic algorithm is used to iteratively calculate and minimize power consumption while meeting brightness requirements. The panel refresh sequence is executed automatically every 12 hours, displaying a full white screen for 30 seconds, followed by a full black screen for 30 seconds to restore the uniformity of the liquid crystal molecule arrangement; Implementing collaborative compensation between adjacent panels: when the brightness attenuation accelerates in the edge area of ​​a certain panel, the brightness of the corresponding edge of the adjacent panel is synchronously adjusted by the splicing controller to form a smooth transition and avoid visible brightness boundaries; All preventative compensation operations are performed during panel standby or low-load periods and managed through a background task scheduler to ensure that normal display functions are not affected. Compensation parameters are dynamically adjusted based on real-time monitoring data and recorded in the maintenance database for model optimization.

Images