A method for detecting brightness uniformity of a display

Abstract

The present invention relates to a method for detecting brightness uniformity of a display. The method comprises the following steps: the display displays a sequence of images in a sub-pixel lighting mode of every Gh columns and every Gv rows, and takes Gh×Gv pictures; takes positioning pictures when four vertex sub-pixel points of the display are lit, and calculates distances dh and dv between the centroids of light spots of adjacent sub-pixel points in the horizontal and vertical directions on the display according to the centroid coordinates of light spots corresponding to the four vertex sub-pixel points and the number of columns and rows of sub-pixel points on the display; calculates the centroid coordinates of light spots corresponding to each sub-pixel point in each sequence of images according to Gh, Gv, dh, and dv to obtain a coordinate matrix, and extracts the brightness integral value of each light spot to obtain a brightness matrix of the centroid of light spots in each picture; integrates the Gh×Gv brightness matrices into a point-by-point brightness matrix Lw of the display, and calculates the brightness uniformity of the display according to the point-by-point brightness matrix Lw of the display; the brightness value of each sub-pixel point of the display can be accurately extracted, thereby accurately calculating the brightness uniformity of the display.

Description

Technical Field

[0001] The present invention belongs to the technical field of display detection, and in particular relates to a method for detecting brightness uniformity of a display. Background Art

[0002] Panel manufacturing involves complex layering processes. These processes can lead to inconsistent output in terms of brightness (luminance) and color uniformity (chromaticity). Brightness uniformity refers to completely uniform brightness across all parts of the display, ensuring no dark spots or unsightly noise. Most uniformity issues occur near the center of the display, with extreme errors occurring at the outer edges of most panels. This is partly due to the integration of backlights into LCD panels, which makes it more difficult to produce a perfectly uniform LCD display. Furthermore, these issues can arise from manufacturing defects such as uneven thickness of the LCD components, issues with the optical backlight and color filters, or uneven backlight illumination. Therefore, uneven brightness and chromaticity can directly affect the display's gamma levels and grayscale. The current method for testing brightness uniformity for LCD displays uses the method described in Section 5.3 of SJ / T11348-2015, "Measurement Methods for Display Performance of Flat-Panel Televisions." This method uses a luminance meter to measure nine brightness values at test points and calculates brightness uniformity using a formula. This method requires repeated movement and adjustment of the measuring device, and can only obtain brightness values at limited measurement points. It is one-sided and cannot obtain and calculate the uniformity of the display point by point. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method for detecting the brightness uniformity of a display, which can accurately extract the brightness value of each sub-pixel of the display, thereby accurately calculating the brightness uniformity of the display.

[0004] In order to solve the above technical problems, the display brightness uniformity detection method of the present invention is as follows:

[0005] Perform vignetting correction on the camera; display a sequence of images in a sub-pixel lighting mode of every Gh columns and every Gv rows and use the camera to take Gh×Gv pictures, generally 2≤Gh≤16, 2≤Gv≤16; take a positioning picture with the four vertex sub-pixels of the display lit and all other sub-pixels turned off, and calculate the coordinates of the centroid of the light spots corresponding to the four vertex sub-pixels; calculate the distance dh between the centroids of the light spots of adjacent horizontal sub-pixels and the distance dv between the centroids of the light spots of adjacent vertical sub-pixels on the display based on the coordinates of the centroid of the light spots corresponding to the four vertex sub-pixels and the number of columns and rows of sub-pixels on the display; calculate the coordinates of the centroid of the light spots corresponding to each sub-pixel in each sequence of images based on Gh, Gv, dh, and dv to obtain a coordinate matrix and extract the integral value of the brightness of each light spot to obtain the brightness matrix of the centroid of the light spot in each image; integrate the Gh×Gv brightness matrices into a point-by-point brightness matrix Lw of the display, and calculate the brightness uniformity of the display based on the point-by-point brightness matrix Lw of the display.

[0006] The display displays sequential images according to the lighting mode of the pth column, (p+Gh)th column, (p+2Gh)th column, ... sub-pixels and the qth row, (q+Gv)th column, (q+2Gv)th column, ... sub-pixels, where p = 1, 2...Gh, q = 1, 2...Gv.

[0007] Assume that the camera resolution is w×h, the display resolution is m×n, and the imaging diameter of each sub-pixel is at least K camera pixels; according to Calculate the display mode that can accurately obtain the brightness value of each sub-pixel.

[0008] Set the coordinate matrix of the centroid of the light spot corresponding to the four vertex sub-pixel points of the display in the image to P' (i，j) ;

[0009]

[0010] where (p'x (0,0) ,p'y (0,0) )、(p'x (0,m-1) ,p'y (0,m-1) )、(p'x (n-1,0) ,p'y (n-1,0) )、(p'x (n-1,m-1) ,p'y (n-1,m-1) ) are the coordinates of the centroid of the light spot corresponding to the four vertex sub-pixel points of the display, namely the upper left, upper right, lower left and lower right; then the distance between the centroids of the light spots corresponding to the adjacent sub-pixel points on the display is dh=(p'x (0,m-1) -p'x (0,0) ) / (m-1), the distance between the center of mass of the adjacent sub-pixel points in the longitudinal direction is dv=(p'y (n-1,0) -p'y (0,0) ) / (n-1).

[0011] Assume that the display displays the sequential image S1 in the lighting mode of the 1st column, (1+Gh)th column, (1+2Gh)th column, ... sub-pixels and the 1st row, (1+Gv)th column, (1+2Gv)th column, ... sub-pixels; then the abscissa px of the centroid of the light spot corresponding to the upper left sub-pixel is S1-(0,0) =p'x (0,0) , vertical coordinate py S1-(0,0) =p'y (0,0) The centroid coordinates p of the light spot corresponding to the i-th horizontal and j-th vertical illuminated sub-pixel point in the sequence image S1 are calculated according to the following formula: S1-(i,j) =(px S1-(i,j) ,py S1-(i,j) ), i=0,1,2,…, j=0,1,2,…;

[0012]

[0013]

[0014] Where, Offset-x=p'x (n-1,0) -p'x (0,0) , is the lateral offset of the display imaging tilt in the captured image sequence, and the lateral offset averagely distributed to each row is Offset_y=p'y (0,m-1) -p'y (0,0) is the longitudinal offset of the tilted display imaging in the captured image sequence, and the lateral offset averagely distributed to each row is

[0015] For any sequence image St, assume that the sub-pixels in the pth column, (p+Gh)th column, (p+2Gh)th column, ... and the sub-pixels in the qth row, (q+Gv)th column, (q+2Gv)th column, ... of the display are lit; calculate the centroid coordinate p of the light spot corresponding to the sub-pixel in the i-th row and j-th column of the sequence image St according to the following formula: St-(i,j) :

[0016]

[0017] Δx=dh,Δy=dv.

[0018] The brightness uniformity D(Lw) of the display is calculated according to the following formula:

[0019]

[0020]

[0021] Wherein Lws represents the brightness value of the s-th sub-pixel in the point-by-point brightness matrix, s=0 to (m×n-1).

[0022] The maximum brightness uniformity among the red, green, and blue sub-pixels is selected to represent the brightness uniformity of the display.

[0023] The present invention uses an array camera (CMOS or CCD) to shoot a specific sequence of images displayed on the display, and cooperates with a post-image processing algorithm to accurately extract the brightness value of each sub-pixel point on the display, thereby accurately calculating the brightness uniformity of the display. The brightness value of each sub-pixel point in the entire display can be obtained by using a specific sequence image combination method, thereby more accurately evaluating the uniformity of the display. The uniformity calculated by the method of the present invention can represent the state of each pixel on the entire screen, which is more representative than the 9-point method of measuring with a luminance meter, and a fixed image sequence can be shot with an array camera, saving time and labor costs. Compared with patent ZL201010613815.9LED "Display Screen Pixel Lighting Chromaticity Information Collection Method", this method does not require the spacing between each sub-pixel light spot to be large enough during measurement, and only needs to ensure that the light spots do not intersect. The method of the present invention displays fewer image sequences and has a short shooting and processing time. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 Flowchart of the present invention.

[0025] Figure 2 Schematic diagram of imaging when two pixel spots are separated by a certain distance.

[0026] Figure 3 Schematic diagram of imaging when two pixel spots interact with each other.

[0027] Figure 4 Schematic diagram of imaging when two pixel spots are aligned to be tangent.

[0028] Figure 5 Schematic diagram of lighting sub-pixel points in sequence image S1 in 2×2 acquisition mode.

[0029] Figure 6 Schematic diagram of the tilt of sequence image S1 in 2×2 acquisition mode.

[0030] Figure 7 Schematic diagram of lighting sub-pixel points in sequence image S2 in 2×2 acquisition mode.

[0031] Figure 8 Schematic diagram of lighting sub-pixel points in sequence image S3 in 2×2 acquisition mode.

[0032] Figure 9 Schematic diagram of sub-pixel lighting in sequence image S4 in 2×2 acquisition mode.

[0033] Figure 10 Schematic diagram of the sub-pixel points of the sequence image S1 illuminated in 5×4 acquisition mode.

[0034] Figure 11 Schematic diagram of lighting sub-pixel points in sequence image S2 in 5×4 acquisition mode.

[0035] Figure 12 Schematic diagram of lighting sub-pixel points in sequence image S3 in 5×4 acquisition mode.

[0036] Figure 13 Schematic diagram of lighting sub-pixel points in sequence image S20 in 5×4 acquisition mode. DETAILED DESCRIPTION

[0037] Example 1

[0038] 2K display (resolution 1920*1080), 60MP camera measures its uniformity. (9344*7000)

[0039] like Figure 1 As shown, the display brightness uniformity detection method of the present invention is specifically as follows:

[0040] Step 1: Correct the camera's vignetting

[0041] Using an integrating sphere or a uniform white board, the array camera is corrected for vignetting. This correction is a common technique in the industry and will not be detailed here. The resulting camera eliminates the problem of bright center and dark edges caused by the camera lens.

[0042] Step 2: Calculate the display mode that just meets the imaging radius requirements

[0043] The camera resolution is w × h = 9344 × 7000 (number of camera pixel columns × number of camera pixel rows), and 90% of its horizontal and vertical area is used as the effective area for imaging. The display resolution is m × n = 1920 × 1080 (number of display subpixel columns × number of display subpixel rows). The imaging diameter of each subpixel (one of R, G, or B colors) is at least K = 8 camera pixels. If the display uses every Gh columns and every Gv rows, the display mode that can accurately obtain the brightness value of each subpixel is:

[0044] Step 3: The display is as follows Figure 5-Figure 8The mode displays a sequence of images S1 to S4 with Gh columns and Gv rows, and uses a camera to capture four images. In each capture, one sub-pixel is lit for every two sub-pixels horizontally, and the rest are off, with a brightness of 0. In the vertical direction, one sub-pixel is lit for every two sub-pixels, and the rest are off, with a brightness of 0. For any sequence of images Si, each lit sub-pixel on the display corresponds to a light spot on the image. Let the coordinate matrix of the center of mass of the light spot on the image be P Si , then the coordinates of the centroid of the light spot corresponding to the lit sub-pixel point in the i-th row and j-th column of the sequence image are p Si-(i,j) =(px Si-(i,j) ,py Si-(i,j) ), i=0,1,2,...,539, j=0,1,2,...,959. p Si-(0,0) =(px Si-(0,0) ,py Si-(0,0) ) is the centroid coordinate of the light spot corresponding to the first sub-pixel point in the upper left corner of the display. This grouping and time-sharing display principle is discussed in detail in patent ZL201010613815.9LED "Display Pixel Brightness Chromaticity Information Collection Method" and will not be repeated here. Here, it is required that the outermost diameter of each imaging light spot can be tangent to the Figure 4 As shown in , there is no requirement for a black interval in the middle, but crosstalk is not allowed. Figure 3 shown.

[0045] Step 4. Since some adjacent sub-pixels in the sequence image are lit at the same time, the two corresponding light spots in the image may be close to being tangent, resulting in cross-light between the black intervals, or partial overlap between the two light spots. The light spot centroid positioning algorithm corresponding to the sub-pixel points cannot correctly segment the light spots corresponding to all sub-pixel points, which will cause the coordinates corresponding to the centroids of some sub-pixel points to be unrecognizable. Therefore, it is necessary to take a positioning picture with the four vertex sub-pixels of the display lit and all other sub-pixels turned off, to calculate the coordinates of the centroid of the light spots corresponding to each sub-pixel point. The light spot corresponding to each sub-pixel point in the positioning picture can be clearly distinguished and accurately positioned. The centroid positioning algorithm is used to calculate the light spot centroid coordinate matrix P' corresponding to the four vertex sub-pixels. (i，j) .in

[0046] where (p'x (0,0) ,p'y (0,0) )、(p'x (0,m-1) ,p'y (0,m-1) )、(p'x (n-1,0) ,p'y (n-1,0) )、(p'x (n-1,m-1) ,p'y (n-1,m-1)) are the centroid coordinates of the light spots corresponding to the four vertex sub-pixel points of the upper left, upper right, lower left, and lower right of the display;

[0047] Step 5: Based on the coordinates of the centroid of each light spot in each of the four images taken in step 3, the number of columns m and rows n of the display sub-pixels, Gh, and Gv, calculate the coordinates of the centroid of each light spot in each of the four images taken in step 3. The brightness of each light spot is extracted based on the coordinates of the centroid of the light spot. The brightness of each light spot is the integral value of the brightness of 8×8 camera pixels, resulting in the image light spot brightness matrix corresponding to the Gh×Gv sequence of images. The Gh×Gv image light spot brightness matrices are integrated into the display brightness matrix L. The calculation method is as follows:

[0048] a. Since each sub-pixel on the display is arranged at equal intervals, Gh, Gv, m, n and the centroid coordinate matrix p' corresponding to the four vertex sub-pixels (i,j) It is known that the first sequence image S1 captured in step 3 has p S1-(0,0) =p' (0,0) The distance between the center of mass of the light spot of adjacent sub-pixel points on the display is dh=(p'x (0,m-1) -p'x (0,0) ) / (m-1), the distance between the center of mass of the adjacent sub-pixel points in the longitudinal direction is dv=(p'y (n-1,0) -p'y (0,0) ) / (n-1); The centroid coordinates p of the light spot corresponding to each sub-pixel point in the sequence image S1 can be obtained by linear calculation S1-(i,j) =(px S1-(i,j) ,py S1-(i,j) ):

[0049]

[0050] For example, Figure 5 middle:

[0051]

[0052] Where Offset-x=p'x (n-1,0) -p'x (0,0) is the lateral offset of the tilted display image in the captured image sequence. The lateral offset averagely distributed to each row is Offset_y=p'y (0,m-1) -p'y (0,0) is the longitudinal offset of the tilted display imaging in the captured image sequence, and the lateral offset averagely distributed to each row is

[0053] According to formula (1), the value of any point p in the sequence image S1 can be calculated: S1-(i,j) The exact coordinates of .

[0054] b. By using formula (1) and the sub-pixel lighting relationship of the sequence images S1-S4, the centroid coordinates p of the light spots corresponding to the sub-pixel points in the sequence images S2-S4 can be calculated. S2-(i,j) 、p S3-(i,j) 、p S4-(i,j) :

[0055]

[0056]

[0057]

[0058] Where (Δx=dh, Δy=dv).

[0059] c. According to the coordinate matrix P of each sequence image Si Si , extract the corresponding spot brightness integral value in the image, and thus calculate the corresponding brightness matrix Li; the four brightness matrices L1-L4 are integrated into the display point-by-point brightness matrix Lw.

[0060] Step 6: Calculate the display uniformity.

[0061] Calculate the corresponding brightness uniformity D(Lw) based on the display's point-by-point brightness matrix Lw. Brightness uniformity can be calculated using the mean square error formula. This can be calculated for all sub-pixels across the entire screen, or for all sub-pixels within a specific area of interest. Lws represents the brightness value of the sth sub-pixel in the point-by-point brightness matrix, s = 0 to (m × n - 1).

[0062]

[0063]

[0064] The calculation process for brightness uniformity of R, G, and B sub-pixels is the same. The maximum brightness uniformity among the red, green, and blue sub-pixels is selected to represent the brightness uniformity of the display.

[0065] Example 2

[0066] 2K display (resolution 1920*1080), 10MP resolution camera for uniformity measurement (3840*2748)

[0067] like Figure 1 As shown, the display brightness uniformity detection method of the present invention is specifically as follows:

[0068] Step 1: Correct the camera's vignetting

[0069] Using an integrating sphere or a uniform white board, the array camera is corrected for vignetting. This correction is a common technique in the industry and will not be detailed here. The resulting camera eliminates the problem of bright center and dark edges caused by the camera lens.

[0070] Step 2: Calculate the display mode that just meets the imaging radius requirements

[0071] The camera resolution is w×h=3840×2748 (number of camera pixel columns × number of camera pixel rows), and 90% of its horizontal and vertical area is used for shooting. The display resolution is m×n=1920*1080 (number of display sub-pixel columns × number of display sub-pixel rows), and the imaging diameter K of each sub-pixel (one of R, G, and B colors) is at least 8 camera pixels. The display mode that can accurately obtain the brightness value of each sub-pixel is: horizontal acquisition mode Vertical acquisition mode

[0072] Step 3: The display shows the sequence images S1-S20 in a pattern of every 5 columns and every 4 rows, and the camera takes 20 pictures. For each picture, one sub-pixel is lit for every 5 sub-pixels horizontally, and the rest are off, with a brightness of 0. One sub-pixel is lit for every 4 sub-pixels vertically, and the rest are off, with a brightness of 0. For any sequence image Si, each lit sub-pixel on the display corresponds to a light spot on the image. Let the coordinate matrix of the center of mass of the light spot on the image be P Si , then the coordinates of the centroid of the light spot corresponding to the lit sub-pixel point in the i-th row and j-th column of the sequence image are p Si-(i,j) =(px Si-(i,j) ,py Si-(i,j) ), i=0,1,2,...,269, j=0,1,2,...,383. p Si-(0,0) =(px Si-(0,0) ,py Si-(0,0) ) is the centroid coordinate of the light spot corresponding to the first sub-pixel point in the upper left corner of the display. This grouping and time-sharing display principle is discussed in detail in patent ZL201010613815.9LED "Display Pixel Brightness Chromaticity Information Collection Method" and will not be repeated here. Here, it is required that the outermost diameter of each imaging light spot can be tangent to the Figure 4 As shown in , there is no requirement for a black interval in the middle, but crosstalk is not allowed. Figure 3 shown.

[0073] Step 4. Since some adjacent sub-pixels in the sequence image are lit at the same time, the two corresponding light spots in the image may be close to being tangent, resulting in cross-light between the black intervals, or partial overlap between the two light spots. The light spot centroid positioning algorithm corresponding to the sub-pixel points cannot correctly segment the light spots corresponding to all sub-pixel points, which will cause the coordinates corresponding to the centroids of some sub-pixel points to be unrecognizable. Therefore, it is necessary to take a positioning picture with the four vertex sub-pixels of the display lit and all other sub-pixels turned off, to calculate the coordinates of the centroid of the light spots corresponding to each sub-pixel point. The light spot corresponding to each sub-pixel point in the positioning picture can be clearly distinguished and accurately positioned. The centroid positioning algorithm is used to calculate the light spot centroid coordinate matrix P' corresponding to the four vertex sub-pixels. (i，j) .in

[0074] where (p'x (0,0) ,p'y (0,0) )、(p'x (0,m-1) ,p'y (0,m-1) )、(p'x (n-1,0) ,p'y (n-1,0) )、(p'x (n-1,m-1) ,p'y (n-1,m-1) ) are the centroid coordinates of the light spots corresponding to the sub-pixel points at the upper left, upper right, lower left, and lower right vertices of the display.

[0075] Step 5: Based on the coordinates of the centroid of the light spot corresponding to the four vertex sub-pixels and the number of columns (m) and rows (n) of the display sub-pixels, calculate the coordinates of the centroid of each light spot in each of the 20 images taken in step 3. The brightness of each light spot is extracted based on the coordinates of the centroid of the light spot. The brightness of each light spot is the integral value of the brightness of 8×8 camera pixels, resulting in the image light spot brightness matrix corresponding to the Gh×Gv sequence of images. The Gh×Gv image light spot brightness matrices are integrated into the display brightness matrix L. The calculation method is as follows:

[0076] a. Since each sub-pixel on the display is arranged at equal intervals, Gh, Gv, m, n and the centroid coordinate matrix p' corresponding to the four vertex sub-pixels (i,j) It is known that the first sequence image S1 captured in step 3 has p S1-(0,0) =p' 0(,0 The distance between the center of mass of the light spot of adjacent sub-pixel points on the display is dh=(p'x (0,m-1) -p'x (0,0) ) / (m-1), the distance between the center of mass of the adjacent sub-pixel points in the longitudinal direction is dv=(p'y (n-1,0) -p'y (0,0) ) / (n-1); The centroid coordinates p of the light spot corresponding to each sub-pixel point in the sequence image S1 can be obtained by linear calculation S1-(i,j) =(pxS1-(i,j) ,py S1-(i,j) ):

[0077]

[0078] For example, Figure 9 middle:

[0079]

[0080] Where Offset_x = p'x (n-1,0) -p'x (0,0) is the lateral offset of the tilted display image in the captured image sequence. The lateral offset averagely distributed to each row is Offset_y=p'y (0,m-1) -p'y (0,0) is the longitudinal offset of the tilted display imaging in the captured image sequence, and the lateral offset averagely distributed to each row is

[0081] According to formula (1), the value of any point p in the sequence image S1 can be calculated: S1-(i,j) The exact coordinates of .

[0082] b. By using formula (1) and the sub-pixel lighting relationship of the sequence images S1-S4, the centroid coordinates p of the light spots corresponding to the sub-pixel points in the sequence images S2-S4 can be calculated. S2-(i,j) 、p S3-(i,j) 、…p St-(i,j) …、p S20-(i,j) :

[0083] For any sequence image St, suppose the sub-pixels in the pth column, (p+Gh)th column, (p+2Gh)th column, ... and the sub-pixels in the qth row, (q+Gv)th column, (q+2Gv)th column, ... of the display are lit; then the centroid coordinates p of the sub-pixel spot in the i-th row and j-th column of the sequence image St are St-(i,j) The calculation formula is as follows:

[0084]

[0085] but:

[0086]

[0087]

[0088] …

[0089]

[0090] Where (Δx=dh, Δy=dv).

[0091] c. According to the coordinate matrix P of each sequence image Si Si , extract the corresponding spot brightness integral value in the image, and thus calculate the corresponding brightness matrix Li; the 20 brightness matrices L1-L20 are integrated into the display point-by-point brightness matrix Lw.

[0092] Step 6: Calculate the display uniformity.

[0093] Calculate the corresponding brightness uniformity D(Lw) based on the display's point-by-point brightness matrix Lw. Brightness uniformity can be calculated using the mean square error formula. This can be calculated for all sub-pixels across the entire screen, or for all sub-pixels within a specific area of interest. Lws represents the brightness value of the sth sub-pixel in the point-by-point brightness matrix, s = 0 to (m × n - 1).

[0094]

[0095]

[0096] The calculation process for brightness uniformity of R, G, and B sub-pixels is the same. The maximum brightness uniformity among the red, green, and blue sub-pixels is selected to represent the brightness uniformity of the display.

Claims

1. A method for detecting brightness uniformity of a display, characterized in that The method is as follows: performing vignetting correction on a camera; displaying a sequence of images in a sub-pixel lighting mode of every Gh columns and every Gv rows and using a camera to shoot Gh×Gv images, 2≤Gh≤16, 2≤Gv≤16, and a sub-pixel refers to a pixel of one color among R, G, and B; shooting a positioning image with four vertex sub-pixels of the display lit and all other sub-pixels turned off, and calculating the coordinates of the centroid of the light spots corresponding to the four vertex sub-pixels; calculating the distance dh between the centroids of the light spots of adjacent horizontal sub-pixels and the distance dv between the centroids of the light spots of adjacent vertical sub-pixels on the display according to the coordinates of the centroid of the light spots corresponding to the four vertex sub-pixels and the number of columns and rows of sub-pixels on the display; calculating the coordinates of the centroid of the light spots corresponding to each sub-pixel in each sequence of images according to Gh, Gv, dh, and dv to obtain a coordinate matrix, and extracting the integral value of the brightness of each light spot to obtain the brightness matrix of the centroid of the light spot in each image; The Gh×Gv brightness matrices are integrated into a display point-by-point brightness matrix Lw, and the brightness uniformity of the display is calculated based on the display point-by-point brightness matrix Lw.

2. The method for detecting brightness uniformity of a display according to claim 1, wherein The display displays sequential images according to the lighting mode of the pth column, (p+Gh)th column, (p+2Gh)th column, ... sub-pixels and the qth row, (q+Gv)th column, (q+2Gv)th column, ... sub-pixels, where p = 1, 2...Gh, q = 1, 2...Gv.

3. The method for detecting brightness uniformity of a display according to claim 1, wherein The camera resolution is w×h, the display resolution is m×n, and the imaging diameter of each sub-pixel is at least K camera pixels; according to Calculate the display mode that can accurately obtain the brightness value of each sub-pixel.

4. The method for detecting brightness uniformity of a display according to claim 1, wherein Set the coordinate matrix of the centroid of the light spot corresponding to the four vertex sub-pixel points of the display in the image to P' (i，j) ; where (p'x (0,0) ,p'y (0,0) )、(p'x (0,m-1) ,p'y (0,m-1) )、(p'x (n-1,0) ,p'y (n-1,0) )、(p'x (n-1,m-1) ,p'y (n-1,m-1) ) are the coordinates of the centroid of the light spot corresponding to the four vertex sub-pixel points of the display, namely the upper left, upper right, lower left and lower right; then the distance between the centroids of the light spots corresponding to the adjacent sub-pixel points on the display is dh=(p'x (0,m-1) -p'x (0,0) ) / (m-1), the distance between the center of mass of the adjacent sub-pixel points in the longitudinal direction is dv=(p'y (n-1,0) -p'y (0,0) ) / (n-1).

5. The method for detecting brightness uniformity of a display according to claim 1, wherein The display displays the sequential image S1 in the lighting mode of the 1st column, (1+Gh)th column, (1+2Gh)th column, ... sub-pixels and the 1st row, (1+Gv)th column, (1+2Gv)th column, ... sub-pixels; the abscissa px of the centroid of the light spot corresponding to the upper left sub-pixel is S1-(0,0) =p'x (0,0) , vertical coordinate py S1-(0,0) =p'y (0,0) The centroid coordinates p of the light spot corresponding to the i-th horizontal and j-th vertical illuminated sub-pixel point in the sequence image S1 are calculated according to the following formula: S1-(i,j) =(px S1-(i,j) ,py S1-(i,j) ), i=0,1,2,…, j=0,1,2,…; Where Offset_x = p'x (n-1,0) -p'x (0,0) , is the lateral offset of the display imaging tilt in the captured image sequence, and the lateral offset averagely distributed to each row is Offset_y=p'y (0,m-1) -p'y (0,0) is the longitudinal offset of the tilted display imaging in the captured image sequence, and the lateral offset averagely distributed to each row is 6. The method for detecting brightness uniformity of a display according to claim 5, wherein For any sequence image St, assume that the sub-pixels in the pth column, (p+Gh)th column, (p+2Gh)th column, ... and the sub-pixels in the qth row, (q+Gv)th column, (q+2Gv)th column, ... of the display are lit; calculate the centroid coordinate p of the light spot corresponding to the sub-pixel in the i-th row and j-th column of the sequence image St according to the following formula: St-(i,j) : Δx=dh,Δy=dv.

7. The method for detecting brightness uniformity of a display according to claim 1, wherein The brightness uniformity D(Lw) of the display is calculated according to the following formula: Wherein Lws represents the brightness value of the s-th sub-pixel in the point-by-point brightness matrix, s=0 to (m×n-1).

8. The method for detecting brightness uniformity of a display according to claim 7, wherein The maximum value of the brightness uniformity corresponding to the red, green, and blue sub-pixels is selected to represent the brightness uniformity of the display.

Images