Artificial intelligence-based optical compensation system, optical compensation method, and display device

Abstract

An artificial intelligence-based optical compensation system, an artificial intelligence-based optical compensation method, and a display apparatus. An artificial intelligence-based optical compensation system according to an embodiment of the disclosure can include a measurement apparatus configured to measure an optical characteristic of a display panel and output measurement result data of the optical characteristic, and an artificial intelligence-based optical compensation controller configured to predict and generate optical compensation result data corresponding to the measurement result data of the optical characteristic based on an artificial intelligence neural network using previous optical compensation result data for at least one other display panel, and store the predicted and generated optical compensation result data in a memory corresponding to the display panel.

Description

[0001] Cross-reference to related applications

[0002] This application claims priority to Korean Patent Application No. 10-2021-0170715, filed on December 2, 2021, which is incorporated herein by reference for all purposes, as if fully set forth herein. Technical Field

[0003] This disclosure relates to optical compensation systems, optical compensation methods, and display devices based on artificial intelligence. Background Technology

[0004] Self-emissive display devices (e.g., organic light-emitting display devices) use self-emissive light-emitting elements, and therefore self-emissive display devices have attracted attention due to advantages such as fast response time, high luminous efficiency, high brightness and wide viewing angle. Summary of the Invention

[0005] Self-emissive display devices, such as organic light-emitting display devices, may have different optical characteristics for each display panel due to various reasons in the process. Therefore, even when the same voltage or current is applied to multiple display panels of the same model, the color coordinates or brightness of the image realized for each display panel may vary.

[0006] In the existing display technology field, there is a problem that image quality deteriorates due to deviations in the optical characteristics of each self-emissive display panel (e.g., organic light-emitting display panel). Various optical compensation techniques have been proposed to address this issue, but they suffer from insufficient optical compensation performance or excessively long processing times. Therefore, the inventors of this specification have invented an artificial intelligence-based optical compensation system, optical compensation method, and display device as an accurate and fast optical compensation technology.

[0007] The embodiments of this disclosure can provide an artificial intelligence-based optical compensation system, optical compensation method, and display device as an accurate and fast optical compensation technology.

[0008] Embodiments of this disclosure can provide an artificial intelligence-based optical compensation system, optical compensation method, and display device, which can also perform display driving by predicting data voltages optimized for the optical characteristics of the display panel.

[0009] The embodiments of this disclosure can provide an AI-based optical compensation system and optical compensation method capable of actively and rapidly responding to changes in the characteristics and conditions of each display panel, as well as a display device applying AI-based optical compensation.

[0010] Embodiments of this disclosure may provide an artificial intelligence-based optical compensation system, the optical compensation system comprising: a measuring device configured to measure the optical characteristics of a display panel and output measurement result data of the optical characteristics; and an artificial intelligence-based optical compensation controller configured to predict and generate optical compensation result data corresponding to the measurement result data of the optical characteristics using previous optical compensation result data for at least one other display panel based on an artificial intelligence neural network, and to store the predicted and generated optical compensation result data in a memory corresponding to the display panel.

[0011] Embodiments of this disclosure may provide an artificial intelligence-based optical compensation method, the method comprising the following operations: measuring the optical characteristics of a display panel using a measuring device and generating measurement result data of the optical characteristics; performing artificial intelligence processing based on an artificial intelligence neural network using previous optical compensation result data for at least one other display panel; predicting and generating data voltage for each frequency band or grayscale as optical compensation result data corresponding to the measurement result data of the optical characteristics based on the result of the artificial intelligence processing; and storing information about the predicted and generated data voltage in a memory corresponding to the display panel.

[0012] Embodiments of this disclosure may provide a display device comprising: a display panel including data lines; a memory configured to store information about data voltages for each frequency band or grayscale; and a data driving circuit configured to output a data voltage corresponding to the current frequency band or grayscale from the data voltages for each frequency band or grayscale to the data lines.

[0013] In the display device according to the embodiments of the present disclosure, the information about the data voltage for each frequency band or grayscale stored in the memory can be optical compensation result data predicted by artificial intelligence processing based on an artificial intelligence neural network and corresponding to the measurement result data of the optical characteristics of the display panel and stored in the memory.

[0014] According to embodiments of this disclosure, an artificial intelligence-based optical compensation system, optical compensation method, and display device can be provided as an accurate and fast optical compensation technology.

[0015] According to embodiments of this disclosure, an artificial intelligence-based optical compensation system, optical compensation method, and display device can be provided, which can also perform display driving by predicting data voltages optimized for the optical characteristics of the display panel.

[0016] According to embodiments of this disclosure, an AI-based optical compensation system and method, capable of actively and rapidly responding to changes in the characteristics and conditions of each display panel, and a display device applying AI-based optical compensation can be provided. Attached Figure Description

[0017] The above and other aspects, features and advantages of this disclosure will become clearer from the following detailed description taken in conjunction with the accompanying drawings, in which:

[0018] Figure 1 This is a system configuration diagram of a display device according to an embodiment of the present disclosure;

[0019] Figure 2 It is the equivalent circuit of the sub-pixel of the display panel according to an embodiment of the present disclosure;

[0020] Figure 3 It is another equivalent circuit of the sub-pixel of the display panel according to an embodiment of the present disclosure;

[0021] Figure 4 This is a diagram illustrating an artificial intelligence-based optical compensation system according to an embodiment of the present disclosure;

[0022] Figure 5 This is a flowchart of an artificial intelligence-based optical compensation method according to an embodiment of the present disclosure;

[0023] Figure 6 This is a diagram illustrating an affine layer neural network for artificial intelligence-based optical compensation according to an embodiment of the present disclosure;

[0024] Figure 7 These are diagrams illustrating machine learning for artificial intelligence-based optical compensation according to embodiments of the present disclosure; and

[0025] Figure 8 This is a schematic diagram of a display device applying artificial intelligence-based optical compensation according to an embodiment of the present disclosure. Detailed Implementation

[0026] In the following description, some embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. When elements in the drawings are indicated by reference numerals, the same elements will be indicated by the same reference numerals, even if they are shown in different drawings. Furthermore, in the following description of the present disclosure, detailed descriptions of known functions and configurations incorporated herein will be omitted where such descriptions would make the subject matter of the disclosure considerably unclear. Where terms such as “comprising,” “having,” “including,” etc., described in this specification are used, additional terms may be added unless more restrictive terms such as “only” are used. Unless otherwise stated, singular terms may include plural forms.

[0027] In addition, when describing the components of this disclosure, terms such as first, second, A, B, (a), (b), etc., may be used herein. Each of these terms is not used to define the nature, order, or sequence of the corresponding component, but only to distinguish the corresponding component from other components.

[0028] When describing a structural element as being "connected to," "linked to," or "in contact with" another structural element, it should be interpreted as the other structural element being "connected to," "linked to," or "in contact with" that structural element, or a structural element being directly connected to or directly in contact with another structural element. Here, other components may be included in one or more of the two or more components that are "connected," "linked," or "connected" to each other.

[0029] In descriptions of temporal relationships relating to components, operating methods, manufacturing methods, etc., such as when time sequence or process sequence is described as “after,” “following,” “next,” “before,” etc., discontinuous instances may be included unless “immediately following” or “directly” is used.

[0030] Furthermore, when referring to numerical values ​​or corresponding information for components (e.g., grades, etc.), although there is no separate explicit description, the numerical values ​​or corresponding information can be interpreted as including a range of errors that may be caused by various factors.

[0031] In the following, various embodiments of the present disclosure will be described in detail with reference to the accompanying drawings.

[0032] Figure 1 This is a system configuration diagram of a display device 100 according to an embodiment of the present disclosure.

[0033] Reference Figure 1 The display driving system of the display device 100 according to the embodiments of the present disclosure may include a display panel 110 and a driving circuit for driving the display panel 110.

[0034] The display panel 110 may include a display area DA for displaying images and a non-display area NDA for not displaying images. The display panel 110 may include a plurality of subpixels SP disposed on the substrate SUB for image display. For example, the plurality of subpixels SP may be disposed in the display area DA. In some cases, at least one subpixel SP may be disposed in the non-display area NDA. At least one subpixel SP disposed in the non-display area NDA is also referred to as a dummy subpixel.

[0035] The display panel 110 may include multiple signal lines disposed on the substrate SUB for driving multiple sub-pixels SP. For example, the multiple signal lines may include multiple data lines DL and multiple gating lines GL. Depending on the structure of the sub-pixels SP, the signal lines may also include signal lines other than the multiple data lines DL and multiple gating lines GL. For example, other signal lines may include driving voltage lines, reference voltage lines, etc.

[0036] Multiple data lines DL and multiple gating lines GL can intersect each other. Each data line DL can be configured to extend in a first direction. Each gating line GL can be configured to extend in a second direction. Here, the first direction can be a column direction, and the second direction can be a row direction. In this specification, the column direction and the row direction are relative. For example, the column direction can be vertical, and the row direction can be horizontal. As another example, the column direction can be horizontal, and the row direction can be vertical. In the following, for ease of description, it is assumed that each data line DL is configured to extend in the vertical direction, and each gating line GL is configured to extend in the horizontal direction.

[0037] The driving circuit may include a data driving circuit 120 for driving multiple data lines DL and a gating driving circuit 130 for driving multiple gating lines GL. The driving circuit may also include a controller 140 for controlling the data driving circuit 120 and the gating driving circuit 130.

[0038] The data driving circuit 120 can be a circuit for driving multiple data lines DL, and can output a data signal (also known as a data voltage) corresponding to the image signal to the multiple data lines DL. The gating driving circuit 130 can be a circuit for driving multiple gating lines GL, and can generate gating signals to output the gating signals to the multiple gating lines GL.

[0039] The controller 140 can initiate scanning based on timing implemented in each frame and drive control data at appropriate times according to the scanning. The controller 140 can convert input image data from external sources to match the data signal format used in the data drive circuit 120 and provide the converted image data to the data drive circuit 120.

[0040] The controller 140 can receive display drive control signals and input image data from the external host system 150. For example, the display drive control signals may include vertical sync signals (VSYNC), horizontal sync signals (HSYNC), input data enable signals (DE), clock signals, etc.

[0041] The controller 140 can generate a data drive control signal DCS and a gating drive control signal GCS based on the display drive control signal input from the host system 150. The controller 140 can control the drive operation and timing of the data drive circuit 120 by providing the data drive control signal DCS to the data drive circuit 120. The controller 140 can control the drive operation and timing of the gating drive circuit 130 by providing the gating drive control signal GCS to the gating drive circuit 130.

[0042] The data driver circuit 120 may include one or more source driver integrated circuits (SDICs). Each source driver integrated circuit (SDIC) may include a shift register, latch circuit, digital-to-analog converter (DAC), output buffer, etc. In some cases, each source driver integrated circuit (SDIC) may also include an analog-to-digital converter (ADC).

[0043] For example, each source driver integrated circuit (SDIC) can be connected to the display panel 110 via a tape auto-bonding (TAB) method, connected to the bonding pads of the display panel 110 via a chip-on-glass (COG) or chip-on-panel (COP) method, or implemented and connected to the display panel 110 via a chip-on-film (COF) method.

[0044] The gating drive circuit 130 can output a gating signal with a conduction level voltage or a gating signal with a cutoff level voltage according to the control of the controller 140. The gating drive circuit 130 can sequentially drive multiple gating lines GL by sequentially providing the gating signal with the conduction level voltage to multiple gating lines GL.

[0045] The gate drive circuit 130 can be connected to the display panel 110 via a tape-on-absence (TAB) method, to the bonding pads of the display panel 110 via a chip-on-glass (COG) method or a chip-on-panel (COP) method, or to the display panel 110 via a chip-on-film (COF) method. Alternatively, the gate drive circuit 130 can be formed in the non-display area NDA of the display panel 110 as a gate-in-panel (GIP) type. The gate drive circuit 130 can be disposed on the substrate or connected to the substrate. That is, in the case of the GIP type, the gate drive circuit 130 can be disposed in the non-display area NDA of the substrate. In the case of the chip-on-glass (COG) type, chip-on-film (COF) type, etc., the gate drive circuit 130 can be connected to the substrate.

[0046] Furthermore, at least one of the driving circuits in the data driving circuit 120 and the gating driving circuit 130 may be disposed in the display area DA. For example, at least one of the driving circuits in the data driving circuit 120 and the gating driving circuit 130 may be configured not to overlap with the sub-pixel SP, and some or all of the driving circuits may be configured to overlap with the sub-pixel SP.

[0047] The data driving circuit 120 can be connected to one side of the display panel 110 (e.g., the top or bottom side). Depending on the driving method, panel design method, etc., the data driving circuit 120 can be connected to both sides of the display panel 110 (e.g., the top and bottom sides), or it can be connected to two or more of the four sides of the display panel 110.

[0048] The gating drive circuit 130 can be connected to one side of the display panel 110 (e.g., the left or right side). Depending on the driving method, panel design method, etc., the gating drive circuit 130 can be connected to both sides of the display panel 110 (e.g., the left and right sides), or it can be connected to two or more of the four sides of the display panel 110.

[0049] The controller 140 can be implemented as a separate component from the data drive circuit 120, or it can be implemented as an integrated circuit by integrating it with the data drive circuit 120. The controller 140 can be a timing controller used in conventional display technology, a control device (including a timing controller) capable of further performing other control functions, a control device different from a timing controller, or circuitry within a control device. The controller 140 can be implemented as various circuits or electronic components, such as integrated circuits (ICs), field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), or processors.

[0050] The controller 140 can be mounted on a printed circuit board, flexible printed circuit board, etc., and can be electrically connected to the data drive circuit 120 and the gating drive circuit 130 via the printed circuit board, flexible printed circuit board, etc. The controller 140 can send signals to and receive signals from the data drive circuit 120 according to one or more predetermined interfaces. Here, for example, the interface may include a low-voltage differential signaling (LVDS) interface, an EPI interface, a serial peripheral interface (SPI), etc.

[0051] The display device 100 according to embodiments of the present disclosure may be a self-emissive display device in which the display panel 110 emits its own light. When the display device 100 according to embodiments of the present disclosure is a self-emissive display device, each of the plurality of sub-pixels SP may include a light-emitting element. For example, the display device 100 according to embodiments of the present disclosure may be an organic light-emitting display device in which the light-emitting element is implemented as an organic light-emitting diode (OLED). As another example, the display device 100 according to embodiments of the present disclosure may be an inorganic light-emitting display device in which the light-emitting element is implemented as a light-emitting diode based on an inorganic material. As yet another example, the display device 100 according to embodiments of the present disclosure may be a quantum dot display device in which the light-emitting element is implemented as a quantum dot, and the quantum dot is a self-emissive semiconductor crystal.

[0052] Figure 2 The equivalent circuit of the sub-pixel SP of the display panel 110 according to an embodiment of the present disclosure is shown, and Figure 3 Another equivalent circuit of the sub-pixel SP of the display panel 110 according to an embodiment of the present disclosure is shown.

[0053] Reference Figure 2 In the display device 100 according to an embodiment of the present disclosure, each sub-pixel SP may include a light-emitting element ED, a driving transistor DRT for driving the light-emitting element ED by controlling the current flowing to the light-emitting element ED, a scan transistor SCT for sending a data voltage Vdata to a first node N1, which is the gate node of the driving transistor DRT, and a storage capacitor Cst for maintaining the voltage for a specific period of time.

[0054] A light-emitting element (ED) may include a pixel electrode (PE), a common electrode (CE), and a light-emitting layer (EL), with the EL located between the pixel electrode (PE) and the common electrode (CE). The pixel electrode (PE) of the ED can be either an anode or a cathode. The common electrode (CE) can be either a cathode or an anode. The ED can be, for example, an organic light-emitting diode (OLED), an inorganic light-emitting diode (LED), or a quantum dot light-emitting element.

[0055] The base voltage EVSS can be applied to the common electrode CE of the light-emitting element ED. Here, the base voltage EVSS can be, for example, ground voltage or a voltage similar to ground voltage.

[0056] The driving transistor DRT can be a transistor used to drive the light-emitting element ED, and can include a first node N1, a second node N2 and a third node N3.

[0057] The first node N1 of the driving transistor DRT can be a node corresponding to the gate node and can be electrically connected to the source node or drain node of the scanning transistor SCT. The second node N2 of the driving transistor DRT can be a source node or a drain node and can be electrically connected to the pixel electrode PE of the light-emitting element ED. The third node N3 of the driving transistor DRT can be a drain node or a source node and can be electrically connected to the driving voltage line DVL that provides the driving voltage EVDD. In the following description, for ease of description, it can be described as if the second node N2 of the driving transistor DRT is a source node and the third node N3 is a drain node.

[0058] The scanning transistor SCT can switch the connection between the data line DL and the first node N1 of the driving transistor DRT.

[0059] In response to the scan signal SCAN provided from the gate line GL, the scan transistor SCT can control the connection between the first node N1 of the drive transistor DRT and the corresponding data line DL among the multiple data lines DL.

[0060] The drain or source node of the scan transistor SCT can be electrically connected to the corresponding data line DL. The source or drain node of the scan transistor SCT can be electrically connected to the first node N1 of the driving transistor DRT. The gate node of the scan transistor SCT can be electrically connected to the gating line GL to receive the scan signal SCAN.

[0061] The scanning transistor SCT can be turned on by the scanning signal SCAN with a turn-on level voltage to send the data voltage Vdata provided from the corresponding data line DL to the first node N1 of the driving transistor DRT.

[0062] The scanning transistor SCT is turned on by a scan signal SCAN with a turn-on voltage and turned off by a scan signal SCAN with a turn-off voltage. Here, when the scanning transistor SCT is an n-type transistor, the turn-on voltage can be a high level voltage and the turn-off voltage can be a low level voltage. When the scanning transistor SCT is a p-type transistor, the turn-on voltage can be a low level voltage and the turn-off voltage can be a high level voltage.

[0063] The storage capacitor Cst can be electrically connected between the first node N1 and the second node N2 of the driving transistor DRT to maintain the data voltage Vdata corresponding to the image signal voltage or the voltage corresponding to the image signal voltage within one frame.

[0064] The storage capacitor Cst may not be a parasitic capacitor (e.g., Cgs or Cgd) existing as an internal capacitor between the first node N1 and the second node N2 of the driving transistor DRT, but may be an external capacitor intentionally designed outside the driving transistor DRT.

[0065] because Figure 2 The sub-pixel SP shown has two transistors DRT and SCT and a capacitor Cst to drive the light-emitting element ED. Therefore, the sub-pixel SP is called a structure with 2T (transistors) and 1C (capacitor).

[0066] Reference Figure 3 In the display device 100 according to an embodiment of the present disclosure, each sub-pixel SP may further include a sensing transistor SENT for initialization operations and sensing operations.

[0067] In this case, due to Figure 3 The sub-pixel SP shown has three transistors DRT, SCT and SENT, and a capacitor Cst to drive the light-emitting element ED. Therefore, the sub-pixel SP is called a 3T (transistor) 1C (capacitor) structure.

[0068] The sensing transistor SENT can switch the connection between the second node N2 of the driving transistor DRT and the reference voltage line RVL.

[0069] The sensing transistor SENT can control the connection between the second node N2 of the driving transistor DRT, which is electrically connected to the pixel electrode PE of the light-emitting element ED, and the corresponding reference voltage line RVL among the multiple reference voltage lines RVL, in response to the sensing signal SENSE.

[0070] The drain or source node of the sensing transistor SENT can be electrically connected to the reference voltage line RVL. The source or drain node of the sensing transistor SENT can be electrically connected to the second node N2 of the driving transistor DRT, and can also be electrically connected to the pixel electrode PE of the light-emitting element ED. The gate node of the sensing transistor SENT can receive the sensing signal SENSE.

[0071] The sensing transistor SENT can be turned on to apply the reference voltage Vref provided from the reference voltage line RVL to the second node N2 of the driving transistor DRT.

[0072] The sensing transistor SENT is turned on by a sensing signal SENSE at a turn-on voltage level and turned off by a sensing signal SENSE at a turn-off voltage level. Here, when the sensing transistor SENT is an n-type transistor, the turn-on voltage level can be high and the turn-off voltage level can be low. When the sensing transistor SENT is a p-type transistor, the turn-on voltage level can be low and the turn-off voltage level can be high.

[0073] Each of the driving transistor DRT, the scanning transistor SCT, and the sensing transistor SENT can be an n-type transistor or a p-type transistor. All of the driving transistor DRT, the scanning transistor SCT, and the sensing transistor SENT can be n-type transistors or p-type transistors. At least one of the driving transistor DRT, the scanning transistor SCT, and the sensing transistor SENT can be an n-type transistor (or a p-type transistor), while the others can be p-type transistors (or n-type transistors).

[0074] The gate node of each of the scan transistor SCT and the sense transistor SENT can be connected to the same single gate line GL. Alternatively, the gate node of each of the scan transistor SCT and the sense transistor SENT can be connected to different gate lines GL.

[0075] A reference voltage line RVL can be configured for one sub-pixel column. Alternatively, a reference voltage line RVL can be configured for two or more sub-pixel columns. When a reference voltage line RVL is configured for two or more sub-pixel columns, multiple sub-pixels SP can receive a reference voltage Vref from one reference voltage line RVL. For example, a reference voltage line RVL can be configured for four sub-pixel columns. That is, the sub-pixels SP included in four sub-pixel columns can share one reference voltage line RVL.

[0076] A driving voltage line (DVL) can be configured for one sub-pixel column. Alternatively, a driving voltage line (DVL) can be configured for two or more sub-pixel columns. When a driving voltage line (DVL) is configured for two or more sub-pixel columns, multiple sub-pixels (SPs) can receive a driving voltage (EVDD) from one driving voltage line (DVL). For example, a driving voltage line (DVL) can be configured for four sub-pixel columns. That is, the sub-pixels (SPs) included in four sub-pixel columns can share one driving voltage line (DVL).

[0077] Figure 3 The 3T1C structure of the sub-pixel SP shown is merely an example for illustration and may also include one or more transistors, or in some cases, one or more capacitors. Alternatively, each of the multiple sub-pixels may have the same structure, and some of the multiple sub-pixels may have different structures.

[0078] Furthermore, the display device 100 according to the embodiments of this disclosure may have a top-emitting structure or a bottom-emitting structure.

[0079] Furthermore, when the display device 100 according to the embodiments of this disclosure is a self-emissive display device, such as an organic light-emitting display device, due to various reasons in the process, the display device 100 may have optical characteristics that are different from the actual desired optical characteristics (e.g., brightness, color coordinates, etc.), and therefore have color coordinates or brightness that are different from the desired color coordinates or brightness of the image.

[0080] Therefore, embodiments of this disclosure can provide an artificial intelligence-based optical compensation system and optical compensation method, which can also perform display driving by predicting data voltage optimized for the optical characteristics (e.g., brightness, etc.) of the display panel 110, and can provide an optical compensation method and a display device applying artificial intelligence-based optical compensation.

[0081] Taking into account the optical characteristics of the display panel 110 (e.g., brightness, color coordinates, etc.), an AI-based optical compensation system and optical compensation method, as well as a display device applying AI-based optical compensation, will be described in more detail as a more accurate and faster optical compensation technology.

[0082] Figure 4 An artificial intelligence-based optical compensation system 400 according to an embodiment of the present disclosure is shown.

[0083] Reference Figure 4 The AI-based optical compensation system 400 according to embodiments of the present disclosure is a system that uses artificial intelligence to perform optical compensation, and the optical compensation system 400 may include a measuring device 410 and an AI-based optical compensation controller 420.

[0084] The measuring device 410 can measure the optical characteristics of the display panel 110 and output measurement result data of the optical characteristics. For example, the measuring device 410 may include a luminance meter, etc.

[0085] The AI-based optical compensation controller 420 can use previous optical compensation result data for at least one other display panel to predict and generate optical compensation result data corresponding to the measurement result data of optical properties based on an AI neural network.

[0086] The AI-based optical compensation controller 420 can use all previous optical compensation result data for at least one other display panel and use an AI neural network to predict the current optical compensation result data, wherein the at least one other display panel is a sample of optical compensation that has been completed based on AI.

[0087] The AI-based optical compensation controller 420 can store the optical compensation result data generated by prediction using AI in a memory 430 corresponding to the display panel 110.

[0088] For example, the optical compensation result data predicted and generated by the AI-based optical compensation controller 420 may include information about the predicted data voltage for each desired target.

[0089] For example, the desired target may include a desired band, brightness, or color coordinates. Here, the band is also referred to as a luminance mode (brightness mode), and the brightness of the display panel 110 can be controlled within one of the various bands.

[0090] For example, the previous optical compensation result data for at least one other display panel may be data obtained as a result of previous optical compensation processing performed on at least one other display panel, and may include information such as data voltage, gamma voltage, etc.

[0091] The AI-based optical compensation controller 420 can generate machine learning result data by performing machine learning (ML) on previous optical compensation result data for at least one other display panel, which is a sample of the AI-based optical compensation that has been completed.

[0092] The AI-based optical compensation controller 420 can use machine learning result data and optical property measurement result data to predict and generate optical compensation result data corresponding to the optical property measurement result data based on an AI neural network.

[0093] The AI-based optical compensation controller 420 can perform log file collection processing using previous optical compensation result data collection for at least one other display panel as a large log file, perform a data processing process to select learning data for machine learning from the collected log file, and perform machine learning based on the selected learning data to generate machine learning result data.

[0094] The AI-based optical compensation controller 420 can perform preprocessing before measuring (main measurement) the optical characteristics of the display panel 110 via the measuring device 410 to obtain measurement result data of the optical characteristics, in order to optimize and set the driving voltage by controlling via the measuring device 410 to initially measure the optical characteristics of the display panel 110.

[0095] The driving voltage can be the voltage used when driving the display panel 110 while measuring the optical characteristics of the display panel 110 (main measurement) through the measuring device 410.

[0096] For example, the driving voltage may include a base voltage EVSS or a black data voltage provided to the sub-pixels SP included in the display panel 110, or it may include a brightness weight for each area in the display panel 110.

[0097] Figure 5 This is a flowchart of an artificial intelligence-based optical compensation method according to an embodiment of the present disclosure.

[0098] Reference Figure 5 The AI-based optical compensation system 400 according to the embodiments of this disclosure can execute an AI-based optical compensation method.

[0099] Reference Figure 5 The artificial intelligence-based optical compensation method according to the embodiments of this disclosure may include a master measurement operation (S520), an artificial intelligence processing execution operation (S560), a data voltage prediction operation (S570), a prediction information storage operation (S590), etc.

[0100] In the main measurement operation (S520), the AI-based optical compensation controller 420 of the AI-based optical compensation system 400 can measure the optical characteristics of the display panel 110 through the measuring device 410 to generate measurement result data of the optical characteristics.

[0101] In the artificial intelligence processing execution operation (S560), the artificial intelligence-based optical compensation controller 420 of the artificial intelligence-based optical compensation system 400 can use previous optical compensation result data for at least one other display panel to perform artificial intelligence processing based on artificial intelligence neural networks.

[0102] In the data voltage prediction operation (S570), the AI-based optical compensation controller 420 of the AI-based optical compensation system 400 can predict and generate data voltage for each frequency band or grayscale as optical compensation result data corresponding to the measurement result data of optical characteristics based on the results of performing AI processing.

[0103] In the prediction information storage operation (S590), the AI-based optical compensation controller 420 of the AI-based optical compensation system 400 can store information about the data voltage generated by the prediction in the data voltage prediction operation (S570) in a memory 430 corresponding to the display panel 110.

[0104] Reference Figure 5The AI-based optical compensation method according to embodiments of the present disclosure may further include a machine learning process operation (S550), which generates machine learning result data by performing machine learning using previous optical compensation result data for at least one other display panel by the AI-based optical compensation controller 420 of the AI-based optical compensation system 400 before performing the AI ​​processing operation (S560).

[0105] In the operation of performing artificial intelligence processing (S560), the artificial intelligence-based optical compensation controller 420 of the artificial intelligence-based optical compensation system 400 can predict and generate optical compensation result data corresponding to the measurement result data of optical characteristics based on machine learning result data and measurement result data of optical characteristics.

[0106] Reference Figure 5 The AI-based optical compensation method according to embodiments of the present disclosure may further include a log file collection operation (S530), wherein the AI-based optical compensation controller 420 of the optical compensation system 400 collects a log file as big data using previous optical compensation result data for at least one other display panel before the machine learning execution operation (S550).

[0107] Reference Figure 5 The AI-based optical compensation method according to the embodiments of this disclosure may further include a data processing operation (S540), wherein the AI-based optical compensation controller 420 of the optical compensation system 400 selects learning data for machine learning from the collected log files after the log file collection operation (S530).

[0108] Reference Figure 5 The AI-based optical compensation method according to the embodiments of the present disclosure may further include a preprocessing operation (S510), wherein the AI-based optical compensation controller 420 of the optical compensation system 400 sets the driving voltage by being controlled via the measuring device 410 to preliminarily measure the optical characteristics of the display panel 110 before generating measurement result data of optical characteristics (S520).

[0109] Reference Figure 5 The AI-based optical compensation method according to the embodiments of this disclosure may further include a cyclic control operation (S580), wherein the AI-based optical compensation system 400’s intelligent optical compensation controller 420 changes the frequency band and point after the operation of predicting and generating data voltage (S570).

[0110] After the cyclic control operation (S580), the AI-based optical compensation controller 420 of the AI-based optical compensation system 400 can repeatedly perform the operation of generating measurement result data of optical characteristics (S520), the operation of performing AI processing (S560), and the operation of predicting and generating data voltage (S570).

[0111] Figure 6 An affine layer neural network 600, as an artificial intelligence neural network for artificial intelligence-based optical compensation, is shown according to an embodiment of the present disclosure.

[0112] Reference Figure 6 For example, the artificial intelligence neural network used for AI-based optical compensation could be an affine layer neural network 600.

[0113] Reference Figure 6 The affine layer neural network 600 may include: an input layer Lin, which includes a plurality of input nodes R, G and B corresponding to the processing information of the first preprocessing; a first intermediate layer Lm1, which includes a plurality of first intermediate nodes R1, G1 and B1 corresponding to the processing information of the second preprocessing; a second intermediate layer Lm2, which includes a plurality of second intermediate nodes R2, G2 and B2 corresponding to the processing information of the third preprocessing; and an output layer Lout, which includes a plurality of output nodes R3, G3 and B3 corresponding to the processing information of the main processing.

[0114] Multiple input nodes R, G, and B can be connected to all or some of the multiple first intermediate nodes R1, G1, and B1; multiple first intermediate nodes R1, G1, and B1 can be connected to all or some of the multiple second intermediate nodes R2, G2, and B2; and multiple second intermediate nodes R2, G2, and B2 can be connected to all or some of the multiple output nodes R3, G3, and B3. For example, the multiple input nodes R, G, and B, the multiple first intermediate nodes R1, G1, and B1, the multiple second intermediate nodes R2, G2, and B2, and the multiple output nodes R3, G3, and B3 can correspond to red image signals (red data), green image signals (green data), and blue image signals (blue data).

[0115] Reference Figure 6 Regarding the affine layer neural network 600, the first preprocessing may be a pre-optical compensation process for each color coordinate or brightness, the second preprocessing may be a pre-optical compensation process for each color using a first brightness value, the third preprocessing may be a pre-optical compensation process for each color using a second brightness value higher than the first brightness value, and the main processing may be a processing for obtaining measurement result data of optical properties.

[0116] Alternatively, regarding the affine layer neural network 600, the main processing can correspond to the optical compensation processing, and the processing information of the main processing can correspond to the optical compensation result data. The second and third preprocessing can be substantially the same as the optical compensation processing, or they can be pre-optical compensation processing performed before the optical compensation processing. Optimization of the driving voltage can be performed through the second and / or third preprocessing.

[0117] The AI-based optical compensation controller 420 of the AI-based optical compensation system 400 can update the AI ​​neural network based on the predicted and generated optical compensation result data.

[0118] Figure 7 Machine learning for artificial intelligence-based optical compensation according to embodiments of the present disclosure is illustrated.

[0119] Reference Figure 7 The AI-based optical compensation system 400's AI-based optical compensation controller 420 can store and manage N log files (a predetermined number) in real time while simultaneously updating them (S700), and perform machine learning using the N log files (S710). Here, the real-time updating, storage, and management of the log files can be combined with… Figure 5 The operations S530 and S540 in the AI-based optical compensation processing correspond to these. Machine learning can be used with... Figure 5 This corresponds to operation S550 in the AI-based optical compensation processing.

[0120] Reference Figure 7 The AI-based optical compensation controller 420 can perform AI-based optical compensation on the new display panel using machine learning result data obtained from the execution of machine learning (S720). Here, AI-based optical compensation can be combined with... Figure 5 Operations S560 and S570 in the AI-based optical compensation processing correspond to each other.

[0121] Reference Figure 7 When the AI-based optical compensation for the new display panel is completed, the AI-based optical compensation controller 420 can generate the command CMD_ML to continue machine learning (ML).

[0122] Furthermore, when AI-based optical compensation for the new display panel is completed, the AI-based optical compensation controller 420 can store the AI-based optical compensation result data for the new display panel as a new log to update the log file in real time. Therefore, the AI-based optical compensation controller 420 can update and store the optical compensation result data stored as a new log as previous optical compensation result data, and delete the log of the oldest previous optical compensation result data to manage log files of N logs (a predetermined number) (S700). In this case, for example, the maintenance of the log files of the N logs can be performed using a first-in, first-out (FIFO) method.

[0123] Reference Figure 7 The AI-based optical compensation controller 420 can re-execute machine learning using a real-time updated log file according to the command CMD_ML to continue machine learning (ML) (S710).

[0124] Figure 8 This is a schematic diagram of a display device 100 that applies artificial intelligence-based optical compensation according to an embodiment of the present disclosure.

[0125] Reference Figure 8 The display device 100 based on artificial intelligence optical compensation according to embodiments of the present disclosure may include: a display panel 110 including a data line DL; a memory 430 storing information about data voltages for each frequency band or grayscale; and a data driving circuit 120 outputting the data voltages of each frequency band or grayscale stored in the memory 430 corresponding to display driving information (e.g., the current frequency band or the current grayscale) to the data line.

[0126] The controller 140 can select a data voltage corresponding to display driving information (e.g., current frequency band or current grayscale) by referring to the data voltage of each frequency band or grayscale stored in the memory 430, and provide the data corresponding to the selected data voltage to the data driving circuit 120.

[0127] For example, based on the execution results of artificial intelligence processing based on artificial intelligence neural networks, information about the data voltage of each frequency band or grayscale stored in memory 430 can be predicted and stored in memory 430 as optical compensation result data corresponding to the measurement result data of the optical characteristics of display panel 110.

[0128] For example, the optical compensation result data stored in memory 430 and predicted based on artificial intelligence-based optical compensation may include information about the predicted data voltage for each desired target. Here, for example, the desired target may include a desired frequency band, brightness, or color coordinates.

[0129] For example, the optical compensation result data stored in memory 430 and predicted based on artificial intelligence-based optical compensation may also include information such as gamma voltage.

[0130] The AI-based optical compensation according to the embodiments of the present disclosure described above is a process performed to achieve the same color coordinates and brightness for each object (display panel) taking into account the optical characteristics of a self-emissive display, such as an OLED display.

[0131] Since the luminance of each of the red, green, and blue subpixels is different for each object (display panel), the image quality of a self-emissive display, such as an OLED display, can be greatly improved by applying AI-based optical compensation according to embodiments of this disclosure.

[0132] According to the AI-based optical compensation technology of this disclosure, the learning data for machine learning used in optical compensation can be automatically updated on the production line. Therefore, by performing optical compensation immediately, it is possible to proactively respond to changes in the characteristics and conditions of each display panel and significantly reduce optical compensation processing time.

[0133] The AI-based optical compensation system 400 according to embodiments of this disclosure can use, for example, an AI neural network of an affine layer neural network 600. The affine layer neural network 600 is structured such that all nodes in the affine layer neural network 600 are connected to all nodes in the next layer. For example, in the viewpoint of the second intermediate node of the second intermediate layer Lm2, the input node of the input layer Lin and the first intermediate node of the first intermediate layer Lm1 are both nodes of the previous layer, and the output nodes of the output layer (Lout) are all nodes of the next layer.

[0134] In the structure of the affine layer neural network 600, since the nodes of the previous layer are connected to all the nodes of the next layer in the optical compensation result, the result of the current desired point can be predicted from multiple points.

[0135] According to embodiments of the present disclosure, an artificial intelligence-based optical compensation system 400 can use the results of previous optical compensation processing (previous optical compensation result data) to predict the results of subsequent optical compensation processing (optical compensation result data).

[0136] Since the AI-based optical compensation system 400 according to embodiments of this disclosure uses an AI neural network, such as an affine layer neural network 600 structure, the result value (optical compensation result data) of the current optical compensation point (e.g., grayscale) can be predicted using all previous optical compensation result data (e.g., color coordinates, brightness, data voltage Vdata, base voltage EVSS, etc.) from previous optical compensation processing for previous samples (other display panels) and all result data (e.g., data voltage Vdata, etc.) from previous optical compensation points (grayscale). For example, the optical compensation result data may include data voltage, etc.

[0137] The AI-based optical compensation system 400 according to embodiments of this disclosure can perform learning processing with big data in advance to perform machine learning, and for this purpose, the learning data can be automatically updated in real time.

[0138] The AI-based optical compensation system 400 according to embodiments of this disclosure can automatically collect log files at each optical compensation completion time to perform machine learning.

[0139] A brief description of the embodiments of this disclosure described above is as follows.

[0140] Embodiments of this disclosure may provide an artificial intelligence-based optical compensation system, comprising: a measuring device configured to measure the optical characteristics of a display panel and output measurement result data of the optical characteristics; and an artificial intelligence-based optical compensation controller configured to predict and generate optical compensation result data corresponding to the measurement result data of the optical characteristics using previous optical compensation result data for at least one other display panel based on an artificial intelligence neural network, and to store the predicted and generated optical compensation result data in a memory corresponding to the display panel.

[0141] The predicted and generated optical compensation results data can include information about the data voltage for each desired target.

[0142] The desired target may include the desired frequency band, brightness, or color coordinates.

[0143] The predicted and generated optical compensation results data may also include information such as gamma voltage.

[0144] Artificial intelligence neural networks can be affine layer neural networks.

[0145] The affine layer neural network may include: an input layer comprising a plurality of input nodes corresponding to processing information of a first preprocessing; a first intermediate layer comprising a plurality of first intermediate nodes corresponding to processing information of a second preprocessing; a second intermediate layer comprising a plurality of second intermediate nodes corresponding to processing information of a third preprocessing; and an output layer comprising a plurality of output nodes corresponding to processing information of a main processing.

[0146] Multiple input nodes can be connected to all of the multiple first intermediate nodes, multiple first intermediate nodes can be connected to all of the multiple second intermediate nodes, and multiple second intermediate nodes can be connected to all of the multiple output nodes.

[0147] The first preprocessing can be a pre-optical compensation process for each color coordinate or brightness, the second preprocessing can be a pre-optical compensation process for each color using a first brightness value, the third preprocessing can be a pre-optical compensation process for each color using a second brightness value that is higher than the first brightness value, and the main processing can be a processing of measurement result data for obtaining optical properties.

[0148] Artificial intelligence-based optical compensation controllers can update artificial intelligence neural networks based on predicted and generated optical compensation result data.

[0149] The previous optical compensation result data of at least one other display panel is obtained as a result of previous optical compensation processing performed on at least one other display panel, and may include information about data voltage and gamma voltage.

[0150] An AI-based optical compensation controller can generate machine learning result data by performing machine learning using previous optical compensation result data for at least one other display panel, and can use the machine learning result data and measurement result data of optical properties to predict and generate optical compensation result data corresponding to the measurement result data of optical properties based on an AI neural network.

[0151] The AI-based optical compensation controller can perform log file collection processing to collect log files as big data using previous optical compensation result data for at least one other display panel, can perform data processing to select learning data for machine learning from the collected log files, and can generate machine learning result data by performing machine learning based on the selected learning data.

[0152] The AI-based optical compensation controller can perform preprocessing to set the driving voltage by controlling the measurement device to initially measure the optical characteristics of the display panel before obtaining measurement result data of the optical characteristics by measuring the optical characteristics of the display panel via the measurement device.

[0153] The driving voltage may include a black data voltage or a base voltage provided to the sub-pixels contained in the display panel, or it may include a brightness weight for each area.

[0154] Embodiments of this disclosure may provide an artificial intelligence-based optical compensation method, the method comprising the following operations: measuring the optical characteristics of a display panel using a measuring device to generate measurement result data of the optical characteristics; performing artificial intelligence processing based on an artificial intelligence neural network using previous optical compensation result data for at least one other display panel; predicting and generating data voltage for each frequency band or grayscale as optical compensation result data corresponding to the measurement result data of the optical characteristics based on the result of the artificial intelligence processing; and storing information about the predicted and generated data voltage in a memory corresponding to the display panel.

[0155] The artificial intelligence-based optical compensation method according to this disclosure may further include a machine learning process operation, which generates machine learning result data by performing machine learning using the previous optical compensation result data of the at least one other display panel before performing the artificial intelligence processing operation.

[0156] In the process of performing artificial intelligence processing, the optical compensation result data corresponding to the measurement result data of optical properties can be predicted and generated based on machine learning result data and measurement result data of optical properties.

[0157] The AI-based optical compensation method according to this disclosure may further include the following operations: a log file collection operation prior to the machine learning process operation, the log file collection operation collecting log files as large data using previous optical compensation result data from at least one other display panel; and a data processing operation selecting learning data for machine learning from the collected log files.

[0158] The artificial intelligence-based optical compensation method according to this disclosure may further include a preprocessing operation, which sets the driving voltage by being controlled via a measuring device to preliminarily measure the optical characteristics of the display panel before generating measurement result data of optical properties.

[0159] The artificial intelligence-based optical compensation method according to this disclosure may also include a cyclic control operation that changes the frequency band and point after the operation of predicting and generating data voltage.

[0160] After the cyclic control operation, the operations of generating measurement results data of optical properties, performing artificial intelligence processing, and predicting and generating data voltages can be repeated.

[0161] Embodiments of this disclosure may provide a display device comprising: a display panel including data lines; a memory storing information about data voltages for each frequency band or grayscale; and a data driving circuit outputting a data voltage corresponding to the current frequency band or grayscale from the data voltages for each frequency band or grayscale to the data lines.

[0162] In the display device according to an embodiment of the present disclosure, the information about the data voltage of each frequency band or grayscale stored in the memory can be information that is predicted and generated as optical compensation result data corresponding to the measurement result data of the optical characteristics of the display panel based on the result of artificial intelligence processing performed based on an artificial intelligence neural network and stored in the memory.

[0163] In a display device according to an embodiment of the present disclosure, the predicted and generated optical compensation result data may include information about the data voltage of each desired target.

[0164] In a display device according to an embodiment of the present disclosure, the desired target may include a desired frequency band, brightness or color coordinates, and the predicted and generated optical compensation result data may also include information such as gamma voltage.

[0165] According to the above embodiments of this disclosure, an artificial intelligence-based optical compensation system, optical compensation method, and display device can be provided as an accurate and fast optical compensation technology.

[0166] According to embodiments of this disclosure, an artificial intelligence-based optical compensation system, optical compensation method, and display device can be provided, which can also perform display driving by predicting data voltages optimized for the optical characteristics of the display panel.

[0167] According to embodiments of this disclosure, an AI-based optical compensation system and method, as well as a display device applying AI-based optical compensation, can be provided that can actively and quickly respond to changes in the characteristics and conditions of each display panel.

[0168] The above description provides examples of the technical concepts of this disclosure for illustrative purposes only. Various modifications and variations will be possible for those skilled in the art to which this disclosure pertains without departing from the essential characteristics of this disclosure. Furthermore, the embodiments disclosed herein are intended to illustrate the scope of the technical concepts of this disclosure, and the scope of this disclosure is not limited to these embodiments. The scope of protection of this disclosure should be interpreted in accordance with the appended claims to include all technical concepts within the scope of the claims equivalents.

Claims

1. An optical compensation system, the optical compensation system comprising: A measuring device configured to measure the optical properties of a display panel and output measurement result data of the optical properties; as well as An AI-based optical compensation controller is configured to use prior optical compensation result data for at least one other display panel to predict and generate optical compensation result data corresponding to measurement result data of the optical characteristics based on an AI neural network. The AI-based optical compensation controller is configured to store the predicted and generated optical compensation result data in a memory corresponding to the display panel. The AI-based optical compensation controller is further configured as follows: The process involves performing a log file collection process that uses the previous optical compensation result data for the at least one other display panel to collect log files as large data, performing a data processing process that selects learning data for machine learning from the collected log files, and generating machine learning result data by performing the machine learning based on the selected learning data.

2. The optical compensation system according to claim 1, wherein, The predicted and generated optical compensation results include information about the predicted data voltage for each frequency band, brightness, or color coordinate.

3. The optical compensation system according to claim 2, wherein, The predicted and generated optical compensation results also include information about the gamma voltage.

4. The optical compensation system according to claim 1, wherein, The artificial intelligence neural network is an affine layer neural network.

5. The optical compensation system according to claim 4, wherein, The affine neural network includes: an input layer comprising multiple input nodes corresponding to first preprocessed processing information; a first intermediate layer comprising multiple first intermediate nodes corresponding to second preprocessed processing information; a second intermediate layer comprising multiple second intermediate nodes corresponding to third preprocessed processing information; and an output layer comprising multiple output nodes corresponding to main processing information. The plurality of input nodes are connected to all of the plurality of first intermediate nodes, the plurality of first intermediate nodes are connected to all of the plurality of second intermediate nodes, and the plurality of second intermediate nodes are connected to all of the plurality of output nodes.

6. The optical compensation system according to claim 5, wherein, The first preprocessing is a pre-optical compensation process for each color coordinate or brightness; the second preprocessing is a pre-optical compensation process for each color using a first brightness value; the third preprocessing is a pre-optical compensation process for each color using a second brightness value that is higher than the first brightness value; and the main processing is a process for obtaining measurement result data of the optical properties.

7. The optical compensation system according to claim 1, wherein, The AI-based optical compensation controller updates the AI ​​neural network based on the predicted and generated optical compensation result data.

8. The optical compensation system according to claim 1, wherein, The previous optical compensation result data for the at least one other display panel is data obtained as a result of previous optical compensation processing performed on the at least one other display panel, and includes information about the data voltage.

9. The optical compensation system according to claim 1, wherein, The AI-based optical compensation controller is configured as follows: The machine learning result data is generated by performing the machine learning using the previous optical compensation result data for the at least one other display panel; as well as Using the machine learning result data and the measurement result data of the optical properties, the optical compensation result data corresponding to the measurement result data of the optical properties is predicted and generated based on the artificial intelligence neural network.

10. The optical compensation system according to claim 1, wherein, The AI-based optical compensation controller is configured as follows: Before measuring the optical characteristics of the display panel using the measuring device to obtain measurement result data of the optical characteristics, a preprocessing step is performed to set the driving voltage by controlling the optical characteristics of the display panel via the measuring device to perform a preliminary measurement.

11. The optical compensation system according to claim 10, wherein, The driving voltage is the voltage used when driving the display panel while measuring the optical characteristics of the display panel via the measuring device.

12. The optical compensation system according to claim 11, wherein, The driving voltage includes a black data voltage or a base voltage provided to the sub-pixels contained in the display panel, or it includes a brightness weight for each region.

13. An optical compensation method, the optical compensation method comprising the following operations: The optical properties of the display panel are measured using a measuring device, and measurement result data of the optical properties are generated. Artificial intelligence processing is performed based on an artificial intelligence neural network using previous optical compensation results data for at least one other display panel; Based on the results of the artificial intelligence processing, the data voltage for each frequency band or grayscale is predicted and generated as optical compensation result data corresponding to the measurement result data of the optical characteristics; as well as Information about the predicted and generated data voltages is stored in a memory corresponding to the display panel. The optical compensation method further includes the following operations: A log file collection operation, which uses the previous optical compensation result data for the at least one other display panel to collect a log file as a large dataset; as well as The data processing operation selects learning data for machine learning from the collected log files.

14. The optical compensation method of claim 13, further comprising a machine learning process operation, wherein the machine learning process operation generates machine learning result data by performing the machine learning on the previous optical compensation result data for the at least one other display panel before performing the artificial intelligence processing operation. in, In the operation of performing the artificial intelligence processing, the optical compensation result data corresponding to the measurement result data of the optical properties is predicted and generated based on the machine learning result data and the measurement result data of the optical properties.

15. The optical compensation method according to claim 13, further comprising a preprocessing operation, wherein the preprocessing operation sets a driving voltage by being controlled via the measuring device to preliminarily measure the optical characteristics of the display panel prior to the operation of generating measurement result data of the optical characteristics.

16. The optical compensation method according to claim 13, further comprising a cyclic control operation, the cyclic control operation changing the frequency band and point after the operation of predicting and generating the data voltage. in, After the cyclic control operation, the operation of generating the measurement result data of the optical properties, the operation of performing the artificial intelligence processing, and the operation of predicting and generating the data voltage are repeated.

17. A display device, the display device comprising: The optical compensation system according to any one of claims 1-12; Display panel, the display panel including data cable; A memory configured to store information about data voltages for each frequency band or grayscale; as well as A data driving circuit is configured to output a data voltage corresponding to the current frequency band or current grayscale value to the data line for each frequency band or grayscale value. The information stored in the memory regarding the data voltage for each frequency band or grayscale is information that is predicted and generated as optical compensation result data corresponding to the measurement result data of the optical characteristics of the display panel, based on the result of artificial intelligence processing performed based on an artificial intelligence neural network. The artificial intelligence neural network is an affine layer neural network. The affine neural network comprises: an input layer including multiple input nodes corresponding to first preprocessed information; a first intermediate layer including multiple first intermediate nodes corresponding to second preprocessed information; a second intermediate layer including multiple second intermediate nodes corresponding to third preprocessed information; and an output layer including multiple output nodes corresponding to main processing information. The first preprocessing is a pre-optical compensation process for each color coordinate or brightness; the second preprocessing is a pre-optical compensation process for each color using a first brightness value; the third preprocessing is a pre-optical compensation process for each color using a second brightness value that is higher than the first brightness value; and the main processing is a process for obtaining measurement result data of the optical characteristics.

18. The display device according to claim 17, wherein, The predicted and generated optical compensation results include information about the predicted data voltage for each desired target.

19. The display device according to claim 18, wherein, The desired target includes desired frequency band, brightness or color coordinates, and The predicted and generated optical compensation results also include information about the gamma voltage.

20. The display device of claim 17, further comprising a controller, the controller selecting a data voltage corresponding to the current frequency band or current grayscale by referring to a data voltage stored in the memory for each frequency band or grayscale, and providing data corresponding to the selected data voltage to the data driving circuit.

Images