Display panel visual effect classification compensation method and device, electronic equipment and storage medium

Abstract

The application provides a display panel visual effect classification compensation method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring and saving visual effect characteristic attributes of N display panels, wherein N is a positive integer; classifying a plurality of display panels into X cluster centers according to the visual effect characteristic attributes and a clustering algorithm, wherein X is a positive integer; and compensating each type of display panel according to a compensation scheme corresponding to the cluster center. The application divides a plurality of display panels into multiple categories through a clustering algorithm, and respectively applies a corresponding visual effect compensation scheme to each category of display panel according to the classification result, so as to optimize the display effect while ensuring the production efficiency.

Description

Technical Field

[0001] This application relates to the field of display technology, and in particular to a display panel visual effect classification compensation method, device, electronic device, and storage medium. Background Technology

[0002] Active-matrix organic light-emitting diodes (AMOLEDs) have been widely used in various high-end display terminals due to their advantages such as self-illumination, high contrast, and low power consumption. As products become more high-end, users' requirements for screen color reproduction accuracy and display consistency continue to increase, making color correction and color gamut mapping technologies the core keys to improving AMOLED display quality.

[0003] Currently, the display industry generally adopts JNCD (Just Noticeable Color Difference) and DE2000 (CIEDE2000, the International Commission on Illumination's 2000 color difference evaluation standard) as color accuracy control standards, and the industry's accuracy threshold continues to tighten. To adapt to high standard requirements, the existing color compensation algorithm architecture is becoming increasingly complex, significantly increasing the workload of debugging and production cycle, and reducing mass production efficiency.

[0004] To balance accuracy and production capacity, the industry generally adopts a generic compensation scheme to uniformly calibrate modules within the same batch. However, due to factors such as manufacturing processes and equipment deviations, individual AMOLED modules within the same batch exhibit variations in display characteristics such as color shift, brightness, and color gamut. The generic scheme cannot achieve precise calibration for each individual module, and even after calibration, modules still suffer from color distortion and poor batch consistency, making it difficult to meet high-precision testing standards. Therefore, how to balance mass production efficiency with individual module calibration accuracy is a pressing technical challenge in this field. Summary of the Invention

[0005] In view of the above problems, the purpose of this application is to provide a display panel visual effect classification compensation method, device, electronic device, and storage medium, which divides multiple display panels into multiple categories through a clustering algorithm, and applies corresponding visual effect compensation schemes to each category of display panels according to the classification results, thereby optimizing the display effect while ensuring production efficiency.

[0006] According to one aspect of this application, a visual effect classification compensation method for display panels is provided, comprising: acquiring and storing visual effect feature attributes of N display panels, where N is a positive integer; classifying the multiple display panels into X cluster centers according to the visual effect feature attributes and a clustering algorithm, where X is a positive integer; and compensating each type of display panel according to a compensation scheme corresponding to the cluster centers.

[0007] Optionally, the method for obtaining the cluster centers includes: acquiring visual effect feature attributes of multiple display panels; clustering the visual effect feature attributes using the clustering algorithm to obtain X cluster centers, where X is a positive integer; verifying the X cluster centers using the visual effect feature attributes of the multiple display panels; outputting X cluster centers if the verification passes; and re-obtaining X cluster centers based on the visual effect feature attributes and the clustering algorithm if the verification fails, until the X cluster centers pass the verification.

[0008] Optionally, it further includes: when the number N of the display panels whose visual effect feature attributes are acquired and saved is greater than the Y display panels, updating the X cluster centers and corresponding compensation schemes according to the visual effect features of the newly acquired display panels.

[0009] Optionally, when the number N of the display panels for which visual effect feature attributes are acquired and saved is greater than the number of display panels Y, the step of updating the X cluster centers and corresponding compensation schemes based on the visual effect features of the newly acquired display panels includes: determining whether the number N of the display panels for which visual effect feature attributes are saved is greater than (2... m-1 Y, m are positive integers initially set to 1; when N is greater than (2 m-1 When Y is reached, the X cluster centers are updated and verified according to the newly saved visual effect feature attributes. Let m = m + 1, and update the compensation scheme of the X cluster centers accordingly. Then, return to determine whether the number N of display panels that save the visual effect feature attributes is greater than (2). m-1 The steps of Y.

[0010] Optionally, updating the X cluster centers and their corresponding compensation schemes requires a transition period, during which the visual effect feature attributes of the display panel obtained use the X cluster centers and their corresponding compensation schemes before the update.

[0011] Optionally, the visual effect characteristics include at least one of the following: CIE1931XYZ color swatch measurements at different brightness, temperatures, and frequencies, IR-drop, mura morphology, and current-voltage characteristics.

[0012] Optionally, the clustering algorithm includes a deep learning algorithm or a machine learning algorithm.

[0013] Optionally, the compensation scheme includes at least one of color gamut mapping, color correction, color temperature compensation, mura compensation, and color gamut conversion.

[0014] According to another aspect of the present invention, a visual effect classification compensation device for a display panel is provided, comprising: a visual effect feature acquisition module for acquiring visual effect feature attributes of N display panels, where N is a positive integer; a classification module connected to the visual effect feature acquisition module for classifying the display panels into X cluster centers according to the acquired visual effect feature attributes, where X is a positive integer; and a compensation module connected to the classification module for compensating each type of display panel according to a compensation scheme corresponding to the cluster centers.

[0015] Optionally, it further includes: a clustering module, connected to the classification module, used to cluster multiple visual effect feature attributes according to a clustering algorithm to obtain X cluster centers; and a storage module, connected to the visual effect feature acquisition module, the compensation module and the clustering module, used to store the acquired visual effect feature attributes and compensation schemes of N display panels.

[0016] Optionally, the clustering module further includes: verifying the obtained X cluster centers; if the verification passes, outputting X cluster centers; if the verification fails, obtaining X cluster centers again based on the visual effect feature attributes and the clustering algorithm, until the X cluster centers pass the verification.

[0017] Optionally, the clustering module further includes: when the number N of the display panels whose visual effect feature attributes are stored in the storage module is greater than the Y display panels, updating X cluster centers and corresponding compensation schemes according to the visual effect features of the newly acquired display panels.

[0018] According to another aspect of the present invention, an electronic device is provided, wherein the above-described visual effect classification compensation device for a display panel is included.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, which stores a computer program, wherein the computer program, when executed by a processor, implements the above-described visual effects classification compensation method.

[0020] The visual effect classification compensation method, apparatus, electronic device, and computer-readable storage medium for display panels provided in this application acquire and save the visual effect feature attributes of multiple display panels, use a clustering algorithm to classify the display panels into multiple cluster centers, and compensate for each type of display panel according to the compensation scheme corresponding to each cluster center. This achieves accurate visual effect compensation based on the individual differences of display panels, avoids the over-correction or under-correction problems caused by uniform parameters in traditional public version compensation schemes, significantly improves the visual effect consistency and compensation accuracy of display panels, and maintains high production efficiency in large-scale mass production.

[0021] Furthermore, in the process of obtaining cluster centers, the cluster centers are verified by using the visual effect feature attributes of multiple display panels. If the verification fails, the clustering is re-clustered until it passes, which ensures the accuracy and reliability of the cluster centers, thereby avoiding invalid classification, improving the effectiveness of subsequent compensation, and making the classification results more realistically reflect the distribution pattern of the visual effect characteristics of the display panels.

[0022] When the number of display panels that acquire and save visual effect feature attributes exceeds a preset threshold, the cluster center and corresponding compensation scheme are updated according to the newly acquired visual effect features. This enables dynamic adjustment of the cluster center and compensation scheme, adapts to changes in display panel characteristics during production, maintains the timeliness and accuracy of the compensation scheme, and continuously optimizes the compensation effect in long-term operation, avoiding the decline in compensation performance caused by data distribution drift.

[0023] The update judgment strategy is based on exponential growth, i.e., when the number of display panels is greater than (2 m-1 When Y is triggered, the value of m is updated and incremented, which balances the relationship between computational resource consumption and model accuracy. It avoids parameter oscillation caused by premature updates or response lag caused by late updates, ensuring the stability of the system and the convergence of the compensation effect, and enabling the cluster centers to be gradually optimized as data accumulates.

[0024] During the transition period required to update the cluster centers and compensation scheme, the new display panels were compensated using the old cluster centers and compensation scheme. This ensured the continuity and stability of the compensation process, avoided fluctuations in the visual effects of the display panels due to interruptions in the update process, and ensured the smooth operation of the production line and the consistency of the compensation effect. Attached Figure Description

[0025] The above and other objects, features and advantages of this application will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0026] Figure 1 A flowchart of a visual effect classification compensation method for a display panel according to an embodiment of this application is shown;

[0027] Figure 2 A flowchart illustrating the process of obtaining cluster centers using a clustering algorithm according to an embodiment of this application is shown;

[0028] Figure 3 A flowchart of the updated clustering algorithm according to an embodiment of this application is shown;

[0029] Figure 4 A schematic diagram of the structure of a visual effect classification compensation device for a display panel according to an embodiment of this application is shown;

[0030] Figure 5A schematic diagram of the clustering module in the visual effect classification compensation device for a display panel according to an embodiment of this application is shown. Detailed Implementation

[0031] The present application will now be described in more detail with reference to the accompanying drawings. In the various drawings, the same elements are indicated by similar reference numerals. For clarity, the various parts in the drawings are not drawn to scale. Furthermore, some well-known parts may not be shown. For simplicity, the semiconductor structure obtained after several steps can be depicted in a single figure.

[0032] It should be understood that when describing the structure of a device, when referring to a layer or region as being "above" or "on top of" another layer or region, it can mean that it is directly above another layer or region, or that there are other layers or regions between it and another layer or region. Furthermore, if the device is flipped, that layer or region will be located "below" or "under" another layer or region.

[0033] To describe a situation where it is located directly on another layer or another area, this article will use expressions such as "directly on top of" or "on and adjacent to".

[0034] Many specific details of this application, such as the structure, materials, dimensions, processing techniques, and methods of the devices, are described below to provide a clearer understanding of the application. However, as those skilled in the art will understand, this application may be implemented without adhering to these specific details.

[0035] The specific implementation methods of this application will be further described in detail below with reference to the accompanying drawings and embodiments.

[0036] Figure 1 A flowchart of a visual effect classification compensation method for a display panel according to an embodiment of this application is shown; Figure 2 A flowchart illustrating the process of obtaining cluster centers using a clustering algorithm according to an embodiment of this application is shown; Figure 3 A flowchart of the updated clustering algorithm according to an embodiment of this application is shown.

[0037] In this application, the display panel is, for example, an AMOLED (Active-matrix Organic Light-Emitting Diode) panel.

[0038] like Figures 1 to 3 As shown, the visual effect classification compensation method for the display panel of this application includes the following specific contents.

[0039] Step S11: Obtain and save the visual effect attribute of N display panels.

[0040] In this step, visual characteristic attributes of N display panels from the same batch are collected and saved. On the production line, the display panels are produced in a continuous flow, therefore the number N of display panels accumulates continuously with the production line's output.

[0041] In some embodiments, visual effect characteristics include at least one of the following: CIE1931 XYZ color swatch measurements at different brightness, temperatures, and frequencies, IR-drop, mura morphology, and current-voltage characteristics.

[0042] The CIE1931 XYZ color swatch refers to the standard color space established by the International Commission on Illumination in 1931. It is the basis for colorimetric measurement. XYZ represents three stimulus values, which represent the amount of light stimulation received by the three types of photoreceptor cells in the human eye.

[0043] IR-drop is a voltage drop, meaning that when power is supplied from one end of the screen to a distant pixel, a portion of the voltage is "eaten up" along the way, resulting in a lower voltage at the distant end. The brightness of pixels in the area with lower voltage decreases, which is one of the most important electrical causes of uneven screen brightness and horizontal / vertical stripes (mura).

[0044] Mura morphology refers to the appearance shape and distribution characteristics of uneven brightness / color in a display panel.

[0045] Current-voltage characteristics refer to the relationship curve between the current of a device and the applied voltage.

[0046] In some embodiments, the process of collecting visual effect feature attributes needs to cover the entire working area of ​​the display panel to ensure data comprehensiveness. After acquiring the data, these multi-dimensional visual effect feature attributes are standardized and saved.

[0047] In some embodiments, the number N of display panels varies from batch to batch.

[0048] By collecting the visual effect features of the display panels, a high-quality sample set is constructed to provide sufficient training materials for subsequent machine learning classification. This ensures that the cluster centers can accurately reflect the overall distribution pattern of the batch of display panels, thereby breaking the limitation of the traditional public version compensation scheme of "one-size-fits-all" and laying the foundation for refined grouping.

[0049] Step S12: Classify the multiple display panels into X cluster centers according to the visual effect feature attributes and clustering algorithm.

[0050] In this step, the display panels are classified into different cluster centers based on the visual effect characteristics of the obtained display panels. In this embodiment, there are, for example, X cluster centers.

[0051] Specifically, by analyzing visual effect feature attributes, the potential structure and distribution pattern within the data are identified. Then, through a specific clustering algorithm, the visual effect feature attributes of multiple display panels are matched one by one with X cluster centers by distance calculation or similarity matching. Finally, each display panel is assigned to the category of the cluster center whose features are closest to its own.

[0052] In clustering, cluster centers are the feature vectors or geometric centers used to represent a particular class of data when grouping data (clustering). Cluster centers serve as a benchmark for measuring sample similarity. New data points are assigned to the nearest cluster based on their distance from the cluster centers.

[0053] In this application, cluster centers refer to, for example, the feature vectors or geometric center points of the visual effect feature attributes of multiple display panels, and each cluster center represents the average feature state of a group of display panels with similar display characteristics.

[0054] In some embodiments, the number of cluster centers X is not fixed, but is continuously adjusted and optimized as the input data is updated in order to better fit the distribution characteristics of the current dataset.

[0055] In some embodiments, the number of cluster centers X varies depending on the set similarity threshold. For example, when the set similarity threshold is large, the number of cluster centers X will be small, while when the set similarity threshold is small, the number of cluster centers X will be large. Those skilled in the art can adjust the number of cluster centers X by setting the similarity threshold according to the visual effect feature attribute range of the display panel, the settings of the visual effect classification compensation device 100, the compensation effect, and the compensation efficiency.

[0056] The process of classifying display panels based on their visual characteristics transforms unordered data into ordered groups, dividing previously diverse display panels into several clusters with highly consistent characteristics. This data-driven classification method can automatically adapt to changes in display characteristics across different batches and production lines, providing a scientific basis for applying independent compensation schemes to each group and significantly improving the accuracy of calibration.

[0057] In some embodiments, the clustering algorithm includes a deep learning algorithm or a machine learning algorithm.

[0058] Among them, such as Figure 2 As shown, the method for obtaining X cluster centers includes the following steps.

[0059] Step S21: Obtain the visual effect characteristic attributes of multiple display panels.

[0060] In this step, when a batch of display panels needs to be compensated, the visual effect feature attributes of multiple display panels are first obtained, and then multiple cluster centers are obtained based on the visual effect feature attributes.

[0061] In some embodiments, if the initial compensation of display panels in the same batch is to be performed, it is necessary to obtain the visual effect feature attributes of Y display panels, where Y is an integer and a preset threshold. The value of Y cannot be too small, otherwise it will easily lead to inaccurate cluster centers obtained subsequently; the value of Y cannot be too large either, otherwise the time to initially obtain cluster centers will be longer.

[0062] In some embodiments, since the display panels are produced in a continuous flow on the production line, it is not possible to rely on the visual effect feature attributes of the initially set Y display panels to obtain X cluster centers. Instead, it is necessary to obtain the visual effect feature attributes of each new display panel produced subsequently in real time and incrementally.

[0063] In some embodiments, as the number of display panels N increases, the cluster centers are updated. Specifically, the update of the cluster centers will be discussed later. Figure 3 The description.

[0064] Step S22: Cluster the visual effect feature attributes using a clustering algorithm to obtain X cluster centers, where X is a positive integer.

[0065] In this step, multiple visual effect feature attributes are analyzed and clustered using a clustering algorithm to obtain X cluster centers, where X is a positive integer.

[0066] In some embodiments, the clustering algorithm automatically finds high-density regions in the data space of failure feature attributes and calculates X optimal "cluster centers".

[0067] In some embodiments, if the compensation of display panels in the same batch is initial, clustering is performed based on the visual effect feature attributes of Y display panels to obtain X cluster centers.

[0068] In some embodiments, if the number N of display panels that have obtained visual effect feature attributes is greater than the initial number Y in the later stage of compensation for display panels in the same batch, the cluster center is updated based on the visual effect feature attributes that are not used by the current cluster center.

[0069] In some embodiments, the number of cluster centers X remains unchanged before and after the cluster centers are updated, but the average feature state of the display panel represented by the cluster centers may change.

[0070] Step S23: Verify the X cluster centers using the visual effect feature attributes of multiple display panels.

[0071] In this step, to ensure that the newly generated cluster centers have sufficient stability and reliability, X cluster centers need to be validated.

[0072] Specifically, the verified visual effect features are input into a cluster center-based classification model to observe the distribution of classification results, convergence speed, and whether they conform to the expected physical characteristics. For example, it checks whether a large number of visual effect features are incorrectly classified into marginal clusters, or whether the cluster centers are too scattered to represent the features of a certain type of module.

[0073] In some embodiments, if the compensation of display panels in the same batch is initial, the visual effect feature attributes of Y display panels can be used to verify X cluster centers, or the visual effect feature attributes specifically reserved in the visual effect feature attributes of Y display panels can be used to verify X cluster centers.

[0074] In some embodiments, if the display panels are in the later stages of compensation within the same batch, X cluster centers are validated using at least a portion of the visual feature attributes that are not currently used by the cluster centers.

[0075] In some embodiments, if the cluster centers have been updated, historical data can be compared to determine whether the offset between the new cluster centers and the old cluster centers is within a reasonable range.

[0076] Step S24: Determine whether the verification passed.

[0077] In this step, if the verification results show good classification performance, high stability, and conformity to engineering intuition, the verification passes, X cluster centers are determined to be valid, and step S25 is executed; if anomalies are found (such as overfitting or underfitting), the verification fails, triggering a re-iteration mechanism, returning to step S22 to adjust the parameters of the clustering algorithm to obtain X cluster centers again, until the verification passes.

[0078] Step S25: Output X cluster centers.

[0079] In this step, if the verification of X cluster centers passes, then X cluster centers are output so that step S12 can directly use the valid X cluster centers.

[0080] Furthermore, the condition for triggering the cluster center update is as follows: when the number N of display panels that have acquired and saved visual effect feature attributes is greater than Y display panels, X cluster centers and corresponding compensation schemes are updated according to the visual effect features of the newly acquired display panels.

[0081] Specifically, refer to Figure 3 The methods for updating clustering include the following.

[0082] Step S31: Determine whether the number N of display panels storing visual effect feature attributes is greater than (2 m-1 )Y.

[0083] This step controls the logical switch for updating cluster centers. In this step, the number N of display panels acquiring visual feature attributes is monitored and compared with (2... m-1 The cluster centers are compared with Y, and the result determines whether to start updating the cluster centers. Here, m is a positive integer initially set to 1.

[0084] In some embodiments, m is a positive integer counter used to identify the current iteration round, and Y is the initial threshold for the number of display panels.

[0085] When N is greater than (2) m-1 When N is less than (2), it indicates that the accumulated visual feature attributes are sufficient to support the next round of cluster center update and verification. The system judgment condition is met, and step S32 is executed; when N is not greater than (2), it indicates that the accumulated visual feature attributes are sufficient to support the next round of cluster center update and verification. m-1 When Y is reached, it indicates that the number of visual effect features accumulated is insufficient. At this time, we will continue to stay in the stage of acquiring visual effect feature attributes or classifying the display panel according to the current cluster center, waiting for more samples to flow in, and only start the iteration when the number requirement is met.

[0086] Step S32: Update the X cluster centers according to the newly saved visual effect feature attributes and verify them.

[0087] In this step, when the number of display panels N satisfies N greater than (2 m-1 When Y is reached, the step of updating X cluster centers is performed. Specifically, the X cluster centers are updated using the method described in steps S21 to S25.

[0088] In some embodiments, since m may change, the newly saved visual effect feature attributes refer, for example, to the criteria used in this judgment (2). m-1 Y and the previous judgment criteria (2) m-1 The newly added visual effect feature attributes between Y and Y.

[0089] Step S33: Let m = m + 1, and update the compensation scheme for the X cluster centers accordingly.

[0090] In this step, after the X cluster centers are updated, let m = m + 1 and return to step S31. After confirming that the updated X cluster centers have passed the verification, the corresponding compensation scheme is updated according to the new X cluster centers.

[0091] That is, the value of m increased by 1, which changed the update judgment criteria in step S31. This also means that triggering the update of X cluster centers again requires more visual feature attributes.

[0092] In some embodiments, such as the second and subsequent updates of the X cluster centers and corresponding compensation schemes, a transition time td is required during the update process. During this transition time td, the visual effect characteristics of the newly acquired display panel are compensated using the X cluster centers and corresponding compensation schemes from before the update. Once the X cluster centers and corresponding compensation schemes are completed, the visual effect characteristics of the newly acquired display panel are compensated using the updated X cluster centers and corresponding compensation schemes.

[0093] For example, based on (2) m-1 When Y=2Y, the cluster center update is triggered. Let m=m+1, then the next update will require (2... m-1 Y = 4Y visual effect attribute data for display panels, but currently only 2Y data are available, requiring an additional 2Y data. However, updating the cluster centers and compensation scheme requires a transition time td. During this transition time td, y display panels acquire new visual effect attribute data. Therefore, of the 2Y display panels that need to acquire visual effect attribute data before triggering the next update, the first y panels still use the original cluster centers and compensation scheme, while the latter (2Y-y) display panels use the new cluster centers and compensation scheme.

[0094] Figure 3 The exponentially growing cluster center update strategy shown not only ensures rapid convergence in the early stages but also guarantees model stability through a large amount of data in the later stages. It effectively balances the relationship between computational resource consumption and model accuracy, avoiding parameter oscillations caused by premature updates or response lag caused by late updates.

[0095] Meanwhile, the update process of cluster centers also demonstrates the effects of online learning and dynamic adjustment, enabling the clustering algorithm to continuously adapt to new process changes as the production line operates over a long period of time, ensuring that it can provide the best visual compensation effect throughout the entire life cycle, and truly achieving the technical goals of "real-time grouping" and "effect convergence".

[0096] Step S13: Compensate for each type of display panel according to the compensation scheme corresponding to the cluster center.

[0097] In this step, the compensation scheme corresponding to the cluster center is executed on the display panels that have been classified into the corresponding cluster center on the production line, and the display panels are compensated accordingly. That is, the compensation scheme used for multiple display panels classified into the same cluster center is consistent.

[0098] In some embodiments, the compensation scheme includes at least one of color gamut mapping, color correction, color temperature compensation, mura compensation, and color gamut conversion.

[0099] In some embodiments, compensation for the display panel is performed in real time. When a display panel is classified under a cluster center, the compensation scheme corresponding to that cluster center is immediately applied to the display panel.

[0100] In some embodiments, compensation for display panels may be achieved by applying the compensation scheme corresponding to the cluster center to all display panels when the number of display panels belonging to a cluster center reaches a threshold.

[0101] In some embodiments, the compensation for the display panel can also be quantitative. For example, after the visual effect classification compensation method for the display panel is enabled, the classified display panels are compensated according to the compensation scheme of the corresponding cluster center for each predetermined number of visual effect feature attributes saved.

[0102] In some embodiments, the compensation for the display panel can also be timed. For example, after the visual effect classification compensation method for the display panel is turned on, the classified display panels are compensated according to the compensation scheme of the corresponding cluster center at predetermined intervals.

[0103] The visual effect classification and compensation method for display panels disclosed in this application classifies display panels into X cluster centers based on their visual effect characteristics. Each cluster center has a matching compensation scheme, thus enabling targeted compensation for display panels with similar defects after classification, unlike traditional reference designs that use averaged parameters, resulting in some display panels being over- or under-corrected. Through this process, each display panel can achieve a display effect that best matches its own characteristics, thereby maximizing production schedule under large-scale mass production conditions.

[0104] Figure 4 A schematic diagram of the structure of a visual effect classification compensation device for a display panel according to an embodiment of this application is shown; Figure 5 A schematic diagram of the clustering module in the visual effect classification compensation device for a display panel according to an embodiment of this application is shown.

[0105] like Figure 4 As shown, the visual effect classification compensation device 100 for the display panel of this application includes: a visual effect feature acquisition module 110, a classification module 120, and a compensation module 130.

[0106] Specifically, the visual effect feature acquisition module 110 is used to acquire the visual effect feature attributes of N display panels, where N is a positive integer. Before compensating multiple display panels in the same batch, the visual effect feature acquisition module 110 acquires the visual effect feature attributes of the display panels.

[0107] In some embodiments, visual effect characteristics include at least one of the following: CIE1931 XYZ color swatch measurements at different brightness levels, temperatures, and frequencies; IR-drop; mura morphology; and current-voltage characteristics.

[0108] The classification module 120 is connected to the visual effect feature acquisition module 110 and is used to classify N display panels into X cluster centers based on the acquired visual effect feature attributes, where X is a positive integer.

[0109] The compensation module 130 is connected to the classification module 120 and is used to compensate various display panels according to the compensation scheme corresponding to the cluster center. Specifically, different cluster centers correspond to different compensation schemes.

[0110] In some embodiments, the compensation scheme includes at least one of color gamut mapping, color correction, color temperature compensation, mura compensation, and color gamut conversion.

[0111] In some embodiments, the compensation module 130 includes a color gamut mapping unit, a color correction unit, and a color temperature compensation unit, etc., and different units are selected to compensate the display panel according to different compensation schemes.

[0112] In some embodiments, the compensation module 130 performs the compensation for the display panel in real time. When the classification module 120 classifies a display panel into a cluster center, the compensation module 130 compensates the display panel according to the compensation scheme corresponding to the cluster center.

[0113] In some embodiments, the compensation module 130 may compensate the display panels by applying the compensation scheme corresponding to the cluster center to a batch of display panels when the number of display panels belonging to a cluster center reaches a threshold.

[0114] In some embodiments, the compensation module 130 may also provide quantitative compensation for the display panel. For example, after the visual effect classification compensation method for the display panel is enabled, the compensation module 130 will perform compensation for the classified display panels according to the compensation scheme of the corresponding cluster center for each predetermined number of visual effect feature attributes saved by the storage module 140.

[0115] In some embodiments, the compensation module 130 may compensate the display panel at regular intervals. For example, after the visual effect classification compensation method of the display panel is enabled, the compensation module 130 will compensate the classified display panels according to the compensation scheme of the corresponding cluster center at predetermined intervals.

[0116] Further, refer to Figure 4 The visual effect classification compensation device 100 for the display panel of this application further includes: a storage module 140 and a clustering module 150.

[0117] The storage module 140 is connected to the visual effect feature acquisition module 110 and the compensation module 130, and is used to store the visual effect feature attributes and compensation schemes of the N display panels acquired by the visual effect feature acquisition module 110.

[0118] Specifically, in the same batch of compensation, all visual effect feature attributes acquired by the visual effect feature acquisition module 110 are stored in the storage module 140.

[0119] The clustering module 150 is connected to the storage module 140 and the classification module 120. It is used to obtain the visual effect feature attributes of Y display panels from the storage module 140, and to cluster the visual effect feature attributes through a clustering algorithm to obtain X cluster centers, where Y and X are positive integers.

[0120] In some embodiments, the Y display panels are a pre-cardinality, which relates to the update trigger threshold of subsequent cluster centers and compensation schemes, and Y is no greater than N.

[0121] In some embodiments, the clustering module 150 further includes: verifying the obtained X cluster centers; if the verification passes, outputting X cluster centers; if the verification fails, obtaining X cluster centers again based on visual effect feature attributes and clustering algorithm, until X cluster centers pass the verification.

[0122] In some embodiments, the clustering module 150 further includes: when the number N of display panels for which the storage module 140 stores visual effect feature attributes is greater than Y display panels, updating X cluster centers and corresponding compensation schemes according to the visual effect features of the newly acquired display panels, where Y is not greater than N.

[0123] In some embodiments, the clustering algorithm includes a deep learning algorithm or a machine learning algorithm.

[0124] Specifically, refer to Figure 5 The clustering module 150 includes an algorithm unit 151, a verification unit 152, and an update unit 153. These three units cooperate to complete at least the following tasks: Figure 2 and Figure 3 The function.

[0125] At the beginning of the classification compensation, the algorithm unit 151 obtains the visual effect feature attributes of multiple display panels from the storage module 140, and then clusters the visual effect feature attributes according to the clustering algorithm to obtain X cluster centers, where X is a positive integer.

[0126] Then, the verification unit 152, connected to the algorithm unit 151, verifies the X cluster centers obtained by the algorithm unit 151 to determine whether the X cluster centers are reliable. Specifically, the verification unit 152 uses the visual effect feature attributes of multiple display panels obtained by the algorithm unit 151 to verify the cluster centers. If the verification passes, X cluster centers are output to the classification module 120; if the verification fails, the algorithm unit 151 re-obtains X cluster centers based on the visual effect feature attributes of multiple display panels and the clustering algorithm, until X cluster centers pass the verification.

[0127] Furthermore, the update unit 153 is connected to the storage module 140 and is used to monitor the number N of display panels corresponding to the visual effect feature attributes stored in the storage module 140, and to determine whether the number N of display panels is greater than (2). m-1 Y, m are positive integers initialized to 1.

[0128] Since m is a positive integer initially set to 1, therefore (2 m-1 The initial value of Y is Y. That is, when the update unit 153 determines that the number N of display panels corresponding to the visual effect feature attributes stored in the storage module 140 is greater than Y, the update is triggered. At this time, the algorithm unit 151 updates X cluster centers according to the visual effect feature attributes of Y display panels, and then the verification unit 152 verifies the X cluster centers. This step is also the initial step in obtaining X cluster centers in this batch of compensation.

[0129] Furthermore, when the update unit 153 determines that the number N of display panels is greater than (2... m-1 After Y, there will be another m = m + 1, therefore, the initial (2 m-1 Y=Y becomes (2) m-1 (Y=2Y); Update unit 153 waits for the next update to be triggered as the number of display panels N increases. Similarly, when the number of display panels N is greater than 2Y, the (2Y) in update unit 153... m-1 Y=2Y becomes (2 m-1 Y=4Y, waiting for the update to be triggered again.

[0130] In some embodiments, the number N of display panels is cumulative.

[0131] In some embodiments, after triggering the update of X cluster centers, the newly obtained visual effect feature attributes are used to update the X cluster centers and the corresponding compensation scheme. For example, if the update is triggered when the number of display panels N is greater than 2Y, the visual effect feature attributes corresponding to the last Y display panels in 2Y are used to update the cluster centers; if the update is triggered when the number of display panels N is greater than 4Y, the visual effect feature attributes corresponding to the last 2Y display panels in 4Y are used to update the cluster centers.

[0132] Furthermore, this application also provides an electronic device, which includes a visual effect classification compensation device for a display panel as described above.

[0133] Furthermore, this application also provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the visual effects classification compensation method as described above.

[0134] The visual effect classification compensation method, apparatus, electronic device, and computer-readable storage medium for display panels provided in this application acquire and save the visual effect feature attributes of multiple display panels, use a clustering algorithm to classify the display panels into multiple cluster centers, and compensate for each type of display panel according to the compensation scheme corresponding to each cluster center. This achieves accurate visual effect compensation based on the individual differences of display panels, avoids the over-correction or under-correction problems caused by uniform parameters in traditional public version compensation schemes, significantly improves the visual effect consistency and compensation accuracy of display panels, and maintains high production efficiency in large-scale mass production.

[0135] Furthermore, in the process of obtaining cluster centers, the cluster centers are verified by using the visual effect feature attributes of multiple display panels. If the verification fails, the clustering is re-clustered until it passes, which ensures the accuracy and reliability of the cluster centers, thereby avoiding invalid classification, improving the effectiveness of subsequent compensation, and making the classification results more realistically reflect the distribution pattern of the visual effect characteristics of the display panels.

[0136] When the number of display panels that acquire and save visual effect feature attributes exceeds a preset threshold, the cluster center and corresponding compensation scheme are updated according to the newly acquired visual effect features. This enables dynamic adjustment of the cluster center and compensation scheme, adapts to changes in display panel characteristics during production, maintains the timeliness and accuracy of the compensation scheme, and continuously optimizes the compensation effect in long-term operation, avoiding the decline in compensation performance caused by data distribution drift.

[0137] The update judgment strategy is based on exponential growth, i.e., when the number of display panels is greater than (2 m-1 When Y is triggered, the value of m is updated and incremented, which balances the relationship between computational resource consumption and model accuracy. It avoids parameter oscillation caused by premature updates or response lag caused by late updates, ensuring the stability of the system and the convergence of the compensation effect, and enabling the cluster centers to be gradually optimized as data accumulates.

[0138] During the transition period required to update the cluster centers and compensation scheme, the new display panels were compensated using the old cluster centers and compensation scheme. This ensured the continuity and stability of the compensation process, avoided fluctuations in the visual effects of the display panels due to interruptions in the update process, and ensured the smooth operation of the production line and the consistency of the compensation effect.

[0139] As described above, these embodiments of the present application do not exhaustively cover all details, nor do they limit the invention to the specific embodiments described. Clearly, many modifications and variations can be made based on the above description. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the present application, thereby enabling those skilled in the art to make good use of the present application and modifications based on it. The present application is limited only by the claims and their full scope and equivalents.

Claims

1. A method for visual effect classification compensation of a display panel, wherein, include: Get and save the visual effect feature attributes of N display panels, where N is a positive integer; Based on the visual effect feature attributes and clustering algorithm, the multiple display panels are classified into X cluster centers, where X is a positive integer; Compensation is applied to various display panels based on the compensation scheme corresponding to the cluster centers.

2. The visual effects classification compensation method according to claim 1, wherein, The method for obtaining the cluster centers includes: Obtain visual effect feature attributes from multiple display panels; The visual effect feature attributes are clustered using the clustering algorithm to obtain X cluster centers, where X is a positive integer. The X cluster centers are verified using the visual effect characteristics of multiple display panels; If the validation passes, output X cluster centers; If the verification fails, X cluster centers will be obtained again based on the visual effect feature attributes and clustering algorithm until the X cluster centers pass the verification.

3. The visual effects classification compensation method according to claim 2, wherein, Also includes: When the number N of the display panels whose visual effect feature attributes are acquired and saved is greater than the number Y display panels, the X cluster centers and corresponding compensation schemes are updated according to the visual effect features of the newly acquired display panels.

4. The visual effects classification compensation method according to claim 3, wherein, When the number N of the display panels whose visual effect feature attributes are acquired and saved is greater than the number Y display panels, the steps of updating the X cluster centers and corresponding compensation schemes based on the visual effect features of the newly acquired display panels include: Determine if the number N of display panels storing visual effect feature attributes is greater than (2). m-1 Y and m are positive integers initially set to 1; When N is greater than (2) m-1 When Y is reached, the X cluster centers are updated and verified according to the newly saved visual effect feature attributes. Let m = m + 1, and update the compensation scheme of the X cluster centers accordingly. Then, return to determine whether the number N of display panels that save the visual effect feature attributes is greater than (2). m-1 The steps of Y.

5. The visual effects classification compensation method according to claim 4, wherein, Updating the X cluster centers and their corresponding compensation schemes requires a transition period. During this transition period, the visual effect feature attributes of the display panel obtained use the X cluster centers and their corresponding compensation schemes before the update.

6. The visual effects classification compensation method according to any one of claims 1-5, wherein, The visual effect characteristics include at least one of the following: CIE1931 XYZ color swatch measurements at different brightness, temperatures, and frequencies, IR-drop, mura morphology, and current-voltage characteristics.

7. The visual effects classification compensation method according to claim 6, wherein, The clustering algorithm includes deep learning algorithms or machine learning algorithms.

8. The visual effects classification compensation method according to claim 6, wherein, The compensation scheme includes at least one of color gamut mapping, color correction, color temperature compensation, mura compensation, and color gamut conversion.

9. A visual effect classification compensation device for a display panel, comprising: The visual effect feature acquisition module is used to acquire the visual effect feature attributes of N display panels, where N is a positive integer; The classification module, connected to the visual effect feature acquisition module, is used to classify the display panel into X cluster centers based on the acquired visual effect feature attributes, where X is a positive integer; The compensation module, connected to the classification module, is used to compensate various display panels according to the compensation scheme corresponding to the cluster center.

10. The visual effects classification compensation device according to claim 9, wherein, Also includes: A clustering module, connected to the classification module, is used to cluster multiple visual effect feature attributes according to a clustering algorithm to obtain X cluster centers; The storage module, connected to the visual effect feature acquisition module, the compensation module, and the clustering module, is used to store the acquired visual effect feature attributes and compensation schemes of the N display panels.

11. The visual effects classification compensation device according to claim 10, wherein, The clustering module further includes: verifying the obtained X cluster centers; if the verification passes, outputting X cluster centers; if the verification fails, obtaining X cluster centers again based on the visual effect feature attributes and clustering algorithm, until the X cluster centers pass the verification.

12. The visual effects classification compensation device according to claim 11, wherein, The clustering module further includes: when the number N of the display panels whose visual effect feature attributes are stored in the storage module is greater than the Y display panels, updating X cluster centers and corresponding compensation schemes according to the visual effect features of the newly acquired display panels.

13. An electronic device, wherein, The display panel includes a visual effect classification compensation device as described in any one of claims 9-12.

14. A computer-readable storage medium storing a computer program, wherein, When the computer program is executed by the processor, it implements the visual effects classification compensation method as described in any one of claims 1-8.

Images