Method Of Compensating Display Device

Abstract

The present invention provides a compensation method for a display device comprising: a step of displaying a first display image of a calibration pattern using a display device including a plurality of display subpixels; a step of generating a first captured image of the calibration pattern by capturing the first display image using a camera including a plurality of sensor subpixels; a step of generating measurement representative position data corresponding to the calibration pattern by measuring the first captured image; a step of generating 3rd-order fitting representative position data by performing 2D 3rd-order fitting on the measurement representative position data; a step of displaying a second display image of a front back display using the display device; a step of generating a second captured image of the front back display by capturing the second display image using the camera; a step of generating a first cropped image by performing a first size adjustment on the second captured image based on the 3rd-order fitting representative position data; a step of calculating a luminance compensation value by comparing the measured luminance of the first cropped image with a target luminance; and a step of generating compensated image data by applying the luminance compensation value to image data.

Description

Technology Field

[0001] The present invention relates to a compensation method for a display device, and more particularly to a compensation method for a display device in which the defect compensation effect is maximized by improved positional accuracy during camera compensation. Background Technology

[0002] In line with the information age, the display field has also developed rapidly, and in response to this, flat panel display devices (FPDs) with advantages of being thin, lightweight, and having low power consumption, such as liquid crystal display devices (LCDs), plasma display panel devices (PDPs), organic light emitting diode display devices (OLEDs), and field emission display devices (FEDs), have been introduced and are rapidly replacing existing cathode ray tubes (CRTs).

[0003] Among these flat panel display devices, organic light-emitting diode (OLED) displays display images by having the light-emitting diode of each subpixel emit light corresponding to the data voltage; however, the light-emitting layer of the LED, which is made of organic material, exhibits brightness variations within the display panel due to material characteristics, and these brightness variations present a problem in that they appear as defects such as stains.

[0004] To solve this problem, a camera compensation method is proposed that captures a display image of a completed display panel with a camera, detects luminance at each location, and modulates the image data to compensate for the luminance deviation at each location.

[0005] In this camera compensation method, since the resolution of multiple display subpixels of the display panel and multiple sensor subpixels of the camera do not match, the captured image of the camera is mapped to correspond to the displayed image of the display panel.

[0006] However, since the accuracy of position data is reduced due to various causes, there is a problem in which the location of defects in the captured image becomes inaccurate, particularly in regions where the modulation transfer function (MTF), which indicates the reproduction accuracy of the camera's captured image relative to the display panel's image, has a relatively high value.

[0007] As a result, not only are defects in the display panel not compensated for, but the defect-free parts of the display panel are distorted, leading to a reduction in the compensation effect and a deterioration in the display quality of the image. The problem to be solved

[0008] The present invention is presented to solve these problems and aims to provide a compensation method for a display device in which errors in the position data are eliminated, continuity is increased, deviation is reduced, and the defect compensation effect is maximized by performing a two-dimensional third-order polynomial fitting on the generated position data.

[0009] Furthermore, another objective of the present invention is to provide a compensation method for a display device in which the alignment of position data and kernels is improved by performing shifting through symmetry operations on generated position data, thereby reducing deviation between captured images and maximizing defect compensation efficiency. means of solving the problem

[0010] To solve the above problem, the present invention provides a compensation method for a display device comprising: a step of displaying a first display image of a calibration pattern using a display device including a plurality of display subpixels; a step of generating a first captured image of the calibration pattern by capturing the first display image using a camera including a plurality of sensor subpixels; a step of generating measured representative position data corresponding to the calibration pattern by measuring the first captured image; a step of generating 3rd-order fitting representative position data by performing 2D 3rd-order fitting on the measured representative position data; a step of displaying a second display image of a front back display using the display device; a step of generating a second captured image of the front back display by capturing the second display image using the camera; a step of generating a first cropped image by performing size adjustment on the second captured image based on the 3rd-order fitting representative position data; a step of calculating a luminance compensation value by comparing the measured luminance of the first cropped image with a target luminance; and a step of generating compensated image data by applying the luminance compensation value to image data.

[0011] Furthermore, the fitting surface (x, y) along the x and y axes of the above 2D cubic fitting can be expressed by the following equation: fitting surface(x, y) = p00 + p10*x + p01*y + p20*x 2 + p11*xy + p02*y 2 + p30*x 3 + p21*x 2 y + p12*xy 2 + p03*y 3

[0012] (p00: 0th-order fitting factor, p10, p01: 1st-order fitting factor, p20, p11, p02: 2nd-order fitting factor, p30, p21, p12, p03: 3rd-order fitting factor)

[0013] In addition, the first display image of the correction pattern may include a plurality of blocks arranged along the x-axis and y-axis, and each of the plurality of blocks may include one light-emitting display subpixel and a plurality of non-light-emitting display subpixels.

[0014] In addition, the compensation method of the display device may further include the step of generating total position data by dividing the distance between the three-dimensional fitting representative position data corresponding to two adjacent light-emitting display subpixels among the plurality of blocks into equal intervals.

[0015] In addition, the compensation method of the display device may further include the step of displaying a third display image of a partial point using the display device; the step of generating a third captured image of the partial point by capturing the third display image using the camera; and the step of calculating an optimal shift value by performing shifting on the third captured image.

[0016] And, the step of calculating the optimal shift value may include: a step of generating a second cropped image by adding a first shift value to the coordinate value of the center of the part point of the third captured image and performing size adjustment on the third captured image; a step of calculating a first comprehensive symmetry value by performing a symmetry operation on the part point of the second cropped image; a step of calculating a plurality of comprehensive symmetry values ​​different from the first comprehensive symmetry value by repeating the size adjustment and the symmetry operation using a plurality of shift values ​​different from the first shift value; and a step of calculating the largest value among the first comprehensive symmetry value and the plurality of comprehensive symmetry values ​​as the optimal shift value.

[0017] Additionally, the part point of the second cropped image includes first to ninth cropped subpixels arranged in 3 rows and 3 columns, and the step of calculating the first comprehensive symmetry value comprises: a step of calculating the average value of the ratio of the minimum luminance value of the first and third cropped subpixels to the maximum luminance value of the first and third cropped subpixels, the ratio of the minimum luminance value of the fourth and sixth cropped subpixels to the maximum luminance value of the fourth and sixth cropped subpixels, and the ratio of the minimum luminance value of the seventh and ninth cropped subpixels to the maximum luminance value of the seventh and ninth cropped subpixels as the left-right symmetry value; The method may include the step of calculating an average value as an upper-lower symmetry value of the ratio of the minimum luminance value of the first and seventh crop subpixels to the maximum luminance value of the first and seventh crop subpixels, the ratio of the minimum luminance value of the second and eighth crop subpixels to the maximum luminance value of the second and eighth crop subpixels, and the ratio of the minimum luminance value of the third and ninth crop subpixels to the maximum luminance value of the third and ninth crop subpixels; and the step of calculating an average value of the left-right symmetry value and the upper-lower symmetry value as the first overall symmetry value.

[0018] In addition, the second cropped image includes red, green, and blue cropped subpixels, and the green cropped subpixel repeats the size adjustment and symmetry operation with the first shift value in the range of -1.0 to +1.0 and the plurality of shift values, and the blue and red cropped subpixels repeat the size adjustment and symmetry operation with the first shift value in the range of -2.0 to +2.0 and the plurality of shift values.

[0019] Meanwhile, the present invention comprises the steps of: displaying a first display image of a calibration pattern using a display device including a plurality of display subpixels; capturing the first display image using a camera including a plurality of sensor subpixels to generate a first captured image of the calibration pattern; measuring the first captured image to generate measured representative position data corresponding to the calibration pattern; displaying a second display image of a partial point using the display device; capturing the second display image using the camera to generate a second captured image of the partial point; performing shifting on the second captured image to calculate an optimal shift value; adding the optimal shift value to the measured representative position data to generate shifting representative position data; displaying a third display image of a front back display using the display device; and capturing the third display image using the camera to generate a third captured image of the front back display. The present invention provides a compensation method for a display device comprising: a step of generating a first cropped image by performing size adjustment on the third captured image based on the above-mentioned shifting representative position data; a step of calculating a brightness compensation value by comparing the measured brightness of the first cropped image with a target brightness; and a step of generating compensated image data by applying the brightness compensation value to image data. Effects of the invention

[0020] The present invention has the effect of removing errors in position data, increasing continuity, reducing deviation, and maximizing defect compensation effects by performing a 2D 3rd-order polynomial fitting on the generated position data.

[0021] Furthermore, the present invention has the effect of reducing deviation between captured images and maximizing defect compensation efficiency by performing shifting through symmetry operations on the generated position data, thereby improving the alignment of position data and kernels. Brief explanation of the drawing

[0022] FIG. 1 is a drawing for explaining a compensation method of a display device according to an embodiment of the present invention. FIG. 2 is a drawing illustrating a captured image and a cropped image of a display device according to an embodiment of the present invention. FIG. 3 is a drawing illustrating a display image and a captured image of a display device according to an embodiment of the present invention. FIG. 4a is a drawing illustrating ideal representative position data and measured representative position data reflecting measurement error of a display device according to an embodiment of the present invention. FIG. 4b is a drawing illustrating the y-axis coordinates and y-axis intervals of the ideal representative position data and the measured representative position data reflecting the measurement error of a display device according to an embodiment of the present invention. FIG. 5a is a drawing illustrating ideal representative position data and primary fitting representative position data of a display device according to an embodiment of the present invention. FIG. 5b is a drawing illustrating the y-axis coordinates and y-axis intervals of the ideal representative position data and the first-order fitting representative position data of a display device according to an embodiment of the present invention. FIG. 6a is a drawing illustrating a shooting system of a display device according to an embodiment of the present invention. FIG. 6b is a drawing illustrating the tilt of a captured image of a display device according to an embodiment of the present invention. FIG. 6c is a drawing illustrating distortion of a captured image of a display device according to an embodiment of the present invention. FIG. 7a is a drawing illustrating ideal representative position data and actual representative position data with distortion reflected of a display device according to an embodiment of the present invention. FIG. 7b is a drawing illustrating the y-axis coordinates and y-axis intervals of the ideal representative position data and the actual representative position data with distortion reflected, of a display device according to an embodiment of the present invention. FIG. 8a is a drawing illustrating measured representative position data and third-order fitting representative position data of a display device according to an embodiment of the present invention. FIG. 8b is a drawing showing the y-axis coordinates and y-axis intervals of the measured representative position data and the third-order fitting representative position data of a display device according to an embodiment of the present invention. FIG. 9a is a diagram illustrating the distribution of x-axis intervals of measured representative position data and third-order fitting representative position data of a display device according to an embodiment of the present invention. FIG. 9b is a drawing showing the x-axis spacing along the first line segment of FIG. 9a. FIG. 9c is a diagram illustrating the distribution of y-axis intervals of measured representative position data and cubic fitting representative position data of a display device according to an embodiment of the present invention. FIG. 9d is a drawing showing the y-axis spacing along the second line segment of FIG. 9c. FIG. 10 is a drawing for explaining the size adjustment of a captured image of a display device according to an embodiment of the present invention. FIG. 11 is a drawing for explaining the shifting of a display device according to an embodiment of the present invention. FIG. 12 is a diagram illustrating the symmetry operation of shifting of a display device according to an embodiment of the present invention. FIGS. 13a and FIGS. 13b are drawings illustrating cropped images on which fitting has been performed according to an embodiment of the present invention. FIG. 14 is a drawing showing a cropped image on which shifting has been performed according to an embodiment of the present invention. FIG. 15 is a diagram showing the distribution of crop subpixels and optimal shift values ​​of a display device according to an embodiment of the present invention. FIGS. 16a and FIGS. 16b are drawings showing the distribution of optimal shift values ​​of different cropped subpixels of a display device according to an embodiment of the present invention. FIG. 17 is a flowchart illustrating a compensation method for a display device according to an embodiment of the present invention. Specific details for implementing the invention

[0023] Hereinafter, a display device and a compensation method according to the present invention will be described with reference to the attached drawings.

[0024] FIG. 1 is a drawing for explaining a compensation method for a display device according to an embodiment of the present invention, and the display device may be an organic light emitting diode display device (OLED display device).

[0025] As illustrated in FIG. 1, in a compensation method for a display device according to an embodiment of the present invention, a camera (170) is placed on the upper part of a display panel (120) of a display device (110), and a display image displayed by the display panel (120) is captured by the camera (170) to generate a captured image including a measured brightness for each position.

[0026] Here, the display image can correspond to the case where all display subpixels of the display panel (120) emit light (front back display).

[0027] Subsequently, the measured luminance at each location in the captured video is compared with the target luminance to calculate the luminance compensation value for each location.

[0028] For example, if the measured luminance is smaller than the target luminance, a positive (+) luminance compensation value can be calculated, and if the measured luminance is larger than the target luminance, a negative (-) luminance compensation value can be calculated.

[0029] Subsequently, the calculated brightness compensation value for each position is stored in the storage unit (154) of the printed circuit board (156) of the display device (110), and the driving unit (150) of the flexible printed circuit (152) of the display device (110) applies the brightness compensation value for each position of the storage unit (154) to the image data to generate compensated image data, and supplies a data voltage according to the compensated image data to the display panel (120).

[0030] For example, the driving unit (150) can generate compensated image data by increasing the grayscale of the image data in response to a positive (+) positional brightness compensation value, and generate compensated image data by decreasing the grayscale of the image data in response to a negative (-) positional brightness compensation value.

[0031] In this way, the brightness deviation of the display device (110) can be minimized and the display quality of the image can be improved through the camera compensation method.

[0032] In this display device, since the resolution of the multiple display subpixels of the display panel (120) and the multiple sensor subpixels of the camera (170) are different, a calibration pattern is captured to generate total position data including a golden position corresponding to the display subpixels, and then the captured image of the camera is mapped to correspond to the display image of the display panel based on the total position data, which is explained with reference to the drawings.

[0033] FIG. 2 is a drawing illustrating a captured image and a cropped image of a display device according to an embodiment of the present invention, and will be explained with reference to FIG. 1.

[0034] As illustrated in FIG. 2, the captured image (SI) of the camera (170) for the display panel (120) of the display device (110) according to an embodiment of the present invention includes a plurality of sensor subpixels (SSP), for example, 6,576 and 4,384 sensor subpixels (SSP) may be arranged along the x-axis and y-axis, respectively.

[0035] Since the resolution of the display panel (120) and the resolution of the camera (170) are different, the size and number of multiple sensor subpixels (SSP) of the captured image (SI) may differ from the size and number of multiple display subpixels (DSP) of the display panel (120). Accordingly, the golden position (e.g., center position) (GP) corresponding to each of the multiple display subpixels (DSP) is not located at the center of each of the multiple sensor subpixels (SSP). That is, the center position of each display subpixel (DSP) does not correspond to the center position of the corresponding sensor subpixel (SSP).

[0036] Afterwards, mapping is performed to generate a cropped image (CI) by cropping the captured image (SI) to correspond to the display panel (120).

[0037] The crop image (CI) includes a number of crop subpixels (CSP), for example, 2,532 and 1,170 crop subpixels (CSP) may be arranged along the x-axis and y-axis, respectively.

[0038] A plurality of crop subpixels (CSP) of a cropped image (CI) may correspond to a plurality of display subpixels (DSP) of a display panel (120). And, a golden position (GP) corresponding to each of the plurality of display subpixels (DSP) is located at the center of each of the plurality of crop subpixels (CSP).

[0039] Therefore, the location of the defect in the display panel (120) can be accurately detected by analyzing the cropped image (CI).

[0040] Mapping of the captured image (SI) is performed based on representative position data generated from the calibration pattern, which will be explained with reference to the drawings.

[0041] FIG. 3 is a drawing illustrating a display image and a captured image of a display device according to an embodiment of the present invention, and will be explained with reference to FIG. 1 and FIG. 2 together.

[0042] As illustrated in FIG. 3, the display image (DI) of the display panel (120) representing the calibration pattern (CP) includes a plurality of blocks (BL) arranged along the x-axis and y-axis, each block (BL) including one light-emitting display subpixel (DSPe) and a plurality of non-light-emitting display subpixels (DSPn).

[0043] For example, each block (BL) may include 12 display subpixels (DSP), one light-emitting display subpixel (DSPe) and 11 non-light-emitting display subpixels (DSPn).

[0044] A camera (170) captures a display panel (120) to generate a captured image (SI) of a calibration pattern (CP), measures the captured image (SI) to detect a measured light-emitting display subpixel (mDSPe) corresponding to a light-emitting display subpixel (DSPe) of each block (BL) as a reference point, and generates representative position data including the reference point, the golden position (GP).

[0045] Then, the distance between the measurement light-emitting display subpixels (mDSPe) of two adjacent blocks (BL) is divided into equal intervals to calculate the golden position (GP) corresponding to all display subpixels (DSP), and the entire position data including all golden positions (GP) is generated.

[0046] For example, the distance between two measurement light-emitting display subpixels (mDSPe) of two adjacent blocks (BL) can be divided into 12 equal parts to calculate two golden positions (GP) corresponding to the two measurement light-emitting display subpixels (mDSPe) and 11 golden positions (GP) between the two measurement light-emitting display subpixels (mDSPe). In this way, one reference point, the golden position (GP) of each block (BL), can be detected from the captured image (SI), and the remaining 11 golden positions (GP) of each block (BL) can be calculated by dividing them into equal intervals for each block (BL).

[0047] In this way, all golden positions (GP) are calculated from multiple blocks (BL) along the x-axis and y-axis, thereby generating all position data of the display panel (120).

[0048] The difference between the ideal representative position data and the measured representative position data can be minimized through fitting, which is explained with reference to the drawing.

[0049] FIG. 4a is a diagram illustrating ideal representative position data and measured representative position data reflecting measurement error of a display device according to an embodiment of the present invention, and FIG. 4b is a diagram illustrating the y-axis coordinates and y-axis intervals of the ideal representative position data and measured representative position data reflecting measurement error of a display device according to an embodiment of the present invention. Additionally, FIG. 5a is a diagram illustrating ideal representative position data and first-order fitting representative position data of a display device according to an embodiment of the present invention, and FIG. 5b is a diagram illustrating the y-axis coordinates and y-axis intervals of ideal representative position data and first-order fitting representative position data of a display device according to an embodiment of the present invention, and will be explained with reference to FIG. 1 to 3 together.

[0050] As illustrated in FIG. 4a, when measuring the image (SI) of the ideal position data corresponding to the ideal coordinates of the light-emitting display subpixel (DSPe) of the calibration pattern (CP), the image (SI) of the measured position data containing the golden position (GP) with coordinates different from the ideal coordinates is generated due to the measurement error.

[0051] As illustrated in FIG. 4b, the light-emitting display subpixels (DSPe) of the first to sixth blocks (BL), which are sequentially arranged along the y-axis in the center of the ideal representative position data, have a linearly increasing y-axis coordinate (y) and a constant y-axis interval (Δy). On the other hand, the measured light-emitting display subpixels (mDSPe) of the first to sixth blocks (BL), which are sequentially arranged along the y-axis in the center of the measured representative position data, have an irregularly increasing y-axis coordinate (y) and an irregular y-axis interval (Δy) that is smaller than, equal to, or larger than a constant value.

[0052] Here, each block (BL) may consist of 12 display subpixels (DSP). Additionally, 11 display subpixels (DSP) may be arranged between the light-emitting display subpixels (DSPe) of two adjacent blocks (BL) and between the measurement light-emitting display subpixels (mDSPe) of the first to sixth blocks (BL). Furthermore, the positions of the 11 display subpixels (DSP) can be calculated by dividing the distance between the light-emitting display subpixels (DSPe) and the distance between the measurement light-emitting display subpixels (mDSPe) by 12.

[0053] Accordingly, in the ideal representative position data, the luminescent display subpixels (DSPe) of each block (BL) are arranged at equal intervals, and the non-luminescent display subpixels (DSPn) of the first to sixth blocks (BL) and their corresponding golden positions (GP) are arranged at equal intervals. On the other hand, in the measured representative position data, the non-luminescent display subpixels (DSPn) of each block (BL) and their corresponding golden positions (GP) are arranged at equal intervals, but the measured luminescent display subpixels (mDSPe) of the first to sixth blocks (BL) are arranged irregularly, and the non-luminescent display subpixels (DSPn) of different blocks (BL) and their corresponding golden positions (GP) are arranged at different intervals.

[0054] Consequently, when mapping is performed based on the measured representative position data and the measured total position data calculated therefrom, the position of the display subpixel (DSP) of the display panel (120) becomes inaccurate, and the defect compensation effect may be reduced.

[0055] As illustrated in FIG. 5a, when the image (SI) of the ideal position data corresponding to the ideal coordinates of the light-emitting display subpixel (DSPe) of the calibration pattern (CP) is measured and first-order fitting is performed, the measurement error is compensated and the image (SI) of the first-order fitting position data containing the golden position (GP) of substantially the same coordinates as the ideal coordinates is generated.

[0056] As illustrated in FIG. 5b, the light-emitting display subpixels (DSPe) of the first to sixth blocks (BL) arranged sequentially along the y-axis in the center of the ideal representative position data have a linearly increasing y-axis coordinate (y) and a constant y-axis interval (Δy), and the light-emitting display subpixels (DSPe) of the first to sixth blocks (BL) arranged sequentially along the y-axis in the center of the first-order fitting representative position data also have a linearly increasing y-axis coordinate (y) and a constant y-axis interval (Δy).

[0057] Here, each block (BL) consists of 12 display subpixels (DSP), and 11 display subpixels (DSP) may be placed between the light-emitting display subpixels (DSPe) of two adjacent blocks (BL), and the positions of the 11 display subpixels (DSP) can be calculated by dividing the distance between the light-emitting display subpixels (DSPe) by 12.

[0058] In addition, the first fitting may be a method of modifying coordinate values ​​so that the x-axis and y-axis coordinates of the measured light-emitting display subpixel (mDSPe) of the measured representative position data become the most similar first-order graph (straight line).

[0059] Accordingly, in the ideal representative position data, the light-emitting display subpixels (DSPe) of each block (BL) are arranged at equal intervals, and the non-light-emitting display subpixels (DSPn) of the first to sixth blocks (BL) and the corresponding golden positions (GP) are arranged at equal intervals. Also, in the first fitting representative position data, the light-emitting display subpixels (DSPe) of each block (BL) are arranged at equal intervals, and the non-light-emitting display subpixels (DSPn) of the first to sixth blocks (BL) and the corresponding golden positions (GP) are arranged at equal intervals.

[0060] As a result, when mapping is performed based on the primary fitting representative position data and the primary fitting overall position data calculated therefrom, the position deviation of the display subpixel (DSP) of the display panel (120) is compensated, and the defect compensation effect can be improved.

[0061] Meanwhile, due to tilt and distortion caused by the shooting system, the ideal position data may differ from the actual position data, which will be explained with reference to the drawings.

[0062] FIG. 6a is a drawing illustrating a shooting system of a display device according to an embodiment of the present invention, FIG. 6b is a drawing illustrating the tilt of a captured image of a display device according to an embodiment of the present invention, and FIG. 6c is a drawing illustrating the distortion of a captured image of a display device according to an embodiment of the present invention, and will be explained with reference to FIG. 1 to 5 together.

[0063] As illustrated in FIGS. 6a and 6b, the shooting system captures the display panel (120) of the display device (110) according to an embodiment of the present invention to generate a captured image (SI). The image is captured with the display panel (120) of the display device (110) and the sensor panel (SP) of the camera (170) positioned so as not to be parallel, thereby generating a captured image (SI) having a tilt.

[0064] For example, a captured image (SI) can be generated in which the horizontal width increases from the top to the bottom along the y-axis.

[0065] As illustrated in FIGS. 6a and 6c, the shooting system includes an optical system that processes light from a display panel (120) and transmits it to a camera (170), and a shooting image (SI) with distortion is generated instead of a shooting image (SI) without distortion by a plurality of lenses (LS) of the optical system.

[0066] For example, in a no-distortion image (SI), the width and height of the center along the x-axis and y-axis are the same as the width and height of the edges, whereas in a barrel distortion image (SI), the width and height of the center along the x-axis and y-axis are larger than the width and height of the edges, and in a pincushion distortion image (SI), the width and height of the center along the x-axis and y-axis may be smaller than the width and height of the edges.

[0067] The difference between the ideal representative location data and the actual representative location data can be minimized through fitting, which will be explained with reference to the drawings.

[0068] FIG. 7a is a diagram illustrating ideal representative position data and actual representative position data reflecting distortion of a display device according to an embodiment of the present invention, and FIG. 7b is a diagram illustrating the y-axis coordinates and y-axis intervals of the ideal representative position data and actual representative position data reflecting distortion of a display device according to an embodiment of the present invention. Additionally, FIG. 8a is a diagram illustrating measured representative position data and third-order fitting representative position data of a display device according to an embodiment of the present invention, and FIG. 8b is a diagram illustrating the y-axis coordinates and y-axis intervals of measured representative position data and third-order fitting representative position data of a display device according to an embodiment of the present invention, and will be explained with reference to FIG. 1 to 6c together.

[0069] As illustrated in FIG. 7a, when capturing a display image (DI) of a calibration pattern (CP), instead of a captured image (SI) of ideal representative position data corresponding to the ideal coordinates of the light-emitting display subpixel (DSPe) of the calibration pattern (CP), a captured image (SI) of real representative position data containing a golden position (GP) of actual coordinates different from the ideal coordinates due to distortion is generated.

[0070] As illustrated in FIG. 7b, the light-emitting display subpixels (DSPe) of the first to sixth blocks (BL), which are sequentially arranged along the y-axis in the center of the ideal representative position data, have a linearly increasing y-axis coordinate (y) and a constant y-axis interval (Δy). On the other hand, the measured light-emitting display subpixels (mDSPe) of the first to sixth blocks (BL), which are sequentially arranged along the y-axis in the center of the actual representative position data, have a y-axis coordinate (y) that decreases and then increases again, and a y-axis interval (Δy) that becomes smaller than a constant value and then increases again.

[0071] Here, each block (BL) may consist of 12 display subpixels (DSP). Additionally, 11 display subpixels (DSP) may be arranged between the light-emitting display subpixels (DSPe) of two adjacent blocks (BL) and between the measurement light-emitting display subpixels (mDSPe) of the first to sixth blocks (BL). Furthermore, the positions of the 11 display subpixels (DSP) can be calculated by dividing the distance between the light-emitting display subpixels (DSPe) and the distance between the measurement light-emitting display subpixels (mDSPe) by 12.

[0072] Accordingly, in ideal representative position data, the luminescent display subpixels (DSPe) of each block (BL) are arranged at equal intervals, so that the non-luminescent display subpixels (DSPn) of the first to sixth blocks (BL) and their corresponding golden positions (GP) can be arranged at equal intervals. On the other hand, in actual representative position data, the non-luminescent display subpixels (DSPn) of each block (BL) and their corresponding golden positions (GP) are arranged at equal intervals, but the luminescent display subpixels (DSPe) of the first to sixth blocks (BL) are arranged at different intervals, so that the non-luminescent display subpixels (DSPn) of different blocks (BL) and their corresponding golden positions (GP) are arranged at different intervals.

[0073] As illustrated in FIG. 8a, a display image (DI) of a calibration pattern (CP) is captured to generate an image (SI) of real position data, and the image (SI) of real position data is measured to generate an image (SI) of measured position data that reflects measurement error and distortion. Then, when third-order fitting is performed on the image (SI) of measured position data, the measurement error is compensated, and an image (SI) of third-order fitting position data is generated that includes a golden position (GP) with coordinates substantially identical to the actual coordinates.

[0074] As illustrated in FIG. 8b, the measured light-emitting display subpixels (mDSPe) of the first to sixth blocks (BL), which are sequentially arranged along the y-axis in the center of the measured representative position data, have a y-axis coordinate (y) in which the slope decreases and then increases again, unlike the ideal representative position data due to measurement error, and a y-axis interval (Δy) in which it becomes smaller than a constant value and then increases again, unlike the ideal representative position data. On the other hand, the light-emitting display subpixels (DSPe) of the first to sixth blocks (BL), which are sequentially arranged along the y-axis in the center of the third-order fitting representative position data, have a y-axis coordinate (y) in which the slope decreases and then increases again, substantially the same as the ideal representative position data, and a y-axis interval (Δy) in which it becomes smaller than a constant value and then increases again, substantially the same as the ideal representative position data.

[0075] Here, each block (BL) may consist of 12 display subpixels (DSP). Additionally, 11 display subpixels (DSP) may be placed between the light-emitting display subpixels (DSPe) of two adjacent blocks (BL). Furthermore, the positions of the 11 display subpixels (DSP) can be calculated by dividing the distance between the light-emitting display subpixels (DSPe) by 12.

[0076] In addition, third-order fitting may be a method of modifying coordinate values ​​so that the x-axis and y-axis coordinates of the measured light-emitting display subpixel (mDSPe) of the measured representative position data become the most similar third-order graph (curve).

[0077] For example, a cubic fitting surface (x, y) for two dimensions of the x-axis and y-axis can be expressed by the following equation.

[0078] fitting surface(x, y) = p00 + p10*x + p01*y + p20*x 2 + p11*xy + p02*y 2 + p30*x 3 + p21*x 2 y + p12*xy 2 + p03*y 3

[0079] (p00: 0th-order fitting factor, p10, p01: 1st-order fitting factor, p20, p11, p02: 2nd-order fitting factor, p30, p21, p12, p03: 3rd-order fitting factor)

[0080] Accordingly, in the measured position data, the light-emitting display subpixels (DSPe) of the first to sixth blocks (BL) and their corresponding golden positions (GP) may be positioned differently from the actual position data with distortion reflected. On the other hand, in the third-order fitting position data, the measurement error is compensated so that the light-emitting display subpixels (DSPe) of the first to sixth blocks (BL) and their corresponding golden positions (GP) are positioned substantially the same as the actual position data with distortion reflected.

[0081] As a result, when mapping is performed based on the two-dimensional cubic fitting representative position data according to the x-axis and y-axis and the two-dimensional cubic fitting overall position data calculated therefrom, the position deviation of the display subpixel (DSP) of the display panel (120) is compensated, and the defect compensation effect can be improved.

[0082] The results of this 2D cubic fitting are explained with reference to the drawings.

[0083] FIG. 9a is a diagram illustrating the distribution of x-axis intervals of measured representative position data and cubic fitting representative position data of a display device according to an embodiment of the present invention, FIG. 9b is a diagram illustrating the x-axis intervals along the first line segment of FIG. 9a, FIG. 9c is a diagram illustrating the distribution of y-axis intervals of measured representative position data and cubic fitting representative position data of a display device according to an embodiment of the present invention, FIG. 9d is a diagram illustrating the y-axis intervals along the second line segment of FIG. 9c, and these will be explained with reference to FIG. 1 to FIG. 8b together.

[0084] As shown in Fig. 9a, the x-axis interval (Δx) of the measured representative position data has an irregular distribution within the captured image (SI), whereas the x-axis interval (Δx) of the third-order fitting representative position data has a uniform distribution within the captured image (SI).

[0085] As illustrated in FIG. 9b, the x-axis interval (Δx) along the first line segment (L1) of the measured representative position data has a small error in the center and a big error in the edge, whereas the x-axis interval (Δx) along the first line segment (L1) of the third-order fitting representative position data has a value that gradually increases without error. In this case, the second and third coefficients of the third-order fitting surface have relatively small values, so the contribution of the second and third terms may be relatively small.

[0086] As shown in Fig. 9c, the y-axis interval (Δy) of the measured representative position data has an irregular distribution within the captured image (SI), whereas the y-axis interval (Δy) of the third-order fitting representative position data has a uniform distribution within the captured image (SI).

[0087] As illustrated in FIG. 9d, the y-axis interval (Δy) along the second line segment (L2) of the measured representative position data has a small error in the center and a big error in the edge, whereas the y-axis interval (Δy) along the second line segment (L2) of the third-order fitting representative position data has a gradually decreasing value without error. In this case, the second and third coefficients of the third-order fitting surface have relatively small values, so the contribution of the second and third terms may be relatively small.

[0088] As described above, in the display device (110) according to the embodiment of the present invention, by performing a two-dimensional third-order polynomial fitting along the x-axis and y-axis on the generated measurement position data, the measurement error is eliminated, the continuity of the position data is increased, the deviation is reduced, and the defect compensation effect is maximized.

[0089] Meanwhile, in the present invention, shifting is performed on a captured image, which is explained with reference to the drawings.

[0090] FIG. 10 is a drawing for explaining the size adjustment of a captured image of a display device according to an embodiment of the present invention, and is explained with reference to FIGS. 1 to 9d together.

[0091] As illustrated in FIG. 10, a display panel (120) displays a display image (DI) having a black dot of a 3x3 size display subpixel (DSP) in the center, and a camera (170) captures the display image (DI) to generate a captured image (SI) having a black dot in the center.

[0092] For convenience, Figure 10 shows an enlarged view of the central part of the captured image (SI).

[0093] Subsequently, the brightness of the sensor subpixel (SSP) of the captured image (SI) having a black spot in the center is measured to generate a discontinuous graph of the intensity of the sensor subpixel (SSP) at different positions, and the golden position (GPi, GPi+1) corresponding to the display subpixel (DSP) can be located between the sensor subpixels (SSP).

[0094] Subsequently, a continuous kernel of positional intensity is generated from a discontinuous graph of positional intensity of the sensor subpixel (SSP) through an interpolation method such as interpolation, and the maximum point of the kernel can be set as the golden position (GP) corresponding to the display subpixel (DSP).

[0095] However, as the modulation transfer function (MTF) increases, there may be cases where the golden position (GP) and the kernel's maximum point do not exactly coincide for each captured image (SI).

[0096] When the golden ratio (GP) and the kernel's maximum point do not coincide, cropping the captured image (SI) to generate a cropped image (CI) results in misalignment where the black spot is not accurately located in the center of the cropped image (CI), thereby reducing the accuracy of defect location detection and decreasing the defect compensation effect.

[0097] In an embodiment of the present invention, shifting is performed on the captured image (SI), and as a result, the golden position (GP) of the display subpixel (DSP) and the crop subpixel (CSP) of the kernel's maximum point can be accurately aligned.

[0098] Furthermore, since the golden ratio (GP) and the kernel's maximum point coincide, cropping the captured image (SI) to generate a cropped image (CI) aligns the black spots within the cropped image (CI) to be accurately positioned in the center, thereby improving the accuracy of defect location detection and maximizing the defect compensation effect.

[0099] Shifting of the captured image (SI) can be performed by repeatedly cropping based on a certain range of shift values, which will be explained with reference to the drawings.

[0100] FIG. 11 is a drawing for explaining shifting of a display device according to an embodiment of the present invention, and FIG. 12 is a drawing for explaining symmetry operation of shifting of a display device according to an embodiment of the present invention, and will be explained with reference to FIG. 1 to FIG. 10 together.

[0101] As illustrated in FIG. 11, a display image (DI) having a black dot of a 3x3 size display subpixel (DSP) in the center is captured to generate a captured image (SI) having a black dot in the center.

[0102] For example, the center of the sunspot in the display image (DI) may have x-axis and y-axis coordinate values ​​of (1356, 614), and the center of the sunspot in the captured image (SI) may have x-axis and y-axis coordinate values ​​of (3325.5, 2203.6).

[0103] A shift value is added to the coordinate value of the center of the sunspot in the captured image (SI), and cropping is performed to generate a cropped image (CI) with a sunspot in the center, and symmetry is calculated by performing a symmetry operation on the sunspot of the cropped image (CI).

[0104] Then, scaling and symmetry operations are iterated using multiple other shift values.

[0105] For example, shift values ​​used for scaling and symmetry operations may be in the range of -2.0 to +2.0.

[0106] And, in iteration 1, a shift value of (-0.5, -0.5) is applied to the center of the sunspot at (3325.5, 2203.6) to perform scaling and symmetry operations (3325.5-0.5, 2203.6-0.5), in iteration 72, a shift value of (+1.0, +1.0) is applied to the center of the sunspot at (3325.5, 2203.6) to perform scaling and symmetry operations, and in iteration 111, a shift value of (+2.0, -0.5) is applied to the center of the sunspot at (3325.5, 2203.6) to perform size It can perform adjustment and symmetry operations.

[0107] As illustrated in FIG. 12, the black spot in the central part of the crop image (CI) may include first to ninth crop subpixels (CSP1 to CSP9) arranged in a 3x3 row. The first, fourth, and seventh crop subpixels (CSP1, CSP4, CSP7) in the left column of the fifth crop subpixel (CSP5) located in the center are compared with the third, sixth, and ninth crop subpixels (CSP3, CSP6, CSP9) in the right column of the fifth crop subpixel (CSP5), respectively, and left-right symmetry is determined from the average value of the comparison results.

[0108] For example, as shown in the equation below, the average value of the ratio of the minimum luminance value (min(CSP1, CSP3)) of the first and third crop subpixels (CSP1, CSP3) to the maximum luminance value (max(CSP1, CSP3)) of the first and third crop subpixels (CSP1, CSP3), the ratio of the minimum luminance value (min(CSP4, CSP6)) of the fourth and sixth crop subpixels (CSP4, CSP6) to the maximum luminance value (max(CSP4, CSP6)) of the fourth and sixth crop subpixels (CSP4, CSP6), and the ratio of the minimum luminance value (min(CSP7, CSP9)) of the seventh and ninth crop subpixels (CSP7, CSP9) to the maximum luminance value (max(CSP7, CSP9)) of the seventh and ninth crop subpixels (CSP7, CSP9) It can be calculated as a symmetry value (LRSV).

[0109] LRSV = AVG[{MIN(CSP1, CSP3) / MAX(CSP1, CSP3)}, {MIN(CSP4, CSP6) / MAX(CSP4, CSP6)}, {MIN(CSP7, CSP9) / MAX(CSP7, CSP9)}]

[0110] Here, it can be determined that the closer the left-right symmetry value (LRSV) is to 1, the more similar the brightness of the crop subpixels (CSP) in the left and right columns is, and the greater the left-right symmetry of the black spots in the cropped image (CI).

[0111] Then, the first, second, and third crop subpixels (CSP1, CSP2, CSP3) in the upper column of the fifth crop subpixel (CSP5) located at the center are compared with the seventh, eighth, and ninth crop subpixels (CSP7, CSP8, CSP9) in the lower column of the fifth crop subpixel (CSP5), respectively, and vertical symmetry is determined from the average value of the comparison results.

[0112] For example, as shown in the equation below, the average value of the ratio of the minimum luminance value (min(CSP1, CSP7)) of the first and seventh crop subpixels (CSP1, CSP7) to the maximum luminance value (max(CSP1, CSP7)) of the first and seventh crop subpixels (CSP1, CSP7), the ratio of the minimum luminance value (min(CSP2, CSP8)) of the second and eighth crop subpixels (CSP2, CSP8) to the maximum luminance value (max(CSP2, CSP8)) of the second and eighth crop subpixels (CSP2, CSP8), and the ratio of the minimum luminance value (min(CSP3, CSP9)) of the third and ninth crop subpixels (CSP3, CSP9) to the maximum luminance value (max(CSP3, CSP9)) of the third and ninth crop subpixels (CSP3, CSP9) is... It can be calculated as a symmetric value (UDSV).

[0113] UDSV = AVG[{MIN(CSP1, CSP7) / MAX(CSP1, CSP7)}, {MIN(CSP2, CSP8) / MAX(CSP2, CSP8)}, {MIN(CSP3, CSP9) / MAX(CSP3, CSP9)}]

[0114] Here, it can be determined that the closer the vertical symmetry value (UDSV) is to 1, the more similar the brightness of the crop subpixels (CSP) in the upper and lower columns is, and the greater the vertical symmetry of the black spots in the cropped image (CI).

[0115] In addition, as shown in the equation below, the average of the left-right symmetry value (LRSV) and the up-down symmetry value (UDSV) of the sunspots in the cropped image (CI) can be calculated as the total symmetry value (TSV).

[0116] TSV = AVG[LRSV, UDSV]

[0117] Here, it can be determined that the closer the total symmetry value (TSV) is to 1, the more similar the brightness of the crop subpixels (CSP) in the top, bottom, left, and right columns is, and the greater the symmetry of the black spots in the cropped image (CI).

[0118] In conclusion, after calculating multiple TSVs by repeating scaling and symmetry operations on multiple shift values, the shift value corresponding to the TSV closest to 1 (largest) can be determined as the optimal shift value.

[0119] For example, if the total symmetry value (TSV) of the 72nd iteration is greater than the total symmetry value (TSV) of the remaining iterations including the 1st and 111th iterations, the cropped image (CI) of the 72nd iteration has greater vertical and horizontal symmetry than the cropped image (CI) of the remaining iterations, and the shift value of (+1.0, +1.0) of the 72nd iteration can be determined as the optimal shift value.

[0120] In an embodiment of the present invention, by performing shifting on the captured image (SI), the golden position (GP) and the maximum point of the kernel are aligned, and the black spot within the cropped image (CI) is aligned to be accurately located in the center, thereby improving the accuracy of defect location detection and maximizing the defect compensation effect.

[0121] The consistency between the display subpixel and the crop subpixel depending on whether fitting and shifting are applied is explained with reference to the drawings.

[0122] FIGS. 13a and FIGS. 13b are drawings illustrating cropped images in which fitting has been performed according to an embodiment of the present invention, and FIGS. 14 is a drawing illustrating a cropped image in which shifting has been performed according to an embodiment of the present invention, and these will be explained with reference to FIGS. 1 to 12 together.

[0123] As shown in FIG. 13a, in the cropped image (CI) of the solid pattern of the comparative example in which shifting is applied and fitting is not applied, the edge of the notch is displayed as a zigzag curve rather than a straight line, causing misalignment with respect to the display panel (120), and in the cropped image (CI) of the 3x3 black dot of the comparative example, the symmetry of the black dot in the central part of the display panel (120) is reduced, causing misalignment with respect to the display panel (120).

[0124] On the other hand, in the cropped image (CI) of the solid pattern of the embodiment of the present invention in which shifting and fitting are applied, the notch edge portion is displayed as a straight line and aligned with the display panel (120), and in the cropped image (CI) of the 3x3 black dot of the embodiment of the present invention, the black dot in the central portion of the display panel (120) has improved symmetry and is aligned with the display panel (120).

[0125] As shown in FIG. 13b, in the cropped image (CI) of the solid pattern of the comparative example in which shifting is applied and fitting is not applied, the right edge of the notch is misaligned by 0.5 cropped subpixels (CSP) with respect to the display panel (120), and the upper right edge is misaligned by 1 cropped subpixel (CSP) with respect to the display panel (120).

[0126] On the other hand, in the cropped image (CI) of the solid pattern of the embodiment of the present invention to which shifting and fitting are applied, the notch edge and the upper right edge are aligned with respect to the display panel (120).

[0127] As shown in FIG. 14, in the cropped image (CI) of the 3x3 grid of Comparative Example 1, in which fitting is applied and shifting is not applied, misalignment occurs in which the width of the high-luminance grid (bright part) in the upper left, center, and upper right sides increases and the width of the low-luminance grid (dark part) decreases, and in the cropped image (CI) of the 3x3 black dot of Comparative Example 1, misalignment occurs in which the low-luminance cropped subpixels (dark part) in the upper left, center, and upper right sides are dispersed.

[0128] In addition, in the cropped image (CI) of the 3x3 grid of Comparative Example 2, which has a fitting applied and no shifting applied with an arbitrary shift value, misalignment occurs in which the width of the high-luminance grid (bright part) in the upper left, center, and upper right sides decreases and the width of the low-luminance grid (dark part) increases, and in the cropped image (CI) of the 3x3 black dot of Comparative Example 2, misalignment occurs in which the low-luminance cropped subpixels (dark part) in the upper left, center, and upper right sides are concentrated in the upper left.

[0129] On the other hand, in the 3x3 grid crop image (CI) of the embodiment of the present invention to which fitting and shifting are applied, the widths of the high-luminance grid (bright part) and low-luminance grid (dark part) in the upper left, center, and upper right sides are aligned to be kept constant, and in the 3x3 black dot crop image (CI) of the embodiment, the low-luminance crop subpixels (dark part) in the upper left, center, and upper right sides are aligned to be concentrated towards the center.

[0130] As described above, in the display device (110) according to the embodiment of the present invention, by performing a two-dimensional third-order polynomial fitting along the x-axis and y-axis on the measured position data, the measurement error is eliminated, the continuity of the position data is increased, the deviation is reduced, and the defect compensation effect is maximized.

[0131] In addition, by performing shifting on the captured image (SI), the golden ratio (GP) and the kernel's maximum point are aligned, and the black spot within the cropped image (CI) is aligned to be accurately located in the center, thereby improving the accuracy of defect location detection and maximizing the defect compensation effect.

[0132] In the embodiments of FIGS. 10 to 14, shifting was performed using a captured image (SI) containing a black dot, but in other embodiments, shifting may be performed using a captured image (SI) containing a white dot.

[0133] Meanwhile, the range of multiple shift values ​​can be set differently for the red, green, and blue crop subpixels (CSP), which will be explained with reference to the drawings.

[0134] FIG. 15 is a diagram showing the distribution of cropped subpixels and optimal shift values ​​of a display device according to an embodiment of the present invention, and FIG. 16a and FIG. 16b are diagrams showing the distribution of optimal shift values ​​of different cropped subpixels of a display device according to an embodiment of the present invention, respectively, and will be explained with reference to FIG. 1 to FIG. 14 together.

[0135] As illustrated in FIG. 15, in the display device (110) of the first case (case 1), the blue (B) crop subpixel (CSP) is separated by xb and yb in the x-axis and y-axis, respectively, with respect to the green (G) crop subpixel (CSP) at the reference position coordinate (Sx, Sy), and the red (R) crop subpixel (CSP) is separated by xr and yr in the x-axis and y-axis, respectively, with respect to the green (G) crop subpixel (CSP).

[0136] Therefore, the blue (B) crop subpixel (CSP) has coordinates (Sx+xb, Sy+yb), where xb is greater than 0 and less than or equal to 2 (0 <xb≤2), yb는 -2보다 크고 0보다 작거나 같을 수 있다(-2<yb≤0).

[0137] The red (R) crop subpixel (CSP) has coordinates (Sx+xr, Sy+yr), where xr is greater than 0 and less than or equal to 2 (0 <xr≤2), yr는 0보다 크고 2보다 작거나 같을 수 있다(0<yr≤2).

[0138] In the display device (110) of the second case (case 2), the blue (B) crop subpixel (CSP) is separated by xb and yb in the x-axis and y-axis, respectively, with respect to the green (G) crop subpixel (CSP) at the reference position coordinate (Sx, Sy), and the red (R) crop subpixel (CSP) is separated by xr and yr in the x-axis and y-axis, respectively, with respect to the green (G) crop subpixel (CSP).

[0139] Therefore, the blue (B) crop subpixel (CSP) has coordinates (Sx+xb, Sy+yb), where xb is greater than or equal to -2 and less than 0 (-2≤xb<0), and yb is greater than or equal to 0 and less than 2 (0≤yb<2).

[0140] The red (R) crop subpixel (CSP) has coordinates (Sx+xr, Sy+yr), where xr is greater than or equal to -2 and less than 0 (-2≤xr<0), and yr is greater than or equal to -2 and less than 0 (-2≤yb<0).

[0141] In the display device (110) of the third case (case 3), the blue (B) crop subpixel (CSP) is separated by xb along the x-axis with respect to the green (G) crop subpixel (CSP) at the reference position coordinate (Sx, Sy), and the red (R) crop subpixel (CSP) is separated by xr along the x-axis with respect to the green (G) crop subpixel (CSP).

[0142] Therefore, the blue (B) crop subpixel (CSP) has coordinates (Sx+xb, Sy), where xb is greater than or equal to 0 and less than 2 (0≤xb<2).

[0143] The red (R) crop subpixel (CSP) has coordinates (Sx+xr, Sy), where xr is greater than or equal to 0 and less than 2 (0≤xr<2).

[0144] In the display device (110) of the fourth case (case 4), the blue (B) crop subpixel (CSP) is separated by yb along the y-axis with respect to the green (G) crop subpixel (CSP) at the reference position coordinate (Sx, Sy), and the red (R) crop subpixel (CSP) is separated by yr along the y-axis with respect to the green (G) crop subpixel (CSP).

[0145] Therefore, the blue (B) crop subpixel (CSP) has coordinates (Sx, Sy+yb), where yb is greater than or equal to -2 and can be less than 0 (-2≤yb<0).

[0146] The red (R) crop subpixel (CSP) has coordinates (Sx, Sy+yr), where yr is greater than or equal to -2 and can be less than 0 (-2≤yr<0).

[0147] That is, in the first to fourth cases (case 1 to case 4), the blue (B) and red (R) crop subpixels (CSP) each have an optimal shift value in the range of -2 to +2 with respect to the green (G) crop subpixel (CSP).

[0148] As illustrated in FIG. 16a, in the first to tenth samples, the green (G) crop subpixel (CSP) has an optimal shift value in the range of about -0.2 to about 0.3, the blue (B) crop subpixel (CSP) has an optimal shift value in the range of about 0.8 to about 1.2, and the red (R) crop subpixel (CSP) has an optimal shift value in the range of about 0.8 to about 1.4.

[0149] As illustrated in FIG. 16b, in the first to tenth samples, the green (G) crop subpixel (CSP) has an optimal shift value in the range of about -0.8 to about 0.2, the blue (B) crop subpixel (CSP) has an optimal shift value in the range of about -1.5 to about -0.4, and the red (R) crop subpixel (CSP) has an optimal shift value in the range of about -0.3 to about 0.7.

[0150] Accordingly, in the display device (110) according to an embodiment of the present invention, in the case of a green (G) crop subpixel (CSP), it is possible to perform an iteration of shifting with a shift value in the range of approximately -1.0 to approximately +1.0 with respect to the reference position coordinate (Sx, Sy). ((Sx, Sy)±1.0 -> (Sx±1.0, Sy±1.0))

[0151] In addition, for the blue (B) and red (R) crop subpixels (CSP), shifting iterations can be performed with shift values ​​ranging from approximately -2.0 to approximately +2.0 with respect to the reference position coordinates (Sx, Sy). ((Sx, Sy)±2.0 -> (Sx±2.0, Sy±2.0))

[0152] Here, the interval of the iteration can be set to about 0.2 or less.

[0153] A method for compensating for such a display device (110) is explained with reference to the drawing.

[0154] FIG. 17 is a flowchart illustrating a compensation method for a display device according to an embodiment of the present invention, which will be explained with reference to FIGS. 1 to 16b.

[0155] As illustrated in FIG. 17, in step 110 (st110), a first display image (DI) of a calibration pattern (CP) is captured by a camera (170) to generate a first captured image (SI). Here, a display panel (120) displays the first display image (DI) of the calibration pattern (CP), and the calibration pattern (CP) may include a plurality of blocks (BL).

[0156] In step 112 (st112), the first captured image is measured to generate first representative measurement position data corresponding to the correction pattern (CP). Here, the first representative measurement position data reflects measurement error and distortion.

[0157] In step 114 (st114), a 2D cubic fitting is performed on the first measurement representative position data to generate cubic fitting representative position data. Here, the measurement error is compensated, and the golden position (GP) of the cubic fitting representative position data can have the same coordinates as the actual coordinates.

[0158] In step 116 (st116), a second display image (DI) of a partial point is captured by a camera (170) to generate a second captured image. Here, the display panel (120) displays the second display image of the partial point, and the partial point may include a black dot of a 3x3 size display subpixel (DSP) in the center.

[0159] In step 118 (st118), the second captured image (SI) is measured to generate second measurement representative position data corresponding to the partial point.

[0160] In step 120 (st120), shifting is performed on the second captured image (SI) to calculate an optimal shift value. Here, multiple TSV values ​​are calculated by repeating scaling and symmetry operations, and the shift value corresponding to the TSV value closest to 1 can be determined as the optimal shift value.

[0161] In step 122 (st122), shifting representative position data is generated by adding an optimal shift value to the second measurement representative position data. Here, the golden position (GP) and the kernel's maximum point coincide, and the partial point can be accurately located in the center of the second captured image.

[0162] In step 124 (st124), a third display image (DI) of a full white display is captured by a camera (170) to generate a third captured image (SI). Here, the display panel (120) can display the third display image (DI) of the full white display.

[0163] In step 126 (st126), the third captured image (SI) is cropped based on the third representative position data and the shifting representative position data to generate the first cropped image (CI).

[0164] In step 128 (st128), the measured luminance of the first cropped image is compared with the target luminance to calculate a luminance compensation value.

[0165] In step 130 (st130), a luminance compensation value is applied to the image data to generate compensated image data.

[0166] In the embodiment of FIG. 17, both the third fitting of step 114 (st114) and the shifting of step 120 (st120) are performed, but in other embodiments, one of the third fitting of step 114 (st114) and the shifting of step 120 (st120) may be omitted.

[0167] In addition, in the embodiment of FIG. 17, both the first display image (DI) of the correction pattern (CP) and the second display image (DI) of the partial dot of the black dot are displayed, but in other embodiments, the display of the second display image (DI) of the partial dot may be omitted. For example, the first display image (DI) of the correction pattern (CP) may include a white dot of a 3x3 size display subpixel (DSP) in the central part, and the first display image (DI) of the white dot may be captured to generate a second captured image (SI).

[0168] Although the present invention has been described above with reference to preferred embodiments, those skilled in the art will understand that various modifications and changes can be made to the invention without departing from the technical spirit and scope of the invention as described in the following claims. Explanation of the symbols

[0169] 110: Display device 120: Display panel DI: Display image SI: Captured image CI: Crop video

Claims

Claim 1 The method comprises the steps of: displaying a first display image of a calibration pattern comprising a plurality of blocks arranged along the x-axis and y-axis using a display device including a plurality of display subpixels; capturing the first display image using a camera including a plurality of sensor subpixels to generate a first captured image of the calibration pattern; measuring the first captured image to generate measurement representative position data corresponding to the calibration pattern; performing a 2D 3D fitting on the measurement representative position data to generate 3D fitting representative position data; displaying a second display image of a front back display using the display device; capturing the second display image using the camera to generate a second captured image of the front back display; performing a first size adjustment on the second captured image based on the 3D fitting representative position data to generate a first cropped image; calculating a luminance compensation value by comparing the measured luminance of the first cropped image with a target luminance; and applying the luminance compensation value to image data to generate compensated image data, wherein the plurality of blocks each Compensation method for a display device comprising one light-emitting display subpixel and a plurality of non-light-emitting display subpixels. Claim 2 In claim 1, the fitting surface (x, y) along the x-axis and y-axis of the 2D cubic fitting is a compensation method for a display device expressed by the following equation: fitting surface(x, y) = p00 + p10*x + p01*y + p20*x 2 + p11*xy + p02*y 2 + p30*x 3 + p21*x 2 y + p12*xy 2 + p03*y 3 (p00: 0th-order fitting factor, p10, p01: 1st-order fitting factor, p20, p11, p02: 2nd-order fitting factor, p30, p21, p12, p03: 3rd-order fitting factor) Claim 3 delete Claim 4 A compensation method for a display device according to claim 1, further comprising the step of generating total position data by dividing the distance between the three-dimensional fitting representative position data corresponding to two adjacent light-emitting display subpixels among the plurality of blocks into equal intervals. Claim 5 A compensation method for a display device according to claim 1, further comprising: a step of displaying a third display image of a partial point of a 3x3 size display subpixel using the display device; a step of generating a third captured image of the partial point by capturing the third display image using the camera; and a step of calculating an optimal shift value by performing shifting on the third captured image. Claim 6 In claim 5, the step of calculating the optimal shift value comprises: a step of generating a second cropped image by adding a first shift value to the coordinate value of the center of the part point of the third captured image and performing a second size adjustment on the third captured image; a step of calculating a first comprehensive symmetry value corresponding to the up-down-left-right symmetry of the part point by performing a symmetry operation on the part point of the second cropped image; a step of calculating a plurality of comprehensive symmetry values ​​different from the first comprehensive symmetry value by repeating the second size adjustment and the symmetry operation using a plurality of shift values ​​different from the first shift value; and a step of calculating the largest value among the first comprehensive symmetry value and the plurality of comprehensive symmetry values ​​as the optimal shift value. Claim 7 In claim 6, the part point of the second cropped image includes first to ninth cropped subpixels arranged in 3 rows and 3 columns, and the step of calculating the first comprehensive symmetry value comprises: a step of calculating the average value of the ratio of the minimum luminance value of the first and third cropped subpixels to the maximum luminance value of the first and third cropped subpixels, the ratio of the minimum luminance value of the fourth and sixth cropped subpixels to the maximum luminance value of the fourth and sixth cropped subpixels, and the ratio of the minimum luminance value of the seventh and ninth cropped subpixels to the maximum luminance value of the seventh and ninth cropped subpixels as the left-right symmetry value; and the ratio of the minimum luminance value of the first and seventh cropped subpixels to the maximum luminance value of the first and seventh cropped subpixels, the ratio of the minimum luminance value of the second and eighth cropped subpixels to the maximum luminance value of the second and eighth cropped subpixels, and the third A compensation method for a display device comprising the steps of: calculating an average value of the ratio of the minimum luminance values ​​of the third and ninth cropped subpixels to the maximum luminance value of the ninth cropped subpixel as an upper and lower symmetrical value; and calculating an average value of the left and right symmetrical value and the upper and lower symmetrical value as a first comprehensive symmetrical value. Claim 8 A compensation method for a display device according to claim 6, wherein the second cropped image includes red, green, and blue cropped subpixels, the green cropped subpixel repeats the second size adjustment and the symmetry operation with the first shift value in the range of -1.0 to +1.0 and the plurality of shift values, and the blue and red cropped subpixels repeat the second size adjustment and the symmetry operation with the first shift value in the range of -2.0 to +2.0 and the plurality of shift values. Claim 9 A step of displaying a first display image of a calibration pattern comprising a plurality of blocks arranged along the x-axis and y-axis using a display device including a plurality of display subpixels; a step of generating a first captured image of the calibration pattern by capturing the first display image using a camera including a plurality of sensor subpixels; a step of generating first representative measurement position data corresponding to the calibration pattern by measuring the first captured image; a step of generating representative fitting position data by performing 2D cubic fitting on the first representative measurement position data; a step of displaying a second display image of a partial point using the display device; a step of generating a second captured image of the partial point by capturing the second display image using the camera; a step of generating second representative measurement position data by measuring the second captured image; a step of calculating an optimal shift value by performing shifting on the second captured image; a step of generating representative shifting position data by adding the optimal shift value to the second representative measurement position data; and a step of a front back display using the display device A compensation method for a display device comprising: a step of displaying a third display image; a step of generating a third captured image of the front back display by capturing the third display image using the camera; a step of generating a first cropped image by performing size adjustment on the third captured image based on the third fitting representative position data and the shifting representative position data; a step of calculating a luminance compensation value by comparing the measured luminance of the first cropped image with a target luminance; and a step of generating compensated image data by applying the luminance compensation value to image data, wherein each of the plurality of blocks includes one light-emitting display subpixel and a plurality of non-light-emitting display subpixels.

Images