Backlight source online detection method based on optical equipment

Abstract

The invention discloses a backlight source online detection method based on optical equipment, relates to the field of backlight source detection, and solves the problems that the brightness uniformity of a whole screen cannot be comprehensively reflected when a current screen is subjected to backlight source detection, and the detection result of a traditional detection method has errors. The method comprises the following steps: preprocessing first real-time video data to obtain a first to-be-tested screen image of the first real-time video data; performing first backlight detection on the first to-be-tested screen image, and obtaining an abnormal type of the to-be-tested screen according to a first backlight detection result; performing second backlight detection on the second video and the second real-time video data, and obtaining the backlight state of the to-be-tested screen according to a second backlight detection result; and visually displaying the first backlight source detection result and the second backlight source detection result of the to-be-tested screen. According to the invention, the backlight source of the screen can be automatically detected under the condition that the detection precision is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of backlight detection, and specifically relates to an on-line backlight detection method based on optical equipment. Background Art

[0002] Backlight detection is a detection method used to evaluate whether the backlight system of a display device (such as an LED matrix screen, a liquid crystal screen, etc.) is working properly. This detection mainly focuses on factors such as the brightness uniformity, color uniformity, stroboscopic situation, and low-brightness performance of the screen. By analyzing the optical characteristics of the screen under specific test videos (such as pure white, pure black, and gray images), and comparing the real-time image captured by the optical equipment with the standard reference image, problems such as uneven backlight brightness, color difference, local abnormal light emission or flicker can be found to ensure the overall visual quality of the display device.

[0003] In the prior art, when detecting the backlight of a screen, it relies on a simple calculation of the maximum-minimum brightness ratio and only uses a small number of test points, which cannot comprehensively reflect the brightness uniformity of the entire screen. At the same time, traditional detection methods require manual observation and subjective judgment, resulting in errors in the backlight detection results of the screen. Therefore, the present invention proposes an on-line backlight detection method based on optical equipment. Summary of the Invention

[0004] The purpose of the present invention is to propose an on-line backlight detection method based on optical equipment to solve the problems raised in the above background art.

[0005] The technical problem to be solved by the present invention is: How to achieve automatic detection of the backlight of a screen while improving the detection accuracy.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions: An on-line backlight detection method based on optical equipment, the method is as follows: Step S1, obtain the aspect ratio of the screen to be tested and the first real-time video data of the screen to be tested when playing the first video through an optical device; Step S2, preprocess the first real-time video data, and the preprocessing obtains the first screen image to be tested of the first real-time video data; Step S3, perform the first backlight detection on the first screen image to be tested, and obtain the abnormal type of the screen to be tested according to the first backlight detection result; Step S4, obtain the second real-time video data of the screen to be tested when playing the second video through an optical device, and perform the second backlight detection on the second video and the second real-time video data, and obtain the backlight state of the screen to be tested according to the second backlight detection result. Step S5, visually display the first backlight detection result and the second backlight detection result of the screen to be tested.

[0007] Furthermore, the first real-time video data is the first video stream and the first video stream frame rate of the first video stream; The second real-time video data includes the second video stream and the second video stream frame rate of the second video stream.

[0008] Furthermore, the said Step S2 includes the following sub-steps: Step S21, decompose the first video stream in the first real-time video data into first video stream frames, specifically: When the first video stream frame rate of the first video stream is greater than or equal to 30 frames per second, decompose the first video stream one by one to obtain first video stream images, and randomly select 15 first video stream images as the first video stream frames of the screen to be tested; when the first video stream frame rate of the first video stream is less than 30 frames per second, decompose the first video stream one by one to obtain the first video stream frames of the screen to be tested; Step S22, denoise the first video stream frames through median filtering technology; Step S23, convert the first video stream frames into first video stream frame grayscale images through a formula, and the formula is specifically as follows: I(x, y) = 0.3R(x, y) + 0.59G(x, y) + 0.11B(x, y), where (x, y) is the coordinate of any pixel point in the first video stream frame grayscale image; Step S24, calculate the horizontal direction gradient Ix and the vertical direction gradient Iy of any pixel point in the first video stream frame grayscale image through a formula, and the formula is specifically as follows: Ix = ∂ I / ∂ x; Iy = ∂ I / ∂ y; Step S25, integrate the calculated horizontal direction gradients and vertical direction gradients corresponding to all pixel points to obtain the gradient matrix M of the first video stream frame grayscale image, and the gradient matrix is specifically as follows: ; Step S26, calculate the corner response value JDX of the first video stream frame grayscale image, and the formula is specifically as follows: JDX = det(M) - k × [trace(M)], where k is an empirical coefficient; Step S27, when the corner response value is greater than the corner response threshold, determine that the corresponding pixel point is a corner point; When the corner response value is less than or equal to the corner response threshold, no operation is performed; Step S28, calculate the polygon area in the grayscale image of the first video stream frame; Step S29, repeat Step S28 to obtain the polygon areas in the grayscale images of the first video stream frames corresponding to all the first video stream frames, and analyze the polygon areas to obtain the first test screen image of the screen to be tested.

[0009] Further, the analysis process of the polygon area is specifically as follows: Step S291, construct the minimum bounding rectangle of the polygon area in the grayscale image of the first video stream frame, and obtain the rectangular vertex coordinates of the minimum bounding rectangle; Step S292, calculate the length and width of the minimum bounding rectangle according to the rectangular vertex coordinates of the minimum bounding rectangle, divide the length and width of the minimum bounding rectangle, and obtain the aspect ratio of the minimum bounding rectangle; Step S293, obtain the aspect ratio of the screen to be tested, subtract the aspect ratio of the minimum bounding rectangle of each grayscale image of the first video stream frame from the aspect ratio of the screen to be tested one by one and take the absolute value to obtain the aspect ratio difference between the minimum bounding rectangle of the grayscale image of the first video stream frame and the screen to be tested; Step S294, use the minimum bounding rectangle corresponding to the minimum aspect ratio difference as the first test screen image of the first real-time video data.

[0010] Further, the step S3 includes the following sub-steps: Step S31, construct the measurement grid of the first test screen image; Step S32, perform brightness uniformity detection on the first test screen image; Step S33, perform brightness stability detection on the first test screen image Step S34, perform color uniformity detection on the first test screen image; Step S35, play the black source video and the gray source video on the screen to be tested respectively, and repeat Steps S31 to S34; If there is no abnormal type in the screen to be tested, enter Step S4; If there is any abnormal type in the screen to be tested, stop the detection work of the first backlight.

[0011] Further, the brightness uniformity detection is as follows: Step S321, obtain the pixel brightness values of all pixel points in the measurement grid; Step S322: Obtain the maximum pixel brightness value and the minimum pixel brightness value of all pixel points within the measurement grid, divide the minimum pixel brightness value by the maximum pixel brightness value to obtain the brightness uniformity value of the measurement grid; Step S323: If the brightness uniformity value of the measurement grid is less than the brightness uniformity threshold, determine that the abnormal type of the screen to be tested is brightness uniformity abnormality; If the brightness uniformity value of the measurement grid is greater than or equal to the brightness uniformity threshold, then proceed to the subsequent detection steps.

[0012] Furthermore, the brightness stability detection is as follows: Step S331: The screen to be tested plays a first video for a first fixed duration, takes a first image of the screen to be tested at intervals of a second fixed duration, and uses the first image of the screen to be tested taken when starting to play the first video as the initial grayscale image; wherein, the first fixed duration is greater than the second fixed duration, and the second fixed duration is greater than zero; Step S332: Obtain the initial average brightness CPL of the initial grayscale image; Step S333: Obtain the node average brightness JPLi corresponding to the first image of the screen to be tested at any time node, where i is the number of the time node, i = 1, 2,..., n, n is a positive integer, and calculate the brightness attenuation value SJZi of the first image of the screen to be tested through the formula. The specific formula is as follows: SJZi = (CPL - JPLi) / CPL; Step S334: When the brightness attenuation value of the first image of the screen to be tested at any time node is greater than or equal to the brightness attenuation threshold, determine that the abnormal type of the screen to be tested is brightness stability abnormality; When the brightness attenuation values of the first images of the screen to be tested at all time nodes are less than the brightness attenuation threshold, then proceed to the subsequent detection steps.

[0013] Furthermore, the color uniformity detection is as follows: Step S341: Obtain the grayscale values of all pixel points within the measurement grid and calculate the grayscale average value of the measurement grid; Step S342: Obtain the grayscale average values of all measurement grids in the first image of the screen to be tested one by one and number all the measurement grids; Step S343: Obtain the maximum grayscale average value ZHP among all measurement grids, and calculate the color difference YSCj between the measurement grid corresponding to the maximum grayscale average value and the grayscale average values CHPj of other measurement grids through the formula. The specific formula is as follows: , where j is the number of the measurement grid, j = 1, 2,..., m, and m is the total number of measurement grids; Step S344: When the color difference value is greater than or equal to the color difference threshold, determine that the abnormal type of the screen to be tested is color uniformity abnormality; When the color difference value is less than the color difference threshold, subsequent detection steps are carried out.

[0014] Furthermore, the step S4 includes the following sub-steps: Step S41: Match the timestamps of the second video and the second video stream; Step S42: Repeat step S2 to obtain the second screen image to be tested of the second real-time video data, and at the same time obtain the second video frame corresponding to the second video at the same timestamp, and convert the second video frame into a second video frame grayscale image; Step S43: Set the measurement grid of the second screen image to be tested, and then obtain the grayscale value HDZk of any pixel point within the measurement grid, where k is the number of pixel points within the measurement grid, k = 1, 2,..., p, and p is the total number of pixel points within the measurement grid, and calculate the grayscale average value HDJl of all pixel points within the measurement grid, where l is the number of the measurement grid, l = 1, 2,..., b, and b is the total number of measurement grids; Step S44: Calculate the local variance JBFl within any measurement grid through the local variance formula. The specific formula is as follows: .

[0015] Furthermore, the step S4 also includes the following sub-steps: Step S45: Repeat steps S43 to S44 to obtain the grayscale average value EHJl of the measurement grid within the second video frame grayscale image, the grayscale value EDJk of any pixel point, and the local variance EBFl of the second video frame grayscale image. Calculate the covariance XFC between the second screen image to be tested and the second video frame grayscale image through the covariance formula. The specific formula is as follows: ; Step S46: Obtain the number A of measurement grids in the second screen image to be tested, A = 1, 2,..., o. Calculate the structural similarity value SSIM between the second screen image to be tested and the second video frame grayscale image through the SSIM formula. The specific formula is as follows: SSIM = (1 / A) × [(2 × HDJl × EHJl) × (2 × XFC × C2)] / [(HDJl 2 + EHJl 2 + C1) × (JBFl 2 + EBFl 2 + C2)], where C1 and C2 are stable constants; Among them, the number of measurement grids in the second screen image to be tested is the same as the number of measurement grids in the second video frame grayscale image; Step S47, if the structural similarity value is greater than the structural similarity threshold, it is determined that the second screen image to be tested and the second video frame grayscale image are similar, and it is determined that the backlight state of the screen to be tested is qualified; If the structural similarity value is less than or equal to the structural similarity threshold, it is determined that the second screen image to be tested and the second video frame grayscale image are not similar, and it is determined that the backlight state of the screen to be tested is unqualified.

[0016] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: 1. The present invention first obtains the aspect ratio of the screen to be tested and the first real-time video data when the screen to be tested plays the first video, and then preprocesses the first real-time video data. The preprocessed first real-time video data is the first screen image to be tested, and the first backlight detection is performed on the first screen image to be tested. According to the first backlight detection result, the abnormal type of the screen to be tested is obtained; 2. The present invention also obtains the second real-time video data when the screen to be tested plays the second video, and performs the second backlight detection on the second video and the second real-time video data. According to the second backlight detection result, the backlight state of the screen to be tested is obtained. Finally, the first backlight detection result and the second backlight detection result of the screen to be tested are visually displayed. The present invention realizes the automatic detection of the backlight of the screen while improving the detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] For the convenience of those skilled in the art to understand, the present invention will be further described below with reference to the drawings.

[0018] Figure 1 It is the overall system block diagram of the present invention; Figure 2 It is the detection flow chart of the first screen image to be tested in the present invention; Figure 3 It is the structural schematic diagram of the electronic device in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0020] Embodiment 1: Please refer to Figure 1 - Figure 2As shown in the figure, the technical solution provided by the present invention is: an on-line detection method for a backlight based on an optical device. This method is used to detect the backlight of an LED screen, and the specific method is as follows: Step S1, obtain the aspect ratio of the screen to be tested and the first real-time video data when the screen to be tested plays the first video through an optical device; Actually, the optical device can be a camera. Among them, the first real-time video data is the first video stream and the first video stream frame rate of the first video stream; the first video stream collected by the optical device is the video stream captured by the optical device when the screen to be tested plays the first video. The first video can be a white source video, a black source video, and a gray source video. The source video pixels of the white source video are (255, 255, 255), the source video pixels of the black source video are (0, 0, 0), and the source video pixels of the gray source video are (128, 128, 128). In this embodiment, the white source video is preferentially analyzed; It should be specifically noted that common backlight devices include direct-lit LED backlights, edge-lit LED backlights, CCFL backlights, OLED backlights, etc. In this embodiment, the direct-lit LED backlight is detected.

[0021] Step S2, preprocess the first real-time video data, and preprocess to obtain the first image of the screen to be tested of the first real-time video data; In this embodiment, the step S2 includes the following sub-steps: Step S21, decompose the first video stream in the first real-time video data into first video stream frames, specifically: When the first video stream frame rate of the first video stream is greater than or equal to 30 frames per second, decompose the first video stream one by one to obtain first video stream images, and randomly select 15 first video stream images as the first video stream frames of the screen to be tested; when the first video stream frame rate of the first video stream is less than 30 frames per second, decompose the first video stream one by one to obtain the first video stream frames of the screen to be tested; Step S22, denoise the first video stream frames through median filtering technology; It should be specifically noted that the first video stream image can be denoised by median filtering, mean filtering or Gaussian filtering. In this embodiment, the direct-lit LED backlight is used for backlight detection, and information such as the brightness and color of the LED screen needs to be detected. Therefore, median filtering is used to denoise the first video stream image; Step S23, convert the first video stream frame into a first video stream frame grayscale image through a formula, and the formula is as follows: I(x, y) = 0.3R(x, y) + 0.59G(x, y) + 0.11B(x, y), where (x, y) is the coordinate of any pixel point in the grayscale image of the first video stream frame; Step S24, calculate the horizontal gradient Ix and vertical gradient Iy of any pixel point in the grayscale image of the first video stream frame through the formula. The specific formula is as follows: Ix = ∂ I / ∂ x; Iy = ∂ I / ∂ y; Step S25, integrate the calculated horizontal gradients and vertical gradients corresponding to all pixel points to obtain the gradient matrix M of the grayscale image of the first video stream frame. The gradient matrix is specifically as follows: ; Step S26, calculate the corner response value JDX of the grayscale image of the first video stream frame through the Harris corner response function formula. The specific formula is as follows: JDX = det(M) - k × [trace(M)], where k is an empirical coefficient. In this embodiment, k = 0.05; Step S27, when the corner response value is greater than the corner response threshold, determine that the corresponding pixel point is a corner; When the corner response value is less than or equal to the corner response threshold, do nothing; Step S28, use the convex hull boundary calculation method to obtain the polygon area in the grayscale image of the first video stream frame; It should be specifically noted that the convex hull boundary calculation method is a prior art; Step S29, repeat the above steps to obtain the polygon areas in the grayscale images corresponding to all the first video stream frames, and analyze the polygon areas to obtain the first test screen image of the screen to be tested. The analysis process of the polygon area is specifically as follows: Step S291, construct the minimum bounding rectangle of the polygon area in the grayscale image of the first video stream frame, and obtain the rectangular vertex coordinates of the minimum bounding rectangle; Step S292, calculate the length and width of the minimum bounding rectangle according to the rectangular vertex coordinates of the minimum bounding rectangle, and divide the length and width of the minimum bounding rectangle to obtain the aspect ratio of the minimum bounding rectangle; Step S293, obtain the aspect ratio of the screen to be tested, subtract the aspect ratio of the minimum bounding rectangle corresponding to all the grayscale images of the first video stream frames from the aspect ratio of the screen to be tested one by one, and take the absolute value to obtain the aspect ratio difference between the minimum bounding rectangle of the grayscale image of the first video stream frame and the screen to be tested; Step S294, use the minimum circumscribed rectangle corresponding to the minimum aspect ratio difference as the first screen image to be tested of the first real-time video data; Specifically, the first screen image to be tested is a grayscale image.

[0022] Step S3, perform a first backlight detection on the first screen image to be tested, and obtain the abnormal type of the screen to be tested according to the first backlight detection result; In this embodiment, step S3 includes the following sub-steps: Step S31, construct a measurement grid for the first screen image to be tested; Among them, the size of the measurement grid of the first screen image to be tested can be 3×3 or 5×5. In this embodiment, the size of the measurement grid is 5×5; Step S32, perform a brightness uniformity detection on the first screen image to be tested. The brightness uniformity detection is as follows: Step S321, obtain the pixel brightness values of all pixel points within the measurement grid; In this embodiment, use the grayscale value of each pixel point as the pixel brightness value of the corresponding pixel point; Step S322, obtain the maximum pixel brightness value and the minimum pixel brightness value of all pixel points within the measurement grid, divide the minimum pixel brightness value by the maximum pixel brightness value to obtain the brightness uniformity value of the measurement grid; Step S323, if the brightness uniformity value of the measurement grid is less than the brightness uniformity threshold, determine that the abnormal type of the screen to be tested is brightness uniformity abnormality; If the brightness uniformity value of the measurement grid is greater than or equal to the brightness uniformity threshold, proceed to the next step; Step S33, perform a brightness stability detection on the first screen image to be tested. The brightness stability detection is as follows: Step S331, the screen to be tested plays a first video for a first fixed duration, takes the first screen image to be tested at intervals of a second fixed duration, and uses the first screen image taken when starting to play the first video as the initial grayscale image; Among them, the first fixed duration is greater than the second fixed duration, and the second fixed duration is greater than zero; Step S332, obtain the initial average brightness CPL of the initial grayscale image; Step S333, obtain the node average brightness JPLi corresponding to the first screen image to be tested at any time node, where i is the number of the time node, i = 1, 2,..., n, n is a positive integer, and the brightness attenuation value SJZi of the first screen image to be tested is calculated through the formula. The specific formula is as follows: SJZi = (CPL - JPLi) / CPL; Step S334: When the brightness attenuation value of the first screen image to be tested at any time node is greater than or equal to the brightness attenuation threshold, determine that the abnormal type of the screen to be tested is brightness stability abnormality; When the brightness attenuation values of the first screen images to be tested at all time nodes are less than the brightness attenuation threshold, proceed to the next step; Step S34: Perform color uniformity detection on the first screen image to be tested. The color uniformity detection is as follows: Step S341: Obtain the gray values of all pixel points within the measurement grid and calculate the gray average value of the measurement grid; Step S342: Sequentially obtain the gray average values of all measurement grids in the first screen image to be tested and number all the measurement grids; Step S343: Obtain the maximum gray average value ZHP among all measurement grids, and calculate the color difference YSCj between the measurement grid corresponding to the maximum gray average value and the gray average values CHPj of other measurement grids through the formula. The specific formula is as follows: , where j is the number of the measurement grid, j = 1, 2,..., m, and m is the total number of measurement grids; It should be specifically noted that the source video played on the screen to be tested is the first video. Therefore, the first screen image to be tested is a pure white image. At this time, the gray value of the pure white image should be 255. When the maximum gray average value is 255, it indicates that there is no abnormality in the corresponding measurement grid; Step S344: When the color difference is greater than or equal to the color difference threshold, determine that the abnormal type of the screen to be tested is color uniformity abnormality; When the color difference is less than the color difference threshold, proceed to the next step; Step S35: The screen to be tested plays the black source video and the gray source video respectively, and repeat steps S31 to S34; If there is no abnormal type in the screen to be tested, proceed to step S4; If there is any abnormal type in the screen to be tested, stop the detection work of the first backlight.

[0023] Step S4: Obtain the second real-time video data when the screen to be tested plays the second video through an optical device, and perform second backlight detection on the second video and the second real-time video data. Obtain the backlight state of the screen to be tested based on the second backlight detection result; It should be specifically noted that the second video is specifically the test source video for detecting the screen to be tested, not a pure color source video; the second real-time video data includes the second video stream and the second video stream frame rate of the second video stream; In this embodiment, the above step S4 includes the following sub-steps: Step S41: Match the timestamps of the second video and the second video stream; Specifically, the timestamp matching is used to make the images of any frame in the second video and the second video stream the same; Step S42: Repeat Step S2 to obtain the second screen image to be tested of the second real-time video data. At the same time, obtain the second video frame corresponding to the second video at the same timestamp, and convert the second video frame into a grayscale image of the second video frame; Step S43: Set the measurement grid of the second screen image to be tested, and then obtain the grayscale value HDZk of any pixel point within the measurement grid, where k is the number of the pixel point within the measurement grid, k = 1, 2, ……, p, and p is the total number of pixel points within the measurement grid. Calculate the average grayscale value HDJl of all pixel points within the measurement grid, where l is the number of the measurement grid, l = 1, 2, ……, b, and b is the total number of measurement grids; Step S44: Calculate the local variance JBFl within any measurement grid through the local variance formula. The specific formula is as follows: ; Step S45: Repeat Step S43 - Step S44 to obtain the average grayscale value EHJl of the pixel points within the measurement grid in the grayscale image of the second video frame, the grayscale value EDJk of any pixel point, and the local variance EBFl of the grayscale image of the second video frame. Calculate the covariance XFC between the second screen image to be tested and the grayscale image of the second video frame through the covariance formula. The specific formula is as follows: ; Step S46: Obtain the number A of measurement grids in the second screen image to be tested, A = 1, 2, ……, o. Calculate the structural similarity value SSIM between the second screen image to be tested and the grayscale image of the second video frame through the SSIM formula. The specific formula is as follows: SSIM = (1 / A) × [(2 × HDJl × EHJl) × (2 × XFC × C2)] / [(HDJl 2 + EHJl 2 + C1) × (JBFl 2 + EBFl 2 + C2)], where C1 and C2 are stable constants, which are specific values used to prevent the mean and variance from being too small and causing excessive calculation errors when the image contrast and brightness are very low. In this embodiment, C1 = 6.5 and C2 = 58.5; Among them, the number of measurement grids in the second screen image to be tested is the same as the number of measurement grids in the grayscale image of the second video frame; Step S47: If the structural similarity value is greater than the structural similarity threshold, it is determined that the second screen image to be tested is similar to the second video frame grayscale image, and it is determined that the backlight state of the screen to be tested is qualified; If the structural similarity value is less than or equal to the structural similarity threshold, it is determined that the second screen image to be tested is not similar to the second video frame grayscale image, and it is determined that the backlight state of the screen to be tested is unqualified.

[0024] Step S5: Visualize the first backlight detection result and the second backlight detection result of the screen to be tested.

[0025] In this application, if there are corresponding calculation formulas, the above calculation formulas are all dimensionless and take their numerical values for calculation. The weight coefficients, proportionality coefficients and other coefficients in the formulas are set to obtain a result value by quantifying each parameter. Regarding the magnitudes of the weight coefficients and proportionality coefficients, as long as they do not affect the proportional relationship between the parameters and the result value.

[0026] Embodiment 2: Figure 3 An example of the structural schematic diagram of an electronic device is shown as Figure 3 shown. The electronic device may include: a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete communication with each other through the communication bus. The processor can call the logical instructions in the memory to execute an online backlight detection method based on an optical device. The method includes: obtaining the aspect ratio of the screen to be tested and the first real-time video data when the screen to be tested plays the first video through the optical device; preprocessing the first real-time video data to obtain the first screen image to be tested of the first real-time video data; performing the first backlight detection on the first screen image to be tested, and obtaining the abnormal type of the screen to be tested according to the first backlight detection result; obtaining the second real-time video data when the screen to be tested plays the second video through the optical device, and performing the second backlight detection on the second video and the second real-time video data, and obtaining the backlight state of the screen to be tested according to the second backlight detection result; visualizing the first backlight detection result and the second backlight detection result of the screen to be tested.

[0027] In addition, when the logical instructions in the above-mentioned memory are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this 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 for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0028] On the other hand, this application also provides a computer program product. The computer program product includes a computer program stored on a computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute an on-line backlight detection method based on an optical device provided by the above-mentioned various methods. The method includes: obtaining the aspect ratio of the screen to be tested and the first real-time video data when the screen to be tested plays a first video through an optical device; preprocessing the first real-time video data, and preprocessing to obtain a first image of the screen to be tested of the first real-time video data; performing a first backlight detection on the first image of the screen to be tested, and obtaining the abnormal type of the screen to be tested according to the first backlight detection result; obtaining the second real-time video data when the screen to be tested plays a second video through the optical device, and performing a second backlight detection on the second video and the second real-time video data, and obtaining the backlight state of the screen to be tested according to the second backlight detection result; visually displaying the first backlight detection result and the second backlight detection result of the screen to be tested.

[0029] In another aspect, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements a backlight online detection method based on an optical device provided above. The method includes: obtaining the aspect ratio of the screen to be tested and the first real-time video data when the screen to be tested plays the first video through the optical device; preprocessing the first real-time video data to obtain a first image of the screen to be tested of the first real-time video data; performing a first backlight detection on the first image of the screen to be tested, and obtaining the abnormal type of the screen to be tested according to the first backlight detection result; obtaining the second real-time video data when the screen to be tested plays the second video through the optical device, and performing a second backlight detection on the second video and the second real-time video data, and obtaining the backlight state of the screen to be tested according to the second backlight detection result; visually displaying the first backlight detection result and the second backlight detection result of the screen to be tested.

[0030] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0031] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0032] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of each embodiment of the present application.

Claims

1. An on-line detection method for a backlight source based on an optical device, characterized in that, The method is as follows: Step S1, obtain the aspect ratio of the screen to be tested and the first real-time video data when the first video is played on the screen to be tested; Step S2, preprocess the first real-time video data, and the preprocessing obtains the first screen image to be tested of the first real-time video data; Step S3, perform the first backlight detection on the first screen image to be tested, and obtain the abnormal type of the screen to be tested according to the first backlight detection result; Step S4, obtain the second real-time video data when the second video is played on the screen to be tested, and perform the second backlight detection on the second video and the second real-time video data, and obtain the backlight state of the screen to be tested according to the second backlight detection result; Step S5, visually display the first backlight detection result and the second backlight detection result of the screen to be tested.

2. The on-line detection method of a backlight source based on an optical device according to claim 1, wherein, The first real-time video data is the first video stream and the first video stream frame rate of the first video stream; The second real-time video data includes the second video stream and the second video stream frame rate of the second video stream.

3. The on-line detection method of a backlight source based on an optical device according to claim 1, characterized in that The said step S2 includes the following sub-steps: Step S21, decompose the first video stream in the first real-time video data into first video stream frames, specifically: When the first video stream frame rate of the first video stream is greater than or equal to 30 frames per second, decompose the first video stream one by one, decompose to obtain first video stream images, and randomly select 15 first video stream images as the first video stream frames of the screen to be tested; When the first video stream frame rate of the first video stream is less than 30 frames per second, decompose the first video stream one by one, decompose to obtain the first video stream frames of the screen to be tested; Step S22, denoise the first video stream frames through the median filtering technology; Step S23, convert the first video stream frames into first video stream frame grayscale images through a formula, and the formula is specifically as follows: I(x, y)=0.3R(x, y)+0.59G(x, y)+0.11B(x, y), where (x, y) is the coordinate of any pixel point in the first video stream frame grayscale image; Step S24, calculate the horizontal direction gradient Ix and the vertical direction gradient Iy of any pixel point in the first video stream frame grayscale image through a formula, and the formula is specifically as follows: Ix= ∂ I / ∂ x; Iy= ∂ I / ∂ y; Step S25, integrate the calculated horizontal direction gradients and vertical direction gradients corresponding to all pixel points, and integrate to obtain the gradient matrix M of the first video stream frame grayscale image, and the gradient matrix is specifically as follows: ； Step S26, calculate the corner response value JDX of the first video stream frame grayscale image, and the formula is specifically as follows: JDX = det(M)-k×[trace(M)], where k is an empirical coefficient; Step S27, when the corner response value is greater than the corner response threshold, determine that the corresponding pixel point is a corner point; When the corner response value is less than or equal to the corner response threshold, do nothing; Step S28, calculate the polygon area in the first video stream frame grayscale image; Step S29, repeat step S28, obtain the polygon areas in the first video stream frame grayscale images corresponding to all the first video stream frames, and analyze the polygon areas, and analyze to obtain the first screen image to be tested of the screen to be tested.

4. The on-line detection method of a backlight source based on an optical device according to claim 3, characterized in that, The analysis process of the polygon area is as follows: Step S291: Construct the minimum bounding rectangle of the polygon area in the grayscale image of the first video stream frame, and obtain the rectangular vertex coordinates of the minimum bounding rectangle; Step S292: Calculate the length and width of the minimum bounding rectangle based on the rectangular vertex coordinates of the minimum bounding rectangle, divide the length and width of the minimum bounding rectangle, and obtain the aspect ratio of the minimum bounding rectangle; Step S293: Obtain the aspect ratio of the screen to be tested, subtract the aspect ratios of the minimum bounding rectangles of all the grayscale images of the first video stream frame from the aspect ratio of the screen to be tested one by one and take the absolute value to obtain the aspect ratio difference between the minimum bounding rectangle of the grayscale image of the first video stream frame and the screen to be tested; Step S294: Use the minimum bounding rectangle corresponding to the minimum aspect ratio difference as the first screen image to be tested of the first real-time video data.

5. A backlight on-line detection method based on an optical device according to claim 3, characterized in that, Step S3 includes the following sub-steps: Step S31: Construct the measurement grid of the first screen image to be tested; Step S32: Perform brightness uniformity detection on the first screen image to be tested; Step S33: Perform brightness stability detection on the first screen image to be tested Step S34: Perform color uniformity detection on the first screen image to be tested; Step S35: Play the black source video and the gray source video on the screen to be tested respectively, and repeat steps S31 to S34; If there is no abnormal type on the screen to be tested, proceed to step S4; If there is any abnormal type on the screen to be tested, stop the detection work of the first backlight.

6. The on-line detection method of a backlight source based on an optical device according to claim 5, characterized in that, The brightness uniformity detection is as follows: Step S321: Obtain the pixel brightness values of all pixel points within the measurement grid; Step S322: Obtain the maximum pixel brightness value and the minimum pixel brightness value of all pixel points within the measurement grid, divide the minimum pixel brightness value by the maximum pixel brightness value, and obtain the brightness uniformity value of the measurement grid; Step S323: If the brightness uniformity value of the measurement grid is less than the brightness uniformity threshold, determine that the abnormal type of the screen to be tested is brightness uniformity abnormality; If the brightness uniformity value of the measurement grid is greater than or equal to the brightness uniformity threshold, proceed to the subsequent detection steps.

7. A backlight on-line detection method based on an optical device according to claim 5, characterized in that, The brightness stability detection is as follows: Step S331: Play the first video for the first fixed duration on the screen to be tested, take the first screen image to be tested captured at intervals of the second fixed duration, and use the first screen image to be tested captured when starting to play the first video as the initial grayscale image; where the first fixed duration is greater than the second fixed duration, and the second fixed duration is greater than zero; Step S332: Obtain the initial average brightness CPL of the initial grayscale image; Step S333: Obtain the node average brightness JPLi corresponding to the first screen image to be tested at any time node, where i is the number of the time node, i = 1, 2,..., n, n is a positive integer, and calculate the brightness attenuation value SJZi of the first screen image to be tested through the formula. The specific formula is as follows: SJZi = (CPL - JPLi) / CPL; Step S334: When the brightness attenuation value of the first screen image to be tested at any time node is greater than or equal to the brightness attenuation threshold, it is determined that the abnormal type of the screen to be tested is brightness stability abnormality; When the brightness attenuation values of the first screen images to be tested at all time nodes are less than the brightness attenuation threshold, the subsequent detection steps are carried out.

8. A backlight on-line detection method based on an optical device according to claim 5, characterized in that, The color uniformity detection is as follows: Step S341: Obtain the gray values of all pixel points within the measurement grid and calculate the gray average value of the measurement grid; Step S342: Sequentially obtain the gray average values of all measurement grids in the first screen image to be tested and number all the measurement grids; Step S343: Obtain the maximum gray average value ZHP among all measurement grids, and calculate the color difference YSCj between the measurement grid corresponding to the maximum gray average value and the gray average value CHPj of other measurement grids through the formula. The specific formula is as follows: , where j is the number of the measurement grid, j = 1, 2, ……, m, and m is the total number of the measurement grids; Step S344: When the color difference is greater than or equal to the color difference threshold, it is determined that the abnormal type of the screen to be tested is color uniformity abnormality; When the color difference is less than the color difference threshold, the subsequent detection steps are carried out.

9. A backlight on-line detection method based on an optical device according to claim 5, characterized in that The step S4 includes the following sub-steps: Step S41: Match the timestamps of the second video and the second video stream; Step S42: Repeat step S2 to obtain the second screen image to be tested of the second real-time video data, and at the same time obtain the second video frame corresponding to the second video at the same timestamp, and convert the second video frame into a second video frame gray image; Step S43: Set the measurement grid of the second screen image to be tested, and then obtain the gray value HDZk of any pixel point within the measurement grid, where k is the number of pixel points within the measurement grid, k = 1, 2,..., p, and p is the total number of pixel points within the measurement grid. Calculate the gray average value HDJl of all pixel points within the measurement grid, where l is the number of the measurement grid, l = 1, 2,..., b, and b is the total number of measurement grids; Step S44: Calculate the local variance JBFl within any measurement grid through the local variance formula. The specific formula is as follows: 。 10. A backlight on-line detection method based on an optical device according to claim 9, characterized in that, The step S4 also includes the following sub-steps: Step S45: Repeat steps S43 to S44 to obtain the gray average value EHJl and the gray value EDJk of any pixel point within the measurement grid in the second video frame gray image, as well as the local variance EBFl of the second video frame gray image. Calculate the covariance XFC between the second screen image to be tested and the second video frame gray image through the covariance formula. The specific formula is as follows: ； Step S46: Obtain the number A of measurement grids in the second screen image to be tested, A = 1, 2,..., o, and calculate the structural similarity value SSIM between the second screen image to be tested and the second video frame gray image through the SSIM formula. The specific formula is as follows: SSIM = (1 / A) × [(2 × HDJl × EHJl) × (2 × XFC × C2)] / [(HDJl 2 + EHJl 2 + C1) × (JBFl 2 + EBFl 2 + C2)], where C1 and C2 are stability constants; Among them, the number of measurement grids in the second screen image to be tested is the same as the number of measurement grids in the second video frame gray image; Step S47: If the structural similarity value is greater than the structural similarity threshold, it is determined that the second screen image to be tested and the second video frame gray image are similar, and it is determined that the backlight state of the screen to be tested is qualified; If the structural similarity value is less than or equal to the structural similarity threshold, it is determined that the second screen image to be tested is not similar to the second video frame grayscale image, and it is determined that the backlight state of the screen to be tested is unqualified.

Images