OLED panel demura compensation method and system

Abstract

The application discloses an OLED panel Demura compensation method and system, and relates to the technical field of display, wherein multi-dimensional characteristic data of an OLED panel is adaptively clustered to obtain an optimal partition number, a partition boundary and characteristic statistical parameters of each partition; based on the characteristic statistical parameters of each partition, luminance compensation coefficients and photographing parameters of each partition are respectively calculated, and a compensation parameter set of each partition is established; based on the partition boundary, distance weighted linear interpolation is adopted to smoothly transition compensation parameters at the boundary of adjacent partitions in the compensation parameter set, and a full-panel compensation parameter matrix is obtained; each partition is assigned a unique identifier, and a mapping relationship between the unique identifier and the compensation parameters of the corresponding partition in the full-panel compensation parameter matrix is established and stored. The application solves the problem of excessive compensation or insufficient compensation caused by differences in panel film thickness and electrical parameters at different positions of a large panel in the prior art.

Description

Technical Field

[0001] This invention relates to the field of display technology, and in particular to a method and system for Demura compensation in OLED panels. Background Technology

[0002] In the production process of OLED panels, color or monochrome cameras are typically used to photograph specific display images (such as images with different brightness and grayscale combinations, such as 500nit L64, 70nit L16, and 2nit L32). The captured images are then processed by a driver IC to achieve color and brightness balance correction, thus correcting for display uniformity. Currently, all panels of the same model are optimized using a unified Demura algorithm.

[0003] However, due to the center-edge effect of the vapor deposition process in OLED panel manufacturing, panels at different locations on a large panel exhibit differences in parameters such as film thickness, threshold voltage (Vth), and subthreshold swing (SS). When a uniform Demura algorithm is used to compensate all panels, panels with better characteristics may be overcompensated, while panels with poorer characteristics may be undercompensated. This can lead to abnormal display effects in some panels and affect product yield. Summary of the Invention

[0004] To overcome the shortcomings of the prior art, the present invention provides an OLED panel Demura compensation method and system, which can adopt a zoned automatic Demura process for all panels. Areas with large differences in film thickness and electrical properties are identified in advance and zoned adjustments are performed. During the Demura production process, the ID is identified and different versions of the shooting / compensation parameters are called. It can quickly and effectively compensate for the visual effects of panels with all characteristics without increasing equipment / manpower.

[0005] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions: In a first aspect, the present invention provides a method for Demura compensation in an OLED panel, comprising: Acquire multi-dimensional characteristic data of the OLED panel and integrate the multi-dimensional characteristic data into a multi-dimensional characteristic data matrix; Adaptive clustering is performed on the multi-dimensional characteristic data matrix to obtain the optimal number of partitions, partition boundaries, and characteristic statistical parameters of each partition; Based on the characteristic statistical parameters of each zone, the brightness compensation coefficient and shooting parameters of each zone are calculated respectively, and a compensation parameter set for each zone is established. Based on the partition boundaries, the compensation parameters at adjacent partition boundaries in the compensation parameter set are processed by distance-weighted linear interpolation to achieve a smooth transition, thus obtaining the full-board compensation parameter matrix. Each partition is assigned a unique identifier, and a mapping relationship is established between the unique identifier and the corresponding partition's compensation parameter in the full-board compensation parameter matrix, and then stored.

[0006] A further technical solution is that the multi-dimensional characteristic data includes film thickness distribution data, electrical parameter data, brightness uniformity data, and color uniformity data.

[0007] Further technical solutions yield the optimal number of partitions, partition boundaries, and characteristic statistical parameters for each partition, specifically including: Normalize the data in each dimension of the multidimensional characteristic data matrix to obtain normalized multidimensional characteristic data; The K-means clustering algorithm is used to cluster the normalized multi-dimensional characteristic data, automatically determine the optimal number of partitions, and obtain the clustering results; Based on the clustering results, the boundary coordinates of each partition are determined, and the characteristic statistical parameters of film thickness, electrical parameters, brightness uniformity and color uniformity in each partition are statistically analyzed.

[0008] A further technical solution is to introduce a partition confidence assessment mechanism to quantitatively assess the reliability of partitions in the clustering results.

[0009] Further technical solutions include establishing compensation parameter sets for each partition, specifically including: Based on the brightness uniformity data of each zone, the brightness compensation coefficient of each zone is calculated using a brightness compensation model. The panel of each zone is divided into grids, and the compensation parameters are calculated independently for each grid. Piecewise linear fitting is performed on the compensation parameters under different gray levels to obtain the brightness compensation coefficient matrix of each zone. Based on the differences in film thickness and electrical parameters of each zone, differentiated imaging parameters for each zone are determined, including exposure time, gain, and test screen configuration. The brightness compensation coefficient matrix of each zone is summarized with the shooting parameters to form the compensation parameter set of each zone.

[0010] A further technical solution is that the brightness compensation model is expressed as follows:

[0011] in, For position The output brightness value after compensation. For position The original input brightness value at that location, The target is a uniform brightness reference value. For position The measured brightness value at the location.

[0012] Further technical solutions yield a full-board compensation parameter matrix, specifically including: The boundary transition region between adjacent partitions is determined based on the partition boundaries. For each pixel within the boundary transition area, a corresponding weight coefficient is calculated based on its distance to the centers of the two adjacent partitions. Based on the weighting coefficients, the compensation parameters of two adjacent partitions in the compensation parameter set are weighted linearly interpolated to obtain the transition compensation value of each pixel in the boundary transition region. The transition compensation values ​​of each pixel within the boundary transition area are combined with the compensation parameters of each partition core area to obtain the full-board compensation parameter matrix.

[0013] In a second aspect, the present invention provides an OLED panel Demura compensation system, comprising: The data acquisition module is configured to: acquire multi-dimensional characteristic data of the OLED panel and integrate the multi-dimensional characteristic data into a multi-dimensional characteristic data matrix; The partitioning decision module is configured to: perform adaptive clustering on the multi-dimensional characteristic data matrix to obtain the optimal number of partitions, partition boundaries, and characteristic statistical parameters of each partition; The compensation parameter generation module is configured to: calculate the brightness compensation coefficient and shooting parameters of each zone based on the characteristic statistical parameters of each zone, and establish a compensation parameter set for each zone; The boundary smoothing module is configured to: based on the partition boundary, perform a distance-weighted linear interpolation method on the compensation parameters at adjacent partition boundaries in the compensation parameter set to smooth the transition, thereby obtaining the full-board compensation parameter matrix; The ID mapping module is configured to: assign a unique identifier to each partition, establish a mapping relationship between the unique identifier and the corresponding partition's compensation parameter in the full-board compensation parameter matrix, and store the mapping.

[0014] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of an OLED panel Demura compensation method as described in the first aspect.

[0015] Fourthly, the present invention provides a computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the OLED panel Demura compensation method as described in the first aspect.

[0016] The above one or more technical solutions have the following beneficial effects: This invention replaces the manually preset fixed partitioning scheme with an adaptive clustering algorithm, which can automatically determine the optimal number and boundaries of partitions based on the actual characteristic data of the panel, adapting to changes in the process capabilities of different batches. By introducing a distance-weighted linear interpolation method at the boundaries of adjacent partitions, the compensation parameters are smoothly transitioned, eliminating abrupt changes in compensation at the boundaries of regions. By monitoring the panel characteristic data in real time and automatically triggering the re-adjustment process of compensation parameters when the data exceeds the threshold, self-learning optimization is achieved, further improving the compensation accuracy. Through the mapping mechanism between panel ID and compensation parameter version, rapid and accurate compensation for panels with various characteristics can be achieved without increasing additional equipment investment and production time, significantly improving product yield and display uniformity.

[0017] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0018] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0019] Figure 1 This is a flowchart of an OLED panel Demura compensation method according to an embodiment of the present invention; Figure 2 This is a flowchart of the data acquisition process according to an embodiment of the present invention; Figure 3 This is a flowchart of the partitioning process in an embodiment of the present invention. Detailed Implementation

[0020] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0021] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0022] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0023] Terminology Explanation: Demura: refers to the brightness uniformity correction technology for display panels, used to eliminate brightness unevenness (Mura defects) that occurs during the panel manufacturing process.

[0024] Film thickness: refers to the thickness of the organic light-emitting layer material in an OLED panel, measured in nanometers (nm). Its uniformity directly affects the luminous efficiency and brightness performance of different areas of the panel.

[0025] Vth: refers to the threshold voltage, which is the minimum gate voltage required for a thin-film transistor (TFT) to start conducting. The unit is volt (V). The higher the Vth, the lower the driving efficiency.

[0026] SS: refers to subthreshold swing, a parameter characterizing the switching characteristics of a thin-film transistor, measured in millivolts per ten-fold change in current (mV / dec). The smaller the SS, the better the switching characteristics.

[0027] In the production and testing of OLED panels, to ensure high-quality display performance, specific techniques are typically employed to precisely test and optimize key display images. Specifically, this involves photographing representative key display images using color or monochrome cameras. These images include combinations of brightness and grayscale, such as 500 nits L64, 70 nits L16, and 2 nits L32. These specific images are chosen because they comprehensively reflect the display performance of the OLED panel under different brightness and grayscale conditions.

[0028] After the photo is taken, the captured image is further processed using a driver integrated circuit (IC). Specifically, this involves a detailed balancing and improvement of the overall color brightness of the image, primarily achieved through calculations of brightness grayscale color coordinates. This calculation allows for precise analysis of the color brightness distribution in various areas of the image, and based on the analysis results, the panel's display parameters are adjusted to achieve more uniform color brightness and a more ideal display effect.

[0029] It is worth noting that in actual production, the same Demura imaging and algorithm solution will be used for all panels of a product.

[0030] However, due to the different positions of different panels on the large board, and the differences in film thickness and derived electrical properties (SS, Vth) caused by process capabilities, the visual effect of panels at different positions on the large board before Demura varies significantly. Even though Demura attempts to cover the entire large board with full mapping for debugging, over-compensation or under-compensation can occur for the best and worst panels, leading to abnormal visual effects. Therefore, this invention proposes a differentiated Demura compensation scheme based on multi-dimensional parameter measurement, adaptive clustering partitioning, and panel ID mapping to address the aforementioned technical shortcomings.

[0031] To address the overcompensation or undercompensation issues caused by the uniform Demura algorithm in existing technologies, this invention provides a Demura compensation method for OLED panels. This method first acquires multi-dimensional data on the OLED panel, including film thickness distribution, electrical parameters, brightness uniformity, and color uniformity, using precision measurement equipment. Then, an adaptive clustering algorithm is used to perform cluster analysis on the multi-dimensional data, automatically determining the optimal number of partitions and partition boundaries. Differentiated compensation parameters are calculated for the characteristics of each partition, and distance-weighted linear interpolation is introduced at the boundaries of adjacent partitions to achieve a smooth transition. A unique ID is assigned to each partition, and a mapping database is established. During production, the corresponding compensation parameters are automatically retrieved by scanning the panel's QR code. Simultaneously, a dynamic update mechanism is introduced, automatically triggering re-adjustment when batch characteristic data exceeds a threshold, achieving self-learning optimization. The core difference between this solution and existing technologies lies in replacing the manually preset uniform scheme with multi-dimensional adaptive clustering partitioning, combined with dynamic updates and a smooth boundary transition mechanism, to achieve precise differentiated compensation.

[0032] Examples of environments in which embodiments of the present invention can be implemented include OLED panels, precision measurement systems, partition decision servers, and production line SIM systems.

[0033] The OLED panel is a trial production panel that is to undergo Demura compensation debugging. It contains multiple panel units in different locations, and each panel unit is marked with a unique QR code.

[0034] The precision measurement system includes a film thickness measurement module, an electrical testing module, an optical inspection module, and a morphology analysis module. These modules are used to collect film thickness mapping data, electrical mapping data, optical mapping data, and morphology mapping data of the OLED panel, respectively. The measurement data from each module are aggregated into a multimodal data fusion node to form a comprehensive characteristic database for use by the difference analysis engine.

[0035] In a preferred embodiment of the present invention, the film thickness measurement module can be implemented using an elliptic polarization spectrometer (such as the Woollam M-2000 series, with a measurement accuracy of ±0.1nm) or a white light interferometer, for high-precision scanning measurement of the organic light-emitting layer film thickness at various locations on the OLED panel; the electrical testing module can be implemented using a probe testing station (such as Cascade Microtech) in conjunction with an OLED photoelectric performance analysis system (such as OLED-200), for measuring the distribution of electrical parameters such as threshold voltage Vth and subthreshold swing SS of the TFT devices at various locations; the optical detection module can be implemented using an image luminance colorimeter (such as Konica Minolta CA-2500) in a standard darkroom environment (temperature 25℃, humidity 50%), for measuring the brightness uniformity and color uniformity data at various locations; the morphology analysis module can be implemented using an atomic force microscope (AFM) or a profilometer, for acquiring spatial distribution data of the panel surface morphology to assist in identifying areas of process abnormalities. The precision measurement system can also be expanded to include other measurement equipment such as a profilometer and a white light interferometer.

[0036] The partition decision server is connected to the precision measurement system via a data bus, receives comprehensive characteristic data, executes an adaptive clustering algorithm, outputs a partitioning scheme, and generates compensation parameters for each partition. The partition decision server can also be expanded into a cloud server or an edge computing node.

[0037] The production line SIM system is connected to the partition decision server via a network. During the mass production process, it scans the panel QR code to identify the panel ID, retrieves the corresponding version of the compensation parameters from the database, and drives the compensation execution device to complete Demura compensation.

[0038] The data collection and processing in this embodiment strictly comply with relevant laws and regulations and are used only for OLED panel Demura compensation, not for any other unauthorized purposes. It should be understood that the above example environment is described for illustrative purposes only and does not imply any limitation on the scope of this disclosure. This invention can also be applied to uniformity correction scenarios for other types of display panels.

[0039] Example 1 like Figure 1 As shown, this embodiment discloses a Demura compensation method for OLED panels, which includes the following steps: S1: Obtain multi-dimensional characteristic data of the OLED panel and integrate the multi-dimensional characteristic data into a multi-dimensional characteristic data matrix; In this embodiment, the purpose of this step is to obtain physical and photoelectric characteristic data of each region of the OLED panel, so as to provide a multi-dimensional data foundation for subsequent zoning decisions.

[0040] Film thickness distribution data refers to the spatial distribution matrix formed by measuring the thickness values ​​of the organic light-emitting layer at multiple locations on the OLED panel, reflecting the uniformity of the evaporation process. Electrical parameter data refers to the spatial distribution data of the threshold voltage Vth and subthreshold swing SS obtained through electrical testing, reflecting the spatial distribution pattern of TFT device characteristics. Brightness uniformity data refers to the spatial distribution data of brightness values ​​at each measurement point under the condition of uniform brightness in a full-screen display. Chromaticity uniformity data refers to the spatial distribution data of chromaticity coordinates at each measurement point when displaying a pure color image.

[0041] like Figure 2 As shown, the data comes from various measurement modules in the precision measurement system. Film thickness distribution data was measured by an ellipsometric spectrometer (such as the Woollam M-2000 series) with a measurement accuracy of ±0.1 nm; electrical parameter data was measured by a probe test stage (such as Cascade Microtech) in conjunction with an OLED photoelectric performance analysis system (such as OLED-200); brightness uniformity data and color uniformity data were measured by an image luminance colorimeter (such as the Konica Minolta CA-2500) in a standard darkroom environment (temperature 25℃, humidity 50%).

[0042] The specific processing procedure is as follows: perform a full-area grid scan on the OLED panel and take multiple measurements at each grid node; calculate the film thickness, Vth value, SS value, brightness uniformity index, and color uniformity index at each measurement point.

[0043] Brightness uniformity index is expressed as:

[0044] in, The lowest brightness value within the measurement area. The highest brightness value within the measurement area. This represents the percentage of brightness uniformity.

[0045] Color uniformity index is expressed as:

[0046] in, , Here are the chromaticity coordinates of the measurement point. , For reference chromaticity coordinates, This represents the chromaticity deviation value. The data from each dimension are integrated into a multimodal data matrix and transmitted to the partitioning decision server.

[0047] Furthermore, the collected film thickness distribution data were statistically analyzed and processed. The specific calculation steps are as follows: Calculate the average film thickness at each measurement point: ,in For the first The film thickness value at each measurement point The total number of measurement points. The average film thickness value of the measurement area; calculate the standard deviation of the film thickness: ,in The standard deviation of the film thickness distribution reflects the degree of dispersion in the film thickness distribution; film thickness uniformity is calculated as follows: ,in This represents the percentage of film thickness uniformity; the closer the value is to 100%, the more uniform the film thickness distribution. A threshold judgment is performed: a process-allowed tolerance range is set (preferably ±5%). Areas exceeding the threshold are marked as abnormal areas, triggering subsequent fine measurement processes; correlation analysis is performed: the relationship between film thickness difference and panel position coordinates and vapor deposition process parameters is analyzed, the spatial distribution pattern of center-edge effect is identified, and quantitative basis is provided for zoning decision-making.

[0048] Furthermore, the electrical parameter measurement also includes current-voltage characteristic testing: measuring the current and luminance values ​​at each measurement point under different driving voltages, analyzing the changes in luminous efficiency in each region, and calculating the efficiency index. ,in For luminous efficiency, To measure the brightness value of the point, This refers to the drive current value; it identifies areas of abnormal efficiency, i.e. Regions whose values ​​deviate significantly from the central region's baseline efficiency (deviation exceeding ±10%) are included in the reference data for zoning decisions to help determine the compensation intensity requirements for that region.

[0049] Output a multi-dimensional characteristic data matrix, including film thickness distribution, Vth distribution, SS distribution, brightness uniformity distribution, and chromaticity uniformity distribution, which is used as input for cluster analysis in step S2.

[0050] This step enables multi-dimensional and high-precision panel characteristic data collection, providing a comprehensive and reliable data foundation for subsequent adaptive clustering partitioning, and overcoming the limitation of existing technologies that rely solely on a single parameter to determine partitioning.

[0051] Preferably, the measurement density is 100 to 1000 measurement points per square centimeter, with the measurement density appropriately increased at the edges and corners of the panel to accurately capture areas of characteristic gradient change.

[0052] S2: Adaptive clustering is performed on the multi-dimensional characteristic data matrix to obtain the optimal number of partitions, partition boundaries, and characteristic statistical parameters of each partition; In this embodiment, the purpose of this step is to automatically determine the optimal number of partitions and partition boundaries based on multi-dimensional characteristic data and an adaptive clustering algorithm, replacing the manual preset fixed partitioning scheme, and achieving an adaptive response to changes in process capability.

[0053] The acquired multi-dimensional characteristic data matrix was preprocessed, with data on film thickness, Vth, SS, brightness uniformity, and chromaticity uniformity being normalized to eliminate the influence of dimensional differences on the clustering results. The K-means clustering algorithm was then used to perform cluster analysis on the normalized multi-dimensional data. The silhouette coefficient was used to evaluate the clustering quality under different numbers of clusters, automatically determining the optimal number of partitions. Figure 3 As shown, you can also choose between a 3×3 grid partitioning mode and a center-edge partitioning mode, which can be flexibly selected according to the actual situation.

[0054] Typically, the partitioning result is four zones: the central region (Zone A), the middle annular region (Zone B), the edge region (Zone C), and the four corner regions (Zone D). The parameter characteristics of each typical zone are as follows: Zone A has a film thickness range of 100 to 120 nm, a target value deviation of ±3%, a Vth range of 1.8 to 2.0 V, an SS range of 60 to 80 mV / dec, a brightness uniformity of not less than 95%, and a color uniformity of... Less than 0.003; Zone B film thickness range is 90 to 110 nm, target value deviation is ±8%, Vth range is 2.0 to 2.3 V, SS range is 80 to 100 mV / dec, brightness uniformity is 90% to 95%, and color uniformity is... Less than 0.005; Zone C film thickness range is 80 to 100 nm, target value deviation ±15%, Vth range is 2.3 to 2.6 V, SS range is 100 to 120 mV / dec, brightness uniformity is 85% to 90%, and color uniformity is... Less than 0.008; Zone D film thickness range is 70 to 90 nm, target value deviation ±20%, Vth range is 2.6 to 3.0 V, SS range is 120 to 150 mV / dec, brightness uniformity is 80% to 85%, and color uniformity is... Less than 0.010.

[0055] Output the partitioning scheme, including the boundary coordinates of each partition, the number of partitions, and the characteristic statistical parameters of each partition.

[0056] This step introduces an adaptive clustering algorithm to replace manually preset fixed partitions, which can automatically adjust the partitioning strategy according to the actual process capability, adapt to the fluctuation of process parameters between different batches, and improve the partitioning accuracy and the adaptability of the scheme.

[0057] Furthermore, this step introduces a partition confidence assessment mechanism to quantitatively evaluate the reliability of the clustering partitioning results. Specifically, it calculates the clustering confidence score for each data point. Its calculation formula is

[0058] in, The distance from a data point to the center of its cluster is the Euclidean distance. The Euclidean distance from the data point to the second nearest cluster center. This represents the partition confidence score for that data point, with a value ranging from 0 to 1. The closer the value is to 1, the more certain the partition a data point belongs to.

[0059] For confidence scores Panel areas corresponding to data points below a preset threshold (preferably 0.6) are automatically marked as "low-confidence areas" by the system, triggering a secondary fine-tuning measurement process for these areas. This increases the measurement density to 500 to 1000 measurement points per square centimeter, re-collects data, and performs local clustering correction until the confidence scores of all data points within the area reach the threshold requirement. This mechanism effectively avoids misclassification of partitions caused by measurement noise or ambiguity in process transition areas, improving the reliability and accuracy of the partitioning scheme.

[0060] Preferably, the confidence threshold can be dynamically adjusted according to the product yield requirements. For products with high yield requirements, the threshold can be increased to 0.75 to obtain more accurate partitioning results.

[0061] S3: Based on the characteristic statistical parameters of each zone, calculate the brightness compensation coefficient and shooting parameters of each zone respectively, and establish the compensation parameter set of each zone; In this embodiment, the purpose of this step is to calculate differentiated brightness compensation coefficients and imaging parameters for each zone based on the characteristics determined in step S2, and to establish a compensation parameter set for each zone.

[0062] The compensation parameters are calculated using a brightness compensation model:

[0063] in, For position The output brightness value after compensation, that is, the target brightness output by the driver chip after Demura correction; For position The original input brightness value at the location, that is, the measured brightness of the panel without compensation; To achieve uniform brightness, the average brightness of the central area (Zone A) of the panel is taken as the ideal target value for the entire panel. For position The measured brightness value at the location is obtained by the image luminance colorimeter in step S1; The horizontal coordinates of the pixel; These are the vertical coordinates of the pixel.

[0064] The OLED panel is divided into 16×16 or 32×32 grids, and compensation parameters are calculated independently for each grid. A linear fitting method is used for the compensation parameters at different gray levels (e.g., a standard gray level sequence from 0 to 255 levels, with 16-level intervals). Perform piecewise fitting, where To compensate for parameter values, Grayscale values, The slope of the linear fit. The intercept is used for linear fitting. The compensation data is burned into the panel driver chip, and a second image acquisition is performed to verify the compensation effect. If color difference... If the uniformity standard of less than 2 is not met, the compensation parameters should be repeatedly adjusted until the requirements are met.

[0065] The brightness gain adjustment strategies for each zone are as follows: Zone A brightness gain is 1.0, using the standard gamma curve; Zone B brightness gain is 1.05 to 1.10, with slight gamma adjustment; Zone C brightness gain is 1.10 to 1.15, with medium gamma adjustment; Zone D brightness gain is 1.20 to 1.30, with significant gamma adjustment.

[0066] The differentiated adjustment strategies for Vth compensation in each zone are as follows: The compensation goal is to make Vth in each zone approach the target value (preferably 2.0V); slight compensation is applied to the center zone (Zone A), with a compensation voltage range of 0 to 0.1V; moderate compensation is applied to the middle ring zone (Zone B), with a compensation voltage range of 0.1 to 0.2V; strong compensation is applied to the edge zone (Zone C), with a compensation voltage range of 0.2 to 0.3V; and the strongest compensation is applied to the four corner zones (Zone D), with a compensation voltage range of 0.3 to 0.4V. The differentiated adjustment strategies for SS compensation in each zone are as follows: The SS compensation principle is to improve the TFT switching characteristics by adjusting the gate voltage slope; the SS in the center zone (Zone A) is close to the ideal value, requiring only slight adjustment; the SS in the middle ring zone (Zone B) and the edge zone (Zone C) is relatively large, requiring moderate gate voltage waveform adjustment; the SS in the four corner zones (Zone D) is the largest, requiring significant gate voltage waveform adjustment to improve switching characteristics and reduce drive power consumption.

[0067] The differentiated design strategy for the photographic images in each zone is as follows: The central area (Zone A) uses a standard test image combination, including a pure white image (for evaluating brightness uniformity), a grayscale step image (for evaluating the performance of each grayscale level), and a solid color image (for evaluating color uniformity); the edge areas (Zone C) and the four corner areas (Zone D) use an enhanced test image combination, including a low-brightness image (for evaluating the performance of Mura defects at low brightness), a high-contrast image (for evaluating edge effects), and a dynamic image (for evaluating response characteristics); the central ring area (Zone B) uses a mixed combination of standard and enhanced test images. The exposure time, gain, and white balance parameters of the photographic images in each zone are configured differently according to the brightness gain of each zone to ensure that the signal-to-noise ratio of the photographed images meets the compensation parameter calculation requirements.

[0068] Output the complete set of compensation parameters for each zone, including the brightness compensation coefficient matrix, shooting parameter configuration, and gamma curve adjustment parameters.

[0069] This step, by establishing an accurate brightness compensation model and a piecewise fitting method, achieves precise calculation of compensation parameters for each grayscale level in each zone, solving the problem of insufficient compensation accuracy in existing technologies and improving the Demura correction effect.

[0070] Furthermore, this step introduces a joint optimization model for chromaticity compensation, simultaneously correcting chromaticity deviation on top of luminance compensation to achieve joint compensation of luminance and chromaticity. Chromaticity compensation employs the chromaticity deviation correction method in the CIE 1976 uniform color space, calculating a chromaticity compensation vector for each pixel. The calculation formula is as follows:

[0071]

[0072] in, , The target chromaticity coordinates are (the average chromaticity coordinates of Zone A in the center area of ​​the panel are taken as the reference). , For position The measured chromaticity coordinates at the location, , They are respectively direction and The chromaticity compensation amount in the direction.

[0073] The chroma gain adjustment strategy for each zone is linked to the luminance gain: Zone A has a chroma compensation of almost zero, requiring only minor adjustments; Zone B has a slight chroma compensation adjustment, with a compensation range not exceeding 0.003; Zone C has a medium chroma compensation adjustment, with a compensation range not exceeding 0.006; and Zone D has a relatively strong chroma compensation adjustment, with a compensation range not exceeding 0.010. After joint optimization of luminance and chroma compensation, the final verification metric requires color difference... Less than 2 and A value less than 0.005 ensures that the overall display effect after compensation meets quality standards in both brightness uniformity and color consistency.

[0074] Preferably, the mesh size is 32×32, and the number of grayscale fitting nodes is no less than 16, so as to achieve the optimal balance between compensation accuracy and computational efficiency.

[0075] S4: Based on the partition boundary, the compensation parameters at adjacent partition boundaries in the compensation parameter set are processed by distance-weighted linear interpolation to achieve a smooth transition, thus obtaining the full-board compensation parameter matrix; In this embodiment, the purpose of this step is to eliminate the visual boundary lines caused by abrupt changes in compensation parameters at the boundaries of adjacent partitions, thereby improving the visual consistency of display uniformity.

[0076] Within the boundary transition region between adjacent partitions, a weight coefficient is calculated for each pixel based on its distance to the center of each adjacent partition. Let the distance from the pixel to the center of partition 1 be... The distance to the center of partition 2 is The corresponding weighting coefficients are respectively and ,in The weighting coefficients for the compensation value of pixel points in partition 1. The weighting coefficients for the pixel-to-partition 2 compensation value. The distance is the Euclidean distance from the pixel to the center of partition 1. Let be the Euclidean distance from the pixel to the center of partition 2, and .

[0077] The formula for calculating the transition compensation value is:

[0078] in, This is the final compensation value for the pixels in the transition region. The compensation parameter value for partition 1. The compensation parameter value for partition 2. , These are the weighting coefficients mentioned above.

[0079] The output includes a complete compensation parameter matrix after boundary smoothing transition processing, ensuring the spatial continuity of the compensation parameters across the entire board.

[0080] Furthermore, the boundary transition region between adjacent partitions is determined by the adaptive clustering partitioning result of step S2. Specifically, the partitioning scheme output by step S2 includes a set of boundary coordinates for each partition, and the common boundary line between adjacent partitions is the initial boundary position. Based on this initial boundary position, the partitions are expanded inward on both sides. 1 pixel ( The transition bandwidth value calculated by the adaptive adjustment mechanism is used to determine the boundary transition region, which is the strip-shaped area formed by the transition bandwidth. The specific determination method is as follows: for any pixel on the panel... Calculate the Euclidean distance from the cluster center of its respective partition. and the Euclidean distance to the second nearest cluster center If the conditions are met Then determine the pixel point Located within the boundary transition region, it needs to be subject to weighted interpolation compensation; if If the pixel is located within the core region of the partition, the compensation parameters for that partition are applied directly without interpolation. The above determination process utilizes the Euclidean distance data from each data point to each cluster center calculated in step S2. No additional calculation is required, and the data can be directly reused, thereby accurately locating the boundary transition region without increasing computational overhead.

[0081] This step effectively eliminates abrupt changes in compensation parameters at the boundaries of adjacent partitions by introducing a distance-weighted linear interpolation method, resulting in a smooth spatial transition of compensation parameters, avoiding visible regional boundaries, and further improving display uniformity.

[0082] Furthermore, this step introduces an adaptive transition bandwidth adjustment mechanism, automatically determining the optimal transition bandwidth based on the characteristic differences between adjacent partitions, thus avoiding insufficient transition effect or excessively large transition areas caused by fixed bandwidth. Transition bandwidth The calculation formula is:

[0083] in, This represents the maximum difference between two adjacent partitions across all compensation parameter dimensions (using the normalized dimensionless value). This is a scaling factor (preferably ranging from 5 to 15 pixels per unit of difference). Transition bandwidth (per pixel).

[0084] When the compensation parameters between adjacent zones differ significantly (e.g., Zone A and Zone D are adjacent), the transition bandwidth is automatically increased to make the changes in compensation parameters smoother. When the compensation parameters between adjacent zones differ slightly (e.g., Zone A and Zone B are adjacent), the transition bandwidth is appropriately reduced to minimize the area occupied by the transition region in the effective compensation region. The preferred adjustment range for the transition bandwidth is 20 to 100 pixels to achieve the optimal balance between visual smoothness and compensation accuracy. This mechanism makes the transition processing of each zone boundary more targeted, further improving the visual consistency of the overall display uniformity.

[0085] S5: Assign a unique identifier to each partition, establish a mapping relationship between the unique identifier and the corresponding partition's compensation parameter in the full board compensation parameter matrix, and store the information.

[0086] In this embodiment, the purpose of this step is to establish a unique identifier system for each partition, providing an index basis for automated production line calls.

[0087] Assign a unique identifier (ID) to each partition determined in step S2, corresponding to the QR code encoding information on the panel. Establish a mapping relationship between the ID and the corresponding partition's photo parameter version number and compensation parameter version number, and store it in the database module of the partition decision server to form a mapping database of ID → (photo parameter version number, compensation parameter version number).

[0088] The Demura system pre-configures partitions and associates partition IDs with corresponding calibration procedures, ensuring that the production line SIM system can quickly retrieve and call up the parameters of the corresponding version based on the scanned panel QR code.

[0089] Output a complete ID mapping database, containing a complete mapping relationship between each partition ID, the corresponding shooting parameter version, and the compensation parameter version.

[0090] During production, the system automatically identifies the panel ID and intelligently calls up the corresponding version of the photography parameters and compensation parameters, achieving rapid response and precise adjustment. This process not only efficiently compensates for the visual display effects of panels with various characteristics, but also significantly improves production efficiency and product yield without increasing additional equipment investment or extending production hours, providing a reliable solution for the OLED panel manufacturing industry.

[0091] This step establishes a precise mapping relationship between panel ID and compensation parameter version, providing a reliable indexing mechanism for automated production line calls and solving the problems of low efficiency and error risk caused by manual switching of compensation parameters in existing technologies.

[0092] In this embodiment, the purpose of this step is to achieve fast and accurate Demura compensation through an automated identification and invocation mechanism during the formal mass production process, without the need for manual intervention.

[0093] The production line SIM system scans the QR code on the panel to be compensated using an industrial camera (such as the Basler Ace series) to identify the panel's unique identifier ID. Based on the identified ID, the SIM system retrieves the corresponding image parameter version and compensation parameter version from the mapping database, automatically switches the camera image parameters (including exposure time, gain, white balance, etc.) and compensation coefficient configuration, and executes the Demura compensation process.

[0094] After compensation is completed, the system automatically performs a second image acquisition and calculates the compensated color difference. Verify whether the compensation effect meets the requirements. The uniformity standard is less than 2; if the standard is not met, the compensation parameters will be automatically adjusted and the process will be repeated until the requirements are met.

[0095] The output panel has completed Demura compensation. The compensation parameters have been burned into the panel driver chip, and a compensation effect verification report is attached.

[0096] Furthermore, this step introduces a compensation effect prediction model. Before programming the compensation parameters into the driver chip, the prediction model pre-evaluates the compensation effect, reducing the number of actual programming-verification iterations and improving production efficiency. The compensation effect prediction model is built based on historical compensation data and uses a multinomial regression method to establish the compensation coefficient and the post-compensation relationship. The mapping relationship between values; the input to the prediction model is the compensation parameter set for the current partition; the output is the predicted compensated color difference. .like If the value is less than 1.5 (the preset prediction pass threshold), the burning operation will be performed directly, skipping the initial verification data collection and saving verification time; if... If the value is between 1.5 and 2.0, a single verification data collection will be performed after the flashing process; if... If the value is greater than 2.0, the compensation parameters are pre-adjusted before programming. The overall scaling factor of the brightness compensation coefficient is adjusted to 0.95 to 1.05 times the original value for optimization before programming. The prediction model is automatically updated after each compensation verification, incorporating the new compensation data into the training set to achieve continuous model optimization. This mechanism reduces the average number of programming-verification iterations from 2 to 3 times to 1 to 1.5 times, significantly improving the processing efficiency of the production line.

[0097] This step automatically identifies the panel ID and calls the corresponding version compensation parameters through the production line SIM system, achieving rapid response and high-precision adjustment, greatly reducing the need for manual intervention, and improving processing consistency and production efficiency without increasing additional equipment investment and production time.

[0098] In this embodiment, the purpose of this step is to automatically trigger parameter updates when process fluctuations exceed a threshold by real-time monitoring of panel characteristic data of production batches, thereby achieving self-learning optimization of the compensation scheme.

[0099] During the production process, film thickness distribution data and electrical parameter data of at least 10 panels to be compensated are collected in real time, and the standard deviation of the film thickness distribution data of the current batch is calculated. and the standard deviation of electrical parameter data The calculated standard deviation is compared with a preset threshold, which is set to 150% of the standard deviation during initial parameter tuning. or When the preset threshold is exceeded, the system automatically triggers the compensation parameter re-adjustment process, re-executes steps S1 to S5, generates updated compensation parameters, and updates the corresponding version in the mapping database.

[0100] Output the updated compensation parameter version and mapping database to ensure that the production line always uses the latest version of the compensation parameters.

[0101] This step introduces an adaptive dynamic update mechanism, enabling the compensation scheme to automatically adjust as process capabilities change. This achieves self-learning optimization and effectively addresses the problem of decreased compensation effectiveness caused by fluctuations in process parameters between different production batches.

[0102] Preferably, the threshold monitoring cycle is set to be once every 50 panels produced, in order to strike a balance between monitoring frequency and system overhead.

[0103] Furthermore, this step introduces a compensation parameter version rollback mechanism to evaluate the effectiveness of dynamically updated compensation parameter versions. When the compensation effect of a new version is inferior to the historical best version, a rollback is automatically triggered to ensure the stability of production quality. The version management strategy is as follows: The system retains the most recent N (preferably N=5) historical compensation parameter versions and their corresponding compensation effect evaluation indicators in the database module, including average... Value, brightness uniformity and color uniformity The statistical mean; after the new version of the compensation parameters is debugged, its compensation effect is verified by small-batch trial production on the production line (preferably 10 to 20 panels), and the average value of the trial production panels is calculated. ;like Greater than the average of the best historical version If the performance rate is 110%, the new version is considered to have degraded, the system will automatically roll back to the best historical version, and issue a warning to the operator indicating abnormal fluctuations in process parameters; if Not greater than If the accuracy reaches 110%, the new version is confirmed to be effective, and the historical best version record is updated. The version rollback mechanism effectively prevents erroneous updates of compensation parameters due to measurement noise or abnormal process fluctuations, ensuring the continuous stability of compensation quality during mass production.

[0104] Preferably, the number N of historical versions to be retained can be flexibly configured according to the production cycle and storage resources. It is recommended to retain at least 3 historical versions to ensure the effectiveness of the rollback mechanism.

[0105] In summary, the beneficial effects of this method are as follows: The system collects multi-dimensional characteristic data such as film thickness distribution, electrical parameters (Vth, SS), brightness uniformity, and color uniformity, and introduces current-voltage characteristic testing to obtain luminous efficiency distribution. Compared with the existing technology that relies on only a single parameter to determine partitioning, the data foundation is more comprehensive and the partitioning basis is more reliable.

[0106] The K-means clustering algorithm is used to adaptively cluster multi-dimensional characteristic data. The optimal number of partitions and partition boundaries are automatically determined by the silhouette coefficient. There is no need to manually preset a fixed partitioning scheme. It can automatically adapt to the changes in process capabilities of different batches, thus improving the partitioning accuracy and the adaptability of the scheme.

[0107] To address the differences in film thickness, Vth, and SS characteristics among different zones, differentiated brightness gain adjustment strategies, Vth compensation strategies, SS compensation strategies, and image capture configurations were developed. This enabled precise matching of the characteristics of each zone, solving the problem of overcompensation or undercompensation caused by uniform compensation parameters in existing technologies, and significantly improving product yield.

[0108] The transition compensation value is calculated by distance-weighted linear interpolation at the boundary of adjacent partitions, and an adaptive adjustment mechanism for transition bandwidth is introduced. The optimal transition bandwidth is automatically determined according to the difference in characteristics between adjacent partitions, which effectively eliminates the abrupt change in compensation parameters at the boundary of adjacent partitions, avoids the visible regional boundary line, and further improves the visual consistency of display uniformity.

[0109] The application value and significance of the method of this invention are as follows: The improvement in display quality effectively enhances the brightness uniformity of OLED displays, improves the consistency of color performance, and corrects display defects generated during the manufacturing process.

[0110] Optimize production efficiency, improve product yield, avoid resource waste caused by overcalibration, and improve production efficiency simultaneously.

[0111] It provides new process ideas for OLED manufacturing and offers technical solutions that the industry can learn from.

[0112] In summary, this innovative Demura process solution for OLED panels represents a significant technological breakthrough for the industry. It not only improves production efficiency and product quality but also opens up new growth opportunities for OLED panel manufacturers. As the OLED market develops and expands, this technology will bring long-term competitive advantages and economic benefits to companies. This solution not only increases yield but also significantly shortens the time from product development to market. By reducing rework and increasing first-pass yield, it also helps manufacturers save substantial costs.

[0113] Example 2 This embodiment discloses an OLED panel Demura compensation system, including: The data acquisition module is configured to: acquire multi-dimensional characteristic data of the OLED panel and integrate the multi-dimensional characteristic data into a multi-dimensional characteristic data matrix; The partitioning decision module is configured to: perform adaptive clustering on the multi-dimensional characteristic data matrix to obtain the optimal number of partitions, partition boundaries, and characteristic statistical parameters of each partition; The compensation parameter generation module is configured to: calculate the brightness compensation coefficient and shooting parameters of each zone based on the characteristic statistical parameters of each zone, and establish a compensation parameter set for each zone; The boundary smoothing module is configured to: based on the partition boundary, perform a distance-weighted linear interpolation method on the compensation parameters at adjacent partition boundaries in the compensation parameter set to smooth the transition, thereby obtaining the full-board compensation parameter matrix; The ID mapping module is configured to: assign a unique identifier to each partition, establish a mapping relationship between the unique identifier and the corresponding partition's compensation parameter in the full-board compensation parameter matrix, and store the mapping.

[0114] Example 3 The purpose of this embodiment is to provide a computing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method of Embodiment 1.

[0115] Example 4 The purpose of this embodiment is to provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the steps of the method of Embodiment 1.

[0116] The steps and methods involved in the apparatuses of Embodiments 3 and 4 above correspond to those in Embodiment 1. For specific implementation details, please refer to the relevant description section of Embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood as including any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and enabling the processor to perform any of the methods in this invention.

[0117] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computer device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. The present invention is not limited to any particular combination of hardware and software.

[0118] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0119] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for Demura compensation in an OLED panel, characterized in that, include: Acquire multi-dimensional characteristic data of the OLED panel and integrate the multi-dimensional characteristic data into a multi-dimensional characteristic data matrix; Adaptive clustering is performed on the multi-dimensional characteristic data matrix to obtain the optimal number of partitions, partition boundaries, and characteristic statistical parameters of each partition; Based on the characteristic statistical parameters of each zone, the brightness compensation coefficient and shooting parameters of each zone are calculated respectively, and a compensation parameter set for each zone is established. Based on the partition boundaries, the compensation parameters at adjacent partition boundaries in the compensation parameter set are processed by distance-weighted linear interpolation to achieve a smooth transition, thus obtaining the full-board compensation parameter matrix. Each partition is assigned a unique identifier, and a mapping relationship is established between the unique identifier and the corresponding partition's compensation parameter in the full-board compensation parameter matrix, and then stored.

2. The OLED panel Demura compensation method as described in claim 1, characterized in that, The multi-dimensional characteristic data includes film thickness distribution data, electrical parameter data, brightness uniformity data, and color uniformity data.

3. The OLED panel Demura compensation method as described in claim 1, characterized in that, The optimal number of partitions, partition boundaries, and characteristic statistical parameters of each partition are obtained, including: Normalize the data in each dimension of the multidimensional characteristic data matrix to obtain normalized multidimensional characteristic data; The K-means clustering algorithm is used to cluster the normalized multi-dimensional characteristic data, automatically determine the optimal number of partitions, and obtain the clustering results; Based on the clustering results, the boundary coordinates of each partition are determined, and the characteristic statistical parameters of film thickness, electrical parameters, brightness uniformity and color uniformity in each partition are statistically analyzed.

4. The OLED panel Demura compensation method as described in claim 3, characterized in that, A partition confidence assessment mechanism is introduced to quantitatively evaluate the reliability of partitions in the clustering results.

5. The OLED panel Demura compensation method as described in claim 1, characterized in that, Establish compensation parameter sets for each partition, specifically including: Based on the brightness uniformity data of each zone, the brightness compensation coefficient of each zone is calculated using a brightness compensation model. The panel of each zone is divided into grids, and the compensation parameters are calculated independently for each grid. Piecewise linear fitting is performed on the compensation parameters under different gray levels to obtain the brightness compensation coefficient matrix of each zone. Based on the differences in film thickness and electrical parameters of each zone, differentiated imaging parameters for each zone are determined, including exposure time, gain, and test screen configuration. The brightness compensation coefficient matrix of each zone is summarized with the shooting parameters to form the compensation parameter set of each zone.

6. The OLED panel Demura compensation method as described in claim 5, characterized in that, The brightness compensation model is expressed as follows: in, For position The output brightness value after compensation. For position The original input brightness value at that location, The target is a uniform brightness reference value. For position The measured brightness value at the location.

7. The OLED panel Demura compensation method as described in claim 1, characterized in that, The full-board compensation parameter matrix is ​​obtained, specifically including: The boundary transition region between adjacent partitions is determined based on the partition boundaries. For each pixel within the boundary transition area, a corresponding weight coefficient is calculated based on its distance to the centers of the two adjacent partitions. Based on the weighting coefficients, the compensation parameters of two adjacent partitions in the compensation parameter set are weighted linearly interpolated to obtain the transition compensation value of each pixel in the boundary transition region. The transition compensation values ​​of each pixel within the boundary transition area are combined with the compensation parameters of each partition core area to obtain the full-board compensation parameter matrix.

8. An OLED panel Demura compensation system, characterized in that, include: The data acquisition module is configured to: acquire multi-dimensional characteristic data of the OLED panel and integrate the multi-dimensional characteristic data into a multi-dimensional characteristic data matrix; The partitioning decision module is configured to: perform adaptive clustering on the multi-dimensional characteristic data matrix to obtain the optimal number of partitions, partition boundaries, and characteristic statistical parameters of each partition; The compensation parameter generation module is configured to: calculate the brightness compensation coefficient and shooting parameters of each zone based on the characteristic statistical parameters of each zone, and establish a compensation parameter set for each zone; The boundary smoothing module is configured to: based on the partition boundary, perform a distance-weighted linear interpolation method on the compensation parameters at adjacent partition boundaries in the compensation parameter set to smooth the transition, thereby obtaining the full-board compensation parameter matrix; The ID mapping module is configured to: assign a unique identifier to each partition, establish a mapping relationship between the unique identifier and the corresponding partition's compensation parameter in the full-board compensation parameter matrix, and store the mapping.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the OLED panel Demura compensation method as described in any one of claims 1-7.

10. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the OLED panel Demura compensation method as described in any one of claims 1-7.

Images