A method and device for quantifying moire patterns in a screen shot and a storage medium

Abstract

The application discloses a screen moire quantization method and device and a storage medium, which are used for improving display screen compensation accuracy. A sub-pixel unit image of a to-be-tested display screen is acquired; a brightness compensation process is performed on the sub-pixel unit image to generate an expected brightness compensation image; a compensation figure spectrum amplitude image is generated according to the brightness compensation image; a stripe band perception image is generated according to a visual distance and a pixel distance corresponding to a target color channel; the compensation figure spectrum amplitude image is regionally divided; a standard deviation of an elliptical region and a plurality of elliptical ring regions is calculated to generate a region standard deviation set; a standard deviation in the region standard deviation set is filled into a blank image to generate a first elliptical ring standard deviation image; the first elliptical ring standard deviation image is subjected to frequency domain filtering using the stripe band perception image to generate a first frequency domain filtering image; and a first moire quantization numerical value corresponding to the brightness compensation image is generated according to the compensation figure spectrum amplitude image and the first frequency domain filtering image.

Description

Technical Field

[0001] This application relates to the field of display screen testing, and in particular to a method, apparatus and storage medium for quantifying moiré patterns in display images. Background Technology

[0002] With technological innovation, new display technologies are constantly being updated and iterated. As a core link in the display industry chain, display quality inspection technology continues to receive attention, especially the detection of Mura defects at the micron level on the display surface, which has become a key factor determining the quality of display products. Optical Demura systems use high-resolution industrial cameras to capture images of illuminated displays and extract sub-pixel unit images from the captured images. This allows for the detection of Mura defects on the panel, enabling high-precision color / brightness compensation, thereby optimizing display quality and improving yield.

[0003] However, moiré patterns caused by the interference between the sampling grid of camera pixels and the display pixel grid significantly degrade the quality of images captured from the screen. Simultaneously, the ever-increasing screen pixel density, the increasingly sophisticated and complex arrangement of pixels, and the reduced sampling rate due to Bayer filters in color cameras further exacerbate the moiré pattern problem in images captured from new displays (such as OLED, Micro-LED, and flexible foldable screens) during automated optical inspection. While defocusing can mitigate or even eliminate some moiré patterns, it severely reduces the accuracy of sub-pixel chromaticity / luminance measurements, significantly diminishing the compensation effect of the Demura system. Therefore, quantifying the moiré pattern in images captured from the screen and designing corresponding demoiré filters based on the quantization results becomes a crucial task for displays in the Demura process.

[0004] In recent years, moiré quantization of displays has employed methods such as Cpd & Contrast and Mura Index. However, these methods primarily target moiré patterns generated by interference between stacked structures, such as metal mesh sensors, rather than those produced during the sampling process of industrial camera sensors. Furthermore, they typically do not consider the interference of the display signal itself on the calculation of moiré quantization values. This makes it impossible to calculate the severity of moiré patterns on the tested display after Demura compensation based solely on the captured image, thus reducing the accuracy of display compensation. Summary of the Invention

[0005] This application discloses a method, apparatus, and storage medium for quantifying moiré patterns in screen capture images, used to improve the compensation accuracy of displays.

[0006] In a first aspect, embodiments of this application provide a method for quantifying moiré patterns in a screen image, comprising: acquiring a sub-pixel unit image of a target color channel of a display screen under test using an industrial camera; performing brightness compensation processing on the sub-pixel unit image to generate a desired brightness compensation image; generating a compensation image spectrum amplitude image based on the brightness compensation image; generating a stripe frequency band sensing image based on the visual distance and the pixel spacing corresponding to the target color channel; dividing the compensation image spectrum amplitude image into regions to generate an elliptical region and several elliptical ring regions; calculating the standard deviations corresponding to the elliptical region and several elliptical ring regions respectively to generate a set of region standard deviations; sequentially filling the standard deviations in the set of region standard deviations into the corresponding regions of a blank image to generate a first elliptical ring standard deviation image of the brightness compensation image; performing frequency domain filtering on the first elliptical ring standard deviation image using the stripe frequency band sensing image to generate a first frequency domain filtered image; generating a first moiré pattern quantization value corresponding to the brightness compensation image based on the gray-scale mean of the acquired compensation image spectrum amplitude image and the gray-scale mean of the first frequency domain filtered image; and quantifying the moiré pattern generated by the industrial camera sampling when there is no moiré pattern caused by interference between multiple internal structures on the display screen based on the first moiré pattern quantization value.

[0007] Optionally, the step of generating a striped frequency band perception image based on the visual distance and the pixel spacing corresponding to the target color channel specifically includes: obtaining a preset visual distance, the horizontal and vertical sub-pixel spacing of the display screen under test in the target color channel; generating horizontal and vertical angular spatial frequencies based on the horizontal and vertical sub-pixel spacing and the screen resolution of the display screen under test; and generating a striped frequency band perception image based on the visual distance, the horizontal and vertical angular spatial frequencies, wherein the resolution of the striped frequency band perception image is the same as that of the compensation map spectrum amplitude image.

[0008] Optionally, the step of dividing the compensation image spectrum amplitude image into regions to generate an elliptical region and several elliptical ring regions specifically includes: generating a minor axis radius length based on the screen parameters of the display screen under test; generating an aspect ratio based on the resolution, horizontal spacing of subpixels, and vertical spacing of subpixels of the display screen under test; generating an elliptical region with the frequency domain center coordinates of the compensation image spectrum amplitude image as the center point, based on the minor axis radius length and aspect ratio; and generating several elliptical ring regions with inner contour minor axis radii that are integer multiples of the minor axis radius length, based on the frequency domain center coordinates of the compensation image spectrum amplitude image as the center point, based on the minor axis radius length and aspect ratio.

[0009] Optionally, the display screen under test contains moiré patterns generated by the interference of multiple internal structures. After the step of generating the first moiré quantization value corresponding to the brightness compensation image based on the gray-scale mean of the acquired compensation image spectrum amplitude image and the gray-scale mean of the first frequency domain filtered image, the quantization method further includes: setting a Gaussian kernel on a zero-value image based on the sub-pixel coordinates in the sub-pixel unit image to generate a Gaussian kernel image; performing a binary mask on the target color channel of the Gaussian kernel image according to the array mode of the industrial camera, and extracting the moiré pattern image from the Gaussian kernel image using deconvolution operation; and based on the moiré pattern... The process involves generating a moiré spectrum amplitude image from a given image; dividing the moiré spectrum amplitude image into elliptical and annular regions; generating a second elliptical annular standard deviation image based on the standard deviations of the elliptical and annular regions; performing frequency domain filtering on the second elliptical annular standard deviation image based on the stripe frequency band sensing image to generate a second frequency domain filtered image; generating a second moiré quantization value based on the amplitudes of the compensation image spectrum amplitude image, the second frequency domain filtered image, and the Gaussian kernel image; and quantizing the moiré generated by the industrial camera sampling when there are multiple layers of internal structures interfering with each other on the display screen based on the second moiré quantization value.

[0010] Optionally, after generating a second moiré quantization value based on the amplitude of the compensation map spectrum amplitude image, the second frequency domain filtered image, and the Gaussian kernel image, and quantizing the moiré generated by the industrial camera sampling when there are multiple layers of internal structures interfering with each other on the display screen based on the second moiré quantization value, the quantization method further includes: performing moiré filtering on the sub-pixel unit image based on the second moiré quantization value and the moiré pattern image.

[0011] Optionally, after the step of generating a moiré pattern spectrum amplitude image from the moiré pattern image, the quantization method further includes: performing upper half peak detection on the moiré pattern spectrum amplitude image to generate a set of peak position coordinates in the upper half; calculating the periodic and directional information of the moiré pattern generated by the industrial camera sampling based on the set of peak position coordinates in the upper half; and performing notch filtering on the sub-pixel unit image based on the calculated periodic and directional information of the moiré pattern generated by the industrial camera sampling.

[0012] Optionally, the steps of acquiring the sub-pixel unit image of the target color channel of the display screen under test using an industrial camera specifically include: inputting a positioning image and a grayscale image into the display screen under test respectively, wherein the positioning image is set with Mark points for color channel positioning; using an industrial camera to acquire images of the display screen under test respectively, generating Mark point images and grayscale images; using the Mark point images to locate the color channel on the grayscale images, generating the sub-pixel unit image of the target color channel.

[0013] Optionally, the step of generating a compensation map spectrum amplitude image based on the brightness compensation image specifically includes: performing a two-dimensional discrete Fourier transform on the brightness compensation image and performing spectrum centering processing to generate a first spectrum image; decomposing the first spectrum image to generate a compensation map spectrum amplitude image.

[0014] Secondly, embodiments of this application provide a device for quantizing moiré patterns in screen-captured images, comprising:

[0015] The system comprises the following components: an acquisition unit for acquiring sub-pixel unit images of the target color channel of the display screen under test using an industrial camera; a compensation unit for performing brightness compensation processing on the sub-pixel unit images to generate the expected brightness compensation image; a first generation unit for generating a compensation map spectrum amplitude image based on the brightness compensation image; a second generation unit for generating a stripe band perception image based on the visual distance and the pixel spacing corresponding to the target color channel; a region division unit for dividing the compensation map spectrum amplitude image into regions to generate an elliptical region and several elliptical ring regions; and a third generation unit for calculating the standard deviations corresponding to the elliptical region and the several elliptical ring regions respectively to generate a region... The system comprises: a standard deviation set; a filling unit, used to sequentially fill the standard deviations from the regional standard deviation set into the corresponding regions of a blank image to generate the first elliptical ring standard deviation image of the brightness compensation image; a first filtering unit, used to perform frequency domain filtering on the first elliptical ring standard deviation image using the stripe frequency band sensing image to generate the first frequency domain filtered image; and a fourth generation unit, used to generate the first moiré morphology quantization value corresponding to the brightness compensation image based on the gray mean of the acquired compensation image spectrum amplitude image and the gray mean of the first frequency domain filtered image, and to quantize the moiré generated by the industrial camera sampling when there is no moiré caused by the mutual interference of multiple internal structures on the display screen based on the first moiré morphology quantization value.

[0016] Optionally, the second generation unit specifically includes: acquiring a preset visual distance, the horizontal and vertical spacing of subpixels in the target color channel of the display screen under test; generating horizontal and vertical angular spatial frequencies based on the horizontal and vertical spacing of subpixels and the screen resolution of the display screen under test; and generating a stripe frequency band perception image based on the visual distance, the horizontal and vertical angular spatial frequencies, wherein the resolution of the stripe frequency band perception image is the same as that of the compensation map spectrum amplitude image.

[0017] Optionally, the region division unit specifically includes: generating a minor axis radius length based on the screen parameters of the display screen under test; generating an aspect ratio based on the resolution, horizontal spacing of subpixels, and vertical spacing of subpixels of the display screen under test; generating an elliptical region with the frequency domain center coordinates of the compensation map spectrum amplitude image as the center point, based on the minor axis radius length and aspect ratio; and generating several elliptical ring regions with inner contour minor axis radii that are integer multiples of the minor axis radius length, based on the frequency domain center coordinates of the compensation map spectrum amplitude image as the center point, based on the minor axis radius length and aspect ratio.

[0018] Optionally, the display screen under test contains moiré patterns caused by the interference of multiple internal structures. After the fourth generation unit, the quantization device further includes: a setting unit, used to set a Gaussian kernel on a zero-value image based on the sub-pixel coordinates in the sub-pixel unit image to generate a Gaussian kernel image; an extraction unit, used to perform a binary mask on the target color channel of the Gaussian kernel image according to the array mode of the industrial camera, and use deconvolution operation to extract the moiré pattern image from the Gaussian kernel image; a fifth generation unit, used to generate a moiré pattern spectrum amplitude image based on the moiré pattern image; a sixth generation unit, used to divide the moiré pattern spectrum amplitude image into elliptical regions and elliptical ring regions, and generate a second elliptical ring standard deviation image based on the standard deviation of the elliptical regions and elliptical ring regions; a second filtering unit, used to perform frequency domain filtering on the second elliptical ring standard deviation image based on the stripe frequency band sensing image to generate a second frequency domain filtered image; and a seventh generation unit, used to generate a second moiré quantization value based on the amplitude of the compensation map spectrum amplitude image, the second frequency domain filtered image, and the Gaussian kernel image, and quantize the moiré pattern generated by the industrial camera sampling when there are multiple internal structures interfering with each other on the display screen based on the second moiré quantization value.

[0019] Optionally, after the seventh generation unit, the quantization device further includes a filtering unit for performing moiré filtering on the sub-pixel unit image based on the second moiré quantization value and the moiré pattern image.

[0020] Optionally, after the fifth generation unit, the quantization device further includes: a detection unit for detecting the peak in the upper half of the moiré spectrum amplitude image and generating a set of peak position coordinates in the upper half; a calculation unit for calculating the periodic and directional information of the moiré generated by the industrial camera sampling based on the set of peak position coordinates in the upper half; and a third filtering unit for performing notch filtering on the sub-pixel unit image based on the calculated periodic and directional information of the moiré generated by the industrial camera sampling.

[0021] Optionally, the acquisition unit specifically includes: inputting the positioning image and the grayscale image into the display screen under test respectively, wherein the positioning image is set with Mark points for color channel positioning; using an industrial camera to acquire images of the display screen under test respectively, generating Mark point images and grayscale images; using the Mark point images to locate the color channels on the grayscale images, generating sub-pixel unit images of the target color channels.

[0022] Optionally, the first generation unit specifically includes: performing a two-dimensional discrete Fourier transform on the brightness compensation image and performing spectrum centering processing to generate a first spectrum image; and decomposing the first spectrum image to generate a compensation image spectrum amplitude image.

[0023] Thirdly, embodiments of this application provide an electronic device, including:

[0024] Processor, memory, input / output units, and bus;

[0025] The processor is connected to memory, input / output units, and a bus;

[0026] The memory stores a program, which the processor calls to execute, such as the first aspect and any optional quantization method of the first aspect.

[0027] Fourthly, embodiments of this application provide a computer-readable storage medium on which a program is stored, which, when executed on a computer, performs the first aspect and any optional quantization method of the first aspect.

[0028] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0029] In this application, an industrial camera is first used to acquire sub-pixel unit images of the target color channel of the display screen under test. Brightness compensation processing is performed on the sub-pixel unit images to generate the expected brightness compensation image. A compensation spectrum amplitude image is generated based on the brightness compensation image. A stripe band sensing image is generated based on the visual distance and the pixel spacing corresponding to the target color channel; the resolution of the stripe band sensing image is the same as that of the compensation spectrum amplitude image. The compensation spectrum amplitude image is divided into regions, generating an elliptical region and several elliptical ring regions. The standard deviations of the elliptical region and the several elliptical ring regions are calculated to generate a set of region standard deviations. The standard deviations from the set of region standard deviations are sequentially filled into the corresponding regions of a blank image to generate the first elliptical ring standard deviation image of the brightness compensation image; the size of the blank image is the same as that of the compensation spectrum amplitude image. The first elliptical ring standard deviation image is frequency domain filtered using the stripe band sensing image to generate a first frequency domain filtered image. The first moiré quantization value corresponding to the brightness compensation image is generated based on the gray mean of the compensation image spectrum amplitude image and the gray mean of the first frequency domain filtered image. The moiré generated by the industrial camera sampling when there is no moiré caused by the mutual interference of multiple internal structures on the display screen is quantized based on the first moiré quantization value.

[0030] The elliptical region and several elliptical ring regions are filtered. Then, the first frequency domain filtered image and the spectrum amplitude image of the compensation image are used to calculate the first moiré quantization value corresponding to the brightness compensation image. This method utilizes the characteristic that the sensing ability within each elliptical ring is basically unchanged, but the sensing ability of different elliptical rings varies greatly. That is, it uses visual perception characteristics to quantify the moiré of the screen image. It effectively quantifies the moiré generated by the sampling process of the industrial camera sensor. It can calculate the severity of the moiré on the screen under test after Demura compensation based on the sub-pixel unit image of the screen image, thereby improving the compensation accuracy of the screen. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a schematic diagram of the method for quantifying moiré patterns in screen-captured images according to this application;

[0033] Figure 2 A schematic diagram illustrating the method for generating stripe band sensing images according to this application;

[0034] Figure 3 This is a schematic diagram of the region division method for the spectral amplitude image of the compensation map in this application;

[0035] Figure 4 A schematic diagram illustrating the moiré pattern quantification method used in this application to eliminate the influence of moiré patterns caused by mutual interference between multi-layer internal structures;

[0036] Figure 5 This is a schematic diagram of the method for treating moiré patterns in this application;

[0037] Figure 6 This is another schematic diagram of the method for treating moiré patterns in this application;

[0038] Figure 7 This is a schematic diagram of the method for generating sub-pixel unit images according to this application;

[0039] Figure 8 A schematic diagram of a method for generating a compensated spectral amplitude image for this application;

[0040] Figure 9 This is a schematic diagram of the quantization device for the moiré pattern of the screen image captured in this application;

[0041] Figure 10 This is a schematic diagram of the electronic device used in this application;

[0042] Figure 11 This is a diagram showing the comparison between the average quantified values ​​and calculated values ​​of the observers in this application;

[0043] Figure 12 This is a schematic diagram of the standard deviation image of the elliptic ring before filtering in this application;

[0044] Figure 13 This is a schematic diagram of the filtered standard deviation image of the elliptic ring in this application;

[0045] Figure 14 This is a partial schematic diagram of the Gaussian kernel image of this application;

[0046] Figure 15 This is a partial schematic diagram of the moiré pattern image obtained from simulation. Detailed Implementation

[0047] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0048] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0049] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0050] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0051] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0052] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0053] In existing technologies, moiré quantization of displays has employed methods such as Cpd & Contrast and Mura Index. However, these methods primarily target moiré patterns generated by interference between stacked structures, such as metal mesh sensors, rather than those generated during the sampling process of industrial camera sensors. Furthermore, they typically do not consider the interference of the display signal itself on the calculation of moiré quantization values. Moreover, moiré patterns generated during the sampling process of industrial camera sensors not only exhibit weaker periodicity and more irregular shapes compared to those generated by interference between stacked structures, but also require the exclusion of the display signal's influence on the quantization value calculation when calculating the quantization index, which is linearly correlated with the degree of moiré, to guide subsequent demoiré filtering algorithms. This makes it impossible to calculate the severity of moiré patterns on the tested display after Demura compensation based solely on the captured image, thus reducing the accuracy of display compensation.

[0054] For the reasons mentioned above, this application provides an efficient screen capture image moiré quantization algorithm based on a contrast sensitivity function, which can calculate the severity of moiré patterns on the display screen after Demura compensation based on the sub-pixel unit image of the screen capture image.

[0055] In addition, this application also simulates the moiré pattern obtained by industrial camera sampling, which can accurately calculate the direction, period and quantization value of the moiré pattern under the current screen placement angle and position, and distinguish the moiré pattern generated by the mutual interference of the stacked structure of the display screen (the moiré pattern that needs to be compensated) and the moiré pattern generated by industrial camera sampling (the moiré pattern that needs to be filtered out). Thus, when filtering the moiré pattern, the moiré signal is retained and the moiré pattern generated by industrial camera sampling is filtered out, thereby improving the compensation accuracy.

[0056] Based on this, this application discloses a method, apparatus and storage medium for quantifying moiré patterns in screen images, which can be used to improve the compensation accuracy of the display screen.

[0057] The technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0058] The method described in this application can be applied to servers, devices, terminals, or other devices with logical processing capabilities; therefore, this application does not limit its application. For ease of description, the following description uses a terminal as the executing entity.

[0059] Please see Figure 1 This application provides an embodiment of a method for quantifying moiré patterns in screen-captured images, comprising:

[0060] 101. Use an industrial camera to acquire sub-pixel unit images of the target color channel of the display screen under test.

[0061] In this embodiment, the terminal uses an industrial camera to acquire images of the display screen under test to obtain sub-pixel unit images of the color channels. This is because conventional Demura systems typically perform defect detection and compensation processing on a single color channel. Therefore, this embodiment also performs moiré quantization processing on the brightness-compensated image of a single color channel after compensation. Specifically, the method of acquiring sub-pixel unit images of the target color channel of the display screen under test by acquiring grayscale images from the display screen under test using an industrial camera, and then generating sub-pixel unit images of the target color channels from the grayscale images, will be described in detail in subsequent embodiments.

[0062] 102. Perform brightness compensation processing on the sub-pixel unit image to generate the desired brightness-compensated image.

[0063] After the terminal acquires sub-pixel unit images at different gray levels, it determines the order based on two adjacent gray levels. and ( Sub-pixel unit image and calculate Grayscale brightness compensation image The specific generation method is as follows:

[0064]

[0065] 103. Generate a compensation spectrum amplitude image based on the brightness compensation image.

[0066] After generating the brightness-compensated image for the target color channel, the terminal generates a compensation map spectral amplitude image based on the brightness-compensated image. The compensated image spectrum amplitude image is an image formed by transforming the brightness compensated image into frequency domain data. The specific method for generating the compensated image spectrum amplitude image will be described in detail in subsequent embodiments.

[0067] 104. Generate a striped frequency band perception image based on the visual distance and the pixel spacing corresponding to the target color channel.

[0068] The stripe frequency band perception image is used to reflect or simulate human eye perception, specifically reflecting the range of stripe frequency bands that the human eye can perceive. In this embodiment, stripe frequency band perception coefficients are generated for each pixel by using visual distance and pixel spacing corresponding to the target color channel. The coefficients are then integrated to generate the stripe frequency band perception image. It can be seen that the resolution of the stripe frequency band perception image is the same as that of the compensation image spectrum amplitude image.

[0069] 105. Divide the compensated image spectrum amplitude into regions to generate an elliptical region and several elliptical ring regions.

[0070] The terminal divides the compensated spectrum amplitude image into regions based on visual perception and the characteristics of the display screen under test. The purpose is to make the visual perception of each region of the compensated spectrum amplitude image similar to that of the human eye, while the visual perception of different regions differs by more than a preset value, that is, to divide regions with different visual perceptions. In this embodiment, the final division shape is an elliptical region and several elliptical ring regions. The specific division method will be described in detail in subsequent embodiments.

[0071] 106. Calculate the standard deviations of the elliptical region and several elliptical annular regions respectively to generate a set of regional standard deviations.

[0072] In this embodiment, the terminal needs to calculate the standard deviation of the elliptical region and several elliptical ring regions.

[0073] 107. Fill the corresponding regions of a blank image with the standard deviations from the set of regional standard deviations to generate the first elliptical ring standard deviation image of the brightness-compensated image.

[0074] The terminal fills the standard deviations of the elliptical region and each elliptical annular region into the corresponding regions of a blank image with height H and width W, obtaining the first elliptical annular standard deviation image of the brightness compensation image. In this embodiment, the first elliptical annular standard deviation image can reflect the severity of moiré patterns in the brightness compensation image at different frequencies; the more severe the moiré patterns, the larger the standard deviation within the corresponding annular region.

[0075] Please refer to Figure 12 , Figure 12 This is a schematic diagram of the standard deviation image of the elliptic ring before filtering (the first standard deviation image of the elliptic ring).

[0076] 108. Use the stripe frequency band sensing image to perform frequency domain filtering on the first elliptical ring standard deviation image to generate the first frequency domain filtered image.

[0077] In this embodiment, the terminal uses the stripe frequency band sensing image to perform frequency domain filtering on the first elliptical ring standard deviation image to obtain the filtered first frequency domain filtered image. .

[0078] Please refer to Figure 13 , Figure 13 The filtered image of the standard deviation of the elliptic ring (first frequency domain filtered image) A schematic diagram of ( ).

[0079] 109. Generate the first moiré quantization value corresponding to the brightness compensation image based on the gray mean value of the obtained compensation image spectrum amplitude image and the gray mean value of the first frequency domain filtered image. Quantize the moiré generated by the industrial camera sampling when there is no moiré generated by the mutual interference of multiple internal structures on the display screen based on the first moiré quantization value.

[0080] Because the brightness-compensated image is interfered with by the Mura signal, the moiré quantization of the grayscale image cannot be performed in the conventional display moiré method by inverse Fourier transforming the filtered first frequency domain image back to the spatial domain and using the standard deviation or contrast of the spatial domain image as the quantization value. In this embodiment, we choose to calculate it directly in the frequency domain:

[0081]

[0082] in, The mean of the spectral amplitude image. Let Q be the mean of the filtered image, and let Q be the moiré quantization value of the brightness compensation image of the display screen under test. The first moiré quantization value is the moiré generated by the industrial camera sampling. In the above scheme, only the moiré generated during the industrial camera acquisition process is quantized. If the display screen under test has a relatively complex hierarchical structure, such as a multi-display layer structure formed by the combination of internal microcircuit layers and pixel layers, some moiré will be generated by the interference of multiple layers. The processing method of this type of moiré is different from that of the moiré generated by the industrial camera acquisition. The specific method will be described in detail in the following embodiments.

[0083] In this embodiment, an industrial camera is first used to acquire sub-pixel unit images of the target color channel of the display screen under test. Brightness compensation processing is performed on the sub-pixel unit images to generate the expected brightness compensation image. A compensation spectrum amplitude image is generated based on the brightness compensation image. A stripe band sensing image is generated based on the visual distance and the pixel spacing corresponding to the target color channel; the resolution of the stripe band sensing image is the same as that of the compensation spectrum amplitude image. The compensation spectrum amplitude image is divided into regions, generating an elliptical region and several elliptical ring regions. The standard deviations of the elliptical region and the several elliptical ring regions are calculated to generate a set of region standard deviations. The standard deviations from the set of region standard deviations are sequentially filled into the corresponding regions of a blank image to generate the first elliptical ring standard deviation image of the brightness compensation image; the size of the blank image is the same as that of the compensation spectrum amplitude image. The stripe band sensing image is used to perform frequency domain filtering on the first elliptical ring standard deviation image to generate a first frequency domain filtered image. The first moiré quantization value corresponding to the brightness compensation image is generated based on the gray-scale mean of the compensation image spectrum amplitude image and the gray-scale mean of the first frequency domain filtered image. The first moiré quantization value corresponding to the brightness compensation image is generated based on the gray-scale mean of the compensation image spectrum amplitude image and the gray-scale mean of the first frequency domain filtered image. The moiré generated by the industrial camera sampling when there is no moiré caused by the mutual interference of multiple internal structures on the display screen is quantized based on the first moiré quantization value.

[0084] The elliptical region and several elliptical ring regions are filtered. Then, the first frequency domain filtered image and the spectrum amplitude image of the compensation image are used to calculate the first moiré quantization value corresponding to the brightness compensation image. This method utilizes the characteristic that the sensing ability within each elliptical ring is basically the same, but the sensing ability of different elliptical rings varies greatly. That is, it uses visual perception characteristics to quantify the moiré of the screen image. It effectively quantifies the moiré generated by the sampling process of the industrial camera sensor. It can calculate the severity of the moiré on the screen under test after Demura compensation based on the sub-pixel unit image of the screen image, thereby improving the compensation accuracy of the screen.

[0085] Please see Figure 2 This application provides an embodiment of a method for generating stripe band sensing images, comprising:

[0086] 201. Obtain the preset visual distance, the horizontal and vertical spacing of subpixels in the target color channel of the display screen under test.

[0087] In this embodiment, the terminal acquires a preset visual distance L and the horizontal spacing of sub-pixels of the display screen under test in the target color channel. and subpixel vertical spacing (Unit: μm) Visual distance L represents the observation distance of the human eye to observe moiré patterns. This distance is preset and is generally designed to be within 200-400 mm.

[0088] 202. Generate the horizontal and vertical angular spatial frequencies based on the horizontal and vertical subpixel spacing and the screen resolution of the display screen under test.

[0089] Next, the terminal calculates the horizontal spacing between sub-pixels. Subpixel vertical spacing and the screen resolution H of the display under test W generates horizontal angular spatial frequency and vertical angular spatial frequency (Unit: rad / μm), the formula is as follows:

[0090]

[0091]

[0092] in, and These represent the horizontal and vertical distances of the current pixel from the center point of the stripe band sensing image, respectively.

[0093] 203. Generate a striped frequency band sensing image based on visual distance, horizontal angular spatial frequency, and vertical angular spatial frequency. The resolution of the striped frequency band sensing image is the same as that of the compensation map spectral amplitude image.

[0094] The terminal calculates the horizontal angular spatial frequency. and vertical angular spatial frequency Then, first based on the horizontal angular spatial frequency and vertical angular spatial frequency The integrated spatial frequency is generated using the following formula:

[0095]

[0096] The terminal then determines the distance L based on the visual distance and the overall spatial frequency. Generate stripe band sensing image The formula is as follows:

[0097]

[0098] As can be seen, the stripe frequency band sensing image in this embodiment is calculated based on the screen parameters of the display screen under test and the visual distance, so that the generated stripe frequency band sensing image can be well filtered in the subsequent process, resulting in better moiré quantization effect. The resolution of the stripe frequency band sensing image is the same as that of the compensation image spectrum amplitude image.

[0099] Please see Figure 3 This application provides an embodiment of a region partitioning method for a compensated image spectral amplitude image, comprising:

[0100] 301. Generate a minor axis radius length based on the screen parameters of the display screen to be tested.

[0101] 302. Generate the aspect ratio based on the resolution, horizontal spacing of subpixels, and vertical spacing of subpixels of the display screen under test.

[0102] 303. Using the frequency domain center coordinates of the compensated image spectrum amplitude as the center point, generate an elliptical region based on the minor axis radius and aspect ratio.

[0103] 304. Using the frequency domain center coordinates of the compensated image spectrum amplitude as the center point, generate several elliptical ring regions whose inner contour minor axis radius is an integer multiple of the minor axis radius length, based on the minor axis radius length and aspect ratio.

[0104] In this embodiment, the compensation graph spectral amplitude image is centered at the frequency domain center point, with a minor axis radius of length b and an aspect ratio of [missing value]. The terminal calculates the standard deviation of the spectral amplitude image of the compensated image within the elliptical region.

[0105] Next, with the frequency domain center point as the center, the width along the minor axis is b, and the aspect ratio of the outer contour is... The minor axis radii of the inner contour are respectively Multiple elliptical rings are used, and the standard deviation of each elliptical ring within the compensated image spectrum amplitude is calculated. The minor axis radius length b is determined by the visual distance L. The smaller the set visual distance L, the larger the minor axis radius length b. This design can well combine the characteristics of the display screen under test and the visual distance, and can better quantify moiré patterns.

[0106] Please see Figure 4 This application provides an embodiment of a moiré quantization method that eliminates the influence of moiré patterns caused by interference between multiple internal structures on the display screen under test. This method can eliminate the interference of moiré patterns caused by interference between multiple internal structures on the display screen under test on the calculation of quantization values, including:

[0107] 401. Based on the sub-pixel coordinates in the sub-pixel unit image, set a Gaussian kernel on a zero-value image to generate a Gaussian kernel image.

[0108] Some new types of displays have multi-layered structures, which are prone to interference, generating numerous moiré patterns. These moiré patterns are Mura defects that need compensation. However, due to their periodic structure, these moiré patterns are composed of straight lines, with each pair of adjacent lines being approximately parallel and equidistant, thus exhibiting strong periodicity. During moiré removal filtering, these patterns are easily mistaken for screen-captured moiré patterns and filtered out, resulting in the corresponding Mura defects not receiving brightness compensation.

[0109] In order to better distinguish between periodic Mura defects and moiré patterns generated by industrial camera sampling during subsequent moiré pattern removal filtering, this embodiment simulates the morphology of screen-captured moiré patterns at the current placement position of the display screen under test based on grayscale images.

[0110] When the terminal locates the coordinates of all sub-pixels of the current color channel in a high-resolution screen-captured image (grayscale image) based on the positioning map (mark point image), Where N is the total number of sub-pixels in the current color channel. These are the horizontal and vertical coordinates of the i-th sub-pixel, respectively. Next, the terminal generates a zero-value image of the same size as the captured image, and places the zero-values ​​in this image... Place a Gaussian kernel at the coordinate position. As shown below:

[0111]

[0112] Where σ and A are the standard deviation and amplitude of the Gaussian kernel, respectively, and x and y are the horizontal and vertical coordinate variables, respectively.

[0113] Please refer to Figure 14 , Figure 14 This is a partial schematic diagram of a Gaussian kernel image.

[0114] 402. Apply a binary mask to the target color channel of the Gaussian kernel image according to the array mode of the industrial camera, and use deconvolution operation to extract the moiré pattern image from the Gaussian kernel image.

[0115] In this embodiment, the terminal applies a binary mask to the target color channel of the Gaussian kernel image according to the array mode of the industrial camera, and uses deconvolution to extract the moiré pattern image from the Gaussian kernel image. Specifically, the terminal applies a binary mask corresponding to the target color channel to the generated Gaussian kernel image according to the current camera's Bayer array mode, setting all pixels in other color channels to 0. Based on the sub-pixel positions... The moiré pattern image is extracted from the Gaussian kernel image using deconvolution, which is the moiré pattern image of the sub-pixel unit image at the current screen placement position obtained through simulation. .

[0116] Please refer to Figure 15 , Figure 15 For the simulated moiré pattern image A partial schematic diagram.

[0117] 403. Generate a moiré pattern spectrum amplitude image based on the moiré pattern image.

[0118] In this embodiment, the terminal performs a two-dimensional discrete Fourier transform on the moiré pattern image and performs spectrum centering processing, shifting the DC component to the center of the frequency domain to obtain the moiré pattern spectrum image. The moiré spectrum image is then decomposed to obtain its moiré spectrum amplitude image.

[0119] 404. Divide the moiré spectrum amplitude image into elliptical regions and elliptical ring regions, and generate a second elliptical ring standard deviation image based on the standard deviation of the elliptical regions and elliptical ring regions.

[0120] The terminal divides the moiré spectrum amplitude image into elliptical regions and elliptical ring regions, and generates a second elliptical ring standard deviation image based on the standard deviation of the elliptical regions and elliptical ring regions. This step is similar to the region division method of the compensation map spectrum amplitude image in Example 3, and will not be described in detail here.

[0121] 405. Based on the stripe frequency band sensing image, perform frequency domain filtering on the standard deviation image of the second elliptical ring to generate the second frequency domain filtered image.

[0122] The terminal performs frequency domain filtering on the second elliptical ring standard deviation image based on the stripe frequency band sensing image to generate a second frequency domain filtered image. This step is similar to step 108 in Embodiment 1, and will not be described in detail here.

[0123] 406. Generate a second moiré quantization value based on the amplitude of the compensated image spectrum amplitude image, the second frequency domain filtered image, and the Gaussian kernel image. Quantize the moiré generated by the industrial camera sampling when there are multiple layers of internal structures interfering with each other on the display screen based on the second moiré quantization value.

[0124] The terminal uses the compensated spectrum amplitude image and the second frequency domain filtered image. The amplitude of the Gaussian kernel image is used to generate a second moiré quantization value. This second moiré quantization value is the quantization value of the moiré generated by the industrial camera sampling after eliminating the interference of moiré caused by the mutual interference of multiple internal structures. The formula is as follows:

[0125]

[0126] in, For image The mean, To compensate for the amplitude of the spectrum image The average gray level, This represents the quantization value of the simulated moiré pattern image.

[0127] Please see Figure 5 This application provides another embodiment of a method for processing moiré patterns, comprising:

[0128] 501. Moiré pattern filtering is performed on the sub-pixel unit image based on the second moiré quantization value and the moiré pattern image.

[0129] After acquiring the quantized moiré pattern, the terminal performs moiré filtering on the sub-pixel unit image based on the second moiré quantization value and the moiré pattern image. Specifically, the terminal filters out the moiré generated by industrial camera sampling in the sub-pixel unit image of the target color channel. One method is to process it using the following formula:

[0130]

[0131] in, It is a sub-pixel unit image with moiré patterns filtered out. This is a sub-pixel unit image of the moiré pattern to be filtered out. It is a moiré pattern image. The mean.

[0132] Please see Figure 6 This application provides an embodiment of a method for processing moiré patterns, comprising:

[0133] 601. Perform peak detection in the upper half of the moiré spectrum amplitude image and generate a set of peak position coordinates in the upper half.

[0134] In this embodiment, the terminal performs peak detection in the upper half of the moiré spectrum amplitude image, generating a set of peak position coordinates in the upper half. Specifically, the terminal uses a thresholding method to detect the peak positions of the moiré spectrum amplitude map in the upper half of the image. M represents the total number of peaks in the upper half of the image. These are the coordinates of the j-th peak after centering, representing the horizontal and vertical coordinates.

[0135] 602. Calculate the periodic and directional information of the moiré pattern generated by the industrial camera sampling based on the set of peak position coordinates in the upper half of the region.

[0136] Next, the terminal calculates the period and direction information of the moiré patterns generated by the industrial camera sampling based on the set of peak position coordinates in the upper half of the region. Specifically, the period and direction of the corresponding moiré patterns can be calculated based on the peak positions as follows:

[0137]

[0138]

[0139] 603. Perform notch filtering on the sub-pixel unit image based on the calculated periodic and directional information of the moiré patterns generated by the industrial camera sampling.

[0140] Another method used by the terminal to filter out moiré patterns is to use the periodic information of the moiré patterns. and direction information A notch filter operation is directly performed on the sub-pixel unit image. During the filtering process, only the moiré components in the direction and period corresponding to the simulated moiré pattern image are filtered out, resulting in a sub-pixel unit image after removing the moiré.

[0141] Please see Figure 7 This application provides an embodiment of a method for generating sub-pixel unit images, comprising:

[0142] 701. Input the positioning screen and grayscale screen into the display screen to be tested respectively. The positioning screen is set with Mark points for color channel positioning.

[0143] 702. Use an industrial camera to acquire images of the display screen under test, and generate Mark point images and grayscale images.

[0144] 703. Locate the color channels on the grayscale image using the Mark point image and generate the sub-pixel unit image of the target color channel.

[0145] Based on the characteristics of the display screen under test, corresponding positioning images and several grayscale images are created. Only one positioning image is needed. This positioning image has marker points set on it for color channel positioning. The positioning image is then combined with several grayscale images. The images are input into the display screen under test, which then displays the positioning image and several grayscale images. In a darkroom environment, an industrial camera is used to capture images of the display screen under test, generating Mark point images and grayscale images.

[0146] Color channels are located in grayscale images using Mark point images and the Demura algorithm. Sub-pixel unit images corresponding to each grayscale level are extracted, specifically the sub-pixel unit images of the target color channels generated from the sub-pixels of the R, G, and B channels. Finally, R, G, and B sub-pixel unit images are generated, and each sub-pixel unit image needs to be quantized for moiré patterns.

[0147] Please see Figure 8 This application provides an embodiment of a method for generating a compensated graph spectral amplitude image, comprising:

[0148] 801. Perform a two-dimensional discrete Fourier transform on the brightness-compensated image and perform spectrum centering to generate the first spectrum image.

[0149] 802. Decompose the first spectrum image to generate a compensated spectrum amplitude image.

[0150] In this embodiment, the terminal performs a two-dimensional discrete Fourier transform on the brightness compensation image and then performs spectrum centering processing on the transformed data. The purpose is to move the DC component to the center of the frequency domain to generate a first spectrum image of the brightness compensation image. Next, the terminal decomposes the first spectrum image to obtain the spectrum amplitude image of the compensation image. .

[0151] After completing the quantization of moiré patterns generated by the industrial camera, a moiré quantization experiment was conducted. Specifically, 12 different compensated images were displayed on the same OLED phone (honor 30 pro). Seven randomly selected observers rated the moiré intensity of each image at a viewing distance of 300mm (0 for no moiré, 7 for the most severe moiré). The average of the data from the seven observers was used to obtain the moiré intensity of each image, which was then compared with the quantization value calculated by the moiré quantization method of this application. Figure 11 As shown, Figure 11 This diagram illustrates the comparison between the observer's average quantified value and the calculated value in this application. The calculated quantified value is normalized to the range of 0-7 by multiplying it by a constant for comparison with the observer's average value. Figure 11 It can be seen that the quantified value and the observer's average value have a strong linear positive correlation (correlation coefficient of 0.9748), indicating that the moiré quantification algorithm proposed in this application can accurately reflect the severity of moiré patterns in screen-captured images.

[0152] Please see Figure 9 This application provides an embodiment of a device for quantizing moiré patterns in screen-captured images, comprising:

[0153] Acquisition unit 901 is used to acquire sub-pixel unit images of the target color channel of the display screen under test using an industrial camera.

[0154] Optionally, the acquisition unit 901 specifically includes:

[0155] The positioning image and grayscale image are input into the display screen under test, respectively. The positioning image has Mark points set on it for color channel positioning.

[0156] An industrial camera is used to acquire images of the display screen under test, generating Mark point images and grayscale images.

[0157] By locating color channels on grayscale images using Mark point images, sub-pixel unit images of the target color channels are generated.

[0158] The compensation unit 902 is used to perform brightness compensation processing on the sub-pixel unit image to generate the expected brightness-compensated image.

[0159] The first generation unit 903 is used to generate a compensation map spectrum amplitude image based on the brightness compensation image.

[0160] Optionally, the first generation unit 903 specifically includes:

[0161] A two-dimensional discrete Fourier transform is performed on the brightness-compensated image, followed by spectral centering to generate the first spectral image.

[0162] The first spectral image is decomposed to generate a compensated spectral amplitude image.

[0163] The second generation unit 904 is used to generate a stripe band perception image based on the visual distance and the pixel spacing corresponding to the target color channel.

[0164] Optionally, the second generation unit 904 specifically includes:

[0165] Obtain the preset visual distance, the horizontal and vertical subpixel spacing of the display screen under test in the target color channel.

[0166] The horizontal and vertical angular spatial frequencies are generated based on the horizontal and vertical subpixel spacing and the screen resolution of the display under test.

[0167] A striped frequency band sensing image is generated based on visual distance, horizontal angular spatial frequency, and vertical angular spatial frequency. The resolution of the striped frequency band sensing image is the same as that of the compensated image spectral amplitude image.

[0168] The region division unit 905 is used to divide the compensated image spectrum amplitude image into regions to generate an elliptical region and several elliptical ring regions.

[0169] Optionally, the regional division unit 905 specifically includes:

[0170] A minor axis radius length is generated based on the screen parameters of the display screen under test.

[0171] The aspect ratio is generated based on the resolution, horizontal spacing of subpixels, and vertical spacing of subpixels of the display screen under test.

[0172] Using the center coordinates of the frequency domain of the compensated image spectrum amplitude as the center point, an elliptical region is generated based on the minor axis radius and aspect ratio.

[0173] Using the frequency domain center coordinates of the compensated image spectrum amplitude as the center point, several elliptical ring regions with inner contour minor axis radii that are integer multiples of the minor axis radius are generated based on the minor axis radius length and aspect ratio.

[0174] The third generation unit 906 is used to calculate the standard deviations corresponding to the elliptical region and several elliptical ring regions respectively, so as to generate a set of regional standard deviations.

[0175] Filling unit 907 is used to fill the standard deviations in the set of regional standard deviations into the corresponding regions of a blank image in sequence to generate the first elliptical ring standard deviation image of the brightness-compensated image.

[0176] The first filtering unit 908 is used to perform frequency domain filtering on the first elliptical ring standard deviation image using the stripe frequency band sensing image to generate a first frequency domain filtered image.

[0177] The fourth generation unit 909 is used to generate a first moiré quantization value corresponding to the brightness compensation image based on the gray average value of the acquired compensation image spectrum amplitude image and the gray average value of the first frequency domain filtered image, and to quantize the moiré generated by the industrial camera sampling when there is no moiré generated by the mutual interference of multiple internal structures on the display screen based on the first moiré quantization value.

[0178] Setting unit 910 is used to set a Gaussian kernel on a zero-value image based on the sub-pixel coordinates in the sub-pixel unit image to generate a Gaussian kernel image. The display screen under test has moiré patterns generated by the mutual interference of multiple internal structures.

[0179] Extraction unit 911 is used to perform binary masking on the target color channel of the Gaussian kernel image according to the array mode of the industrial camera, and to extract the moiré pattern image from the Gaussian kernel image using deconvolution operation.

[0180] The fifth generation unit 912 is used to generate a moiré spectrum amplitude image based on the moiré pattern image.

[0181] The sixth generation unit 913 is used to divide the moiré spectrum amplitude image into elliptical regions and elliptical ring regions, and generate a second elliptical ring standard deviation image based on the standard deviation of the elliptical regions and elliptical ring regions.

[0182] The second filtering unit 914 is used to perform frequency domain filtering on the second elliptical ring standard deviation image based on the stripe frequency band sensing image to generate a second frequency domain filtered image.

[0183] The seventh generation unit 915 is used to generate a second moiré quantization value based on the amplitude of the compensation map spectrum amplitude image, the second frequency domain filtered image and the Gaussian kernel image, and to quantize the moiré generated by the industrial camera sampling when there are multiple layers of internal structures interfering with each other on the display screen based on the second moiré quantization value.

[0184] The filtering unit 916 is used to perform moiré filtering on the sub-pixel unit image based on the second moiré quantization value and the moiré pattern image.

[0185] The detection unit 917 is used to perform peak detection in the upper half of the moiré spectrum amplitude image and generate a set of peak position coordinates in the upper half.

[0186] The calculation unit 918 is used to calculate the periodic and directional information of the moiré pattern generated by the industrial camera sampling based on the set of peak position coordinates in the upper half of the region.

[0187] The third filtering unit 919 is used to perform notch filtering on the sub-pixel unit image based on the periodic and directional information of the moiré pattern generated by the industrial camera sampling.

[0188] Please see Figure 10 This application provides an electronic device, including: a processor 1001, a memory 1002, an input / output unit 1003, and a bus 1004. The processor 1001 is connected to the memory 1002, the input / output unit 1003, and the bus 1004. The memory 1002 stores a program, and the processor 1001 calls the program to execute... Figure 1 , Figure 2 and Figure 3 , Figure 4 , Figure 5 , Figure 6 , Figure 7 and Figure 8 Quantization methods in [the context].

[0189] This application provides a computer-readable storage medium on which a program is stored, and when the program is executed on a computer, it performs the following... Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 5 , Figure 6 , Figure 7 and Figure 8 Quantization methods in [the context].

[0190] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0191] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0192] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0193] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0194] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for quantifying moiré patterns in screen-captured images, characterized in that, include: Use an industrial camera to acquire sub-pixel unit images of the target color channel of the display screen under test; The sub-pixel unit image is subjected to brightness compensation processing to generate the desired brightness-compensated image; Generate a compensation map spectrum amplitude image based on the brightness compensation image; A stripe band perception image is generated based on visual distance and the pixel spacing corresponding to the target color channel; The compensated image spectrum amplitude image is divided into regions to generate an elliptical region and several elliptical ring regions; Calculate the standard deviations of the elliptical region and several elliptical annular regions respectively to generate a set of regional standard deviations; The standard deviations in the set of regional standard deviations are sequentially filled into the corresponding regions of a blank image to generate the first elliptical ring standard deviation image of the brightness-compensated image; The first elliptical ring standard deviation image is frequency domain filtered using the stripe frequency band sensing image to generate a first frequency domain filtered image. The first moiré quantization value corresponding to the brightness compensation image is generated based on the gray mean value of the obtained compensation image spectrum amplitude image and the gray mean value of the first frequency domain filtered image. The moiré generated by the industrial camera sampling when there is no moiré generated by the mutual interference of multiple internal structures on the display screen is quantized based on the first moiré quantization value.

2. The quantization method according to claim 1, characterized in that, The step of generating a striped frequency band perception image based on visual distance and pixel spacing corresponding to the target color channel specifically includes: Obtain the preset visual distance, the horizontal and vertical subpixel spacing of the display screen under test in the target color channel; The horizontal angular spatial frequency and the vertical angular spatial frequency are generated based on the horizontal spacing of the sub-pixels, the vertical spacing of the sub-pixels, and the screen resolution of the display screen under test. A striped frequency band sensing image is generated based on the visual distance, the horizontal angular spatial frequency, and the vertical angular spatial frequency. The resolution of the striped frequency band sensing image is the same as the resolution of the compensated image spectral amplitude image.

3. The quantization method according to claim 1, characterized in that, The step of dividing the compensated image spectrum amplitude image into regions to generate an elliptical region and several elliptical ring regions specifically includes: A minor axis radius length is generated based on the screen parameters of the display screen under test; The aspect ratio is generated based on the resolution, horizontal spacing of subpixels, and vertical spacing of subpixels of the display screen under test; Using the frequency domain center coordinates of the compensated image spectrum amplitude as the center point, an elliptical region is generated based on the minor axis radius and the aspect ratio. Using the frequency domain center coordinates of the compensated image spectrum amplitude as the center point, several elliptical ring regions with inner contour minor axis radii that are integer multiples of the minor axis radius are generated according to the minor axis radius length and the aspect ratio.

4. The quantization method according to claim 1, characterized in that, The display screen under test has moiré patterns caused by the interference of multiple internal structures. After the step of generating the first moiré quantization value corresponding to the brightness compensation image based on the gray-scale mean of the acquired compensation image spectrum amplitude image and the gray-scale mean of the first frequency domain filtered image, the quantization method further includes: A Gaussian kernel image is generated by setting a Gaussian kernel on a zero-value image based on the sub-pixel coordinates in the sub-pixel unit image. The target color channel of the Gaussian kernel image is subjected to a binary mask according to the array mode of the industrial camera, and the moiré pattern image is extracted from the Gaussian kernel image using deconvolution operation. Generate a moiré pattern spectrum amplitude image based on the moiré pattern image; The moiré spectrum amplitude image is divided into elliptical regions and elliptical ring regions, and a second elliptical ring standard deviation image is generated based on the standard deviation of the elliptical regions and the elliptical ring regions. The second elliptical ring standard deviation image is frequency domain filtered based on the stripe frequency band sensing image to generate a second frequency domain filtered image; A second moiré quantization value is generated based on the amplitude of the compensated image spectrum amplitude image, the second frequency domain filtered image, and the Gaussian kernel image. The moiré generated by the industrial camera sampling when there are multiple layers of internal structures interfering with each other on the display screen is quantized based on the second moiré quantization value.

5. The quantization method according to claim 4, characterized in that, After the steps of generating a second moiré quantization value based on the amplitude of the compensated image spectrum amplitude image, the second frequency domain filtered image, and the Gaussian kernel image, and quantizing the moiré generated by industrial camera sampling when there are multiple layers of internal structures interfering with each other on the display screen based on the second moiré quantization value, the quantization method further includes: Moiré pattern removal is performed on the sub-pixel unit image based on the second moiré quantization value and the moiré pattern image.

6. The quantization method according to claim 4, characterized in that, After the step of generating a moiré spectrum amplitude image based on the moiré pattern image, the quantization method further includes: Peak detection is performed on the upper half of the moiré spectrum amplitude image to generate a set of peak position coordinates in the upper half. The periodic and directional information of the moiré patterns generated by the industrial camera sampling is calculated based on the set of peak position coordinates in the upper half of the region. Notch filtering is performed on the sub-pixel unit image based on the calculated periodic and directional information of the moiré patterns generated by the industrial camera sampling.

7. The quantization method according to any one of claims 1 to 6, characterized in that, The step of acquiring the sub-pixel unit image of the target color channel of the display screen under test using an industrial camera specifically includes: The positioning image and grayscale image are input into the display screen to be tested, and the positioning image is provided with Mark points for color channel positioning. An industrial camera is used to acquire images of the display screen under test, generating Mark point images and grayscale images. The color channels are located on the grayscale image using the Mark point image, and a sub-pixel unit image of the target color channel is generated.

8. The quantization method according to any one of claims 1 to 6, characterized in that, The step of generating a compensation map spectral amplitude image based on the brightness compensation image specifically includes: The brightness-compensated image is subjected to a two-dimensional discrete Fourier transform and then spectral centering is performed to generate a first spectral image. The first spectral image is decomposed to generate a compensated spectral amplitude image.

9. A device for quantifying moiré patterns in screen-captured images, characterized in that, include: The acquisition unit is used to acquire sub-pixel unit images of the target color channel of the display screen under test using an industrial camera; A compensation unit is used to perform brightness compensation processing on the sub-pixel unit image to generate the expected brightness-compensated image. The first generation unit is used to generate a compensated image spectrum amplitude image based on the brightness compensated image; The second generation unit is used to generate a stripe frequency band perception image based on the visual distance and the pixel spacing corresponding to the target color channel; A region division unit is used to divide the compensated image spectrum amplitude image into regions to generate an elliptical region and several elliptical ring regions. The third generation unit is used to calculate the standard deviations corresponding to the elliptical region and several elliptical annular regions respectively, so as to generate a set of regional standard deviations; A filling unit is used to sequentially fill the standard deviations in the set of regional standard deviations into the corresponding regions of a blank image to generate the first elliptical ring standard deviation image of the brightness-compensated image; The first filtering unit is used to perform frequency domain filtering on the first elliptical ring standard deviation image using the stripe frequency band sensing image to generate a first frequency domain filtered image. The fourth generation unit is used to generate a first moiré quantization value corresponding to the brightness compensation image based on the gray-scale mean value of the acquired compensation image spectrum amplitude image and the gray-scale mean value of the first frequency domain filtered image, and to quantize the moiré generated by the industrial camera sampling when there is no moiré generated by the mutual interference of multiple internal structures on the display screen based on the first moiré quantization value.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium contains a program that, when executed on a computer, performs the quantization method as described in any one of claims 1 to 8.

Images