AI-based analysis-based method for high-definition signal adaptation and mini LED backlight control display

Abstract

This invention discloses a high-definition signal adaptation method and display screen based on AI analysis for mini LED backlight control. Relating to the field of AI analysis technology, this invention improves the viewing angle adaptation accuracy and cross-zone halo suppression coverage dimension during mini LED backlight control. It can adapt to the differentiated light field diffusion response requirements and halo transmission coefficient iterative convergence characteristics under different viewing angle conditions, thereby achieving efficient and accurate viewing angle compensation calculation and halo interference suppression. This significantly improves the image quality output accuracy and multi-viewer display experience reliability in the mini LED display process, effectively preventing insufficient response of fixed threshold parameter settings to actual viewing angle light field diffusion changes, avoiding brightness deviation and halo misjudgment caused by experience-based judgment, and effectively improving the targeting and display effect of the final backlight control matrix generation, ensuring a high degree of fit between backlight control measures and actual multi-viewer viewing angle requirements.

Description

Technical Field

[0001] This invention relates to the field of AI analysis technology, specifically to a method for adjusting the backlight of a mini LED based on high-definition signals using AI analysis, and a display screen. Background Technology

[0002] As a core technology of the next generation of high-end display systems, mini LED backlight display technology requires precise adaptive control of backlight zone brightness output and differentiated suppression analysis of cross-zone halo interference under dynamic viewing angles of multiple viewers. Therefore, it is essential to construct an intelligent backlight control system based on a prediction model of the spatial distribution characteristics of the light field from the viewing angle.

[0003] In existing technologies, mini LED backlight control relies on fixed zone dimming strategies and single image content analysis modes, which cannot cover the dynamic changes in the spatial distribution characteristics of the light field and the evolution of the spatial distribution of the halo transfer coefficient under different viewing angles. Although existing technologies have introduced basic local dimming mechanisms and image brightness analysis functions, due to the single setting of their control parameters and their high dependence on empirical threshold configuration, it is difficult to accurately judge the differences in viewing angles of multiple viewers, the matching of personalized light flux spatial distribution, and the risk of halo crosstalk between backlight zones. When faced with complex viewing conditions of multiple viewers or uneven distribution of backlight demand caused by high dynamic range image content, problems such as insufficient viewing angle compensation, poor halo suppression effect, and conflict between single zone control and cross-zone collaborative control are likely to occur.

[0004] Furthermore, existing technologies lack a mechanism for assessing and predicting the differential impact of differences in the spatial distribution characteristics of light fields from different viewing angles and the backlight zone halo transmission link during the control process. This makes it impossible to effectively optimize backlight brightness compensation and accurately identify personalized color reproduction based on the viewer's viewing position. In addition, there are defects such as weak halo attenuation index inversion modeling capability, insufficient multi-viewer viewing angle fusion calculation, lack of coupling analysis of viewing angle influence and image content influence, improper handling of backlight zone halo interference boundary, and low convergence accuracy of iterative correction. As a result, high-precision backlight control matrix cannot be effectively generated and its quality cannot be guaranteed. At the same time, there is a lack of adaptive brightness correction strategies for different backlight zone halo interference levels, and the backlight output scheme cannot be adaptively adjusted according to dynamic changes in halo suppression parameters, viewing angle compensation coefficients, etc. Summary of the Invention

[0005] The purpose of this invention is to provide a method for adjusting the backlight of a mini LED based on high-definition signal adaptation using AI analysis, and to provide a display screen that solves the problems existing in the background technology.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The first aspect of the present invention provides a high-definition signal adaptation mini LED backlight control method based on AI analysis, including: Step 1: Before the mini LED leaves the factory, the diffusion layer thickness parameters, scattering particle size parameters and concentration parameters of the mini LED backlight module are pre-collected, and multi-scene optical simulation modeling is performed accordingly, so as to pre-construct the light field spatial distribution characteristics and halo transmission attenuation model of the mini LED backlight module under various viewing angles.

[0007] Step 2: While each viewer is watching the mini LED, the high-definition image signal to be displayed is acquired in real time, and the spatial position and eye angle of each viewer relative to the display screen are identified by the visual acquisition device, thereby obtaining the visual reception area characteristic parameters of each viewer.

[0008] Step 3: Perform content AI feature analysis on the high-definition image signal to extract the brightness level distribution features and color saturation distribution features of the high-definition image signal. Based on the light field spatial distribution features and halo transmission attenuation model of the mini LED backlight module under various viewing angles, and combined with the visual reception area feature parameters of each viewer, perform fusion inference calculation to obtain the peak brightness requirement value, color reproduction accuracy requirement coefficient, viewing angle compensation coefficient, and halo suppression parameters of each backlight zone of the mini LED.

[0009] Step 4: Generate a mini LED backlight control matrix based on the peak brightness requirement, color reproduction accuracy requirement coefficient, viewing angle compensation coefficient, and halo suppression parameters of each backlight zone of the mini LED.

[0010] Step 5: Convert the mini LED backlight control matrix into a driving signal to drive the mini LED display to complete the brightness output in real time.

[0011] A second aspect of the present invention provides a high-definition signal adaptation mini LED backlight control display screen based on AI analysis, including a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the high-definition signal adaptation mini LED backlight control method based on AI analysis as described above.

[0012] The beneficial effects of this invention are as follows: This invention improves the viewing angle adaptation accuracy and cross-zone halo suppression coverage dimension in the mini LED backlight control process. It can adapt to the differentiated light field diffusion response requirements and halo transmission coefficient iterative convergence characteristics under different viewing angle conditions, thereby achieving efficient and accurate viewing angle compensation calculation and halo interference suppression. It significantly improves the image quality output accuracy and multi-viewer display experience reliability in the mini LED display process. It can dynamically adjust the backlight zone brightness correction value and halo suppression parameter calculation strategy according to the difference in halo attenuation index and effective halo distance between different viewers. It effectively prevents the fixed threshold parameter setting from being insufficient in response to the actual viewing angle light field diffusion changes, avoids brightness deviation and halo misjudgment caused by experience judgment, realizes intelligent allocation and differentiated output of backlight resources, effectively improves the pertinence and display effect of the final backlight control matrix generation, determines the backlight zones that affect the display image quality and their halo interference levels, and ensures that the backlight control measures are highly consistent with the actual multi-viewer viewing angle requirements. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, 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 the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] Reference Figure 1 As shown, the first aspect of the present invention provides a method for high-definition signal adaptation and mini LED backlight control based on AI analysis, including: Step 1: Before the mini LED leaves the factory, the diffusion layer thickness parameters, scattering particle size parameters and concentration parameters of the mini LED backlight module are pre-collected, and multi-scene optical simulation modeling is performed accordingly, so as to pre-construct the light field spatial distribution characteristics and halo transmission attenuation model of the mini LED backlight module under various viewing angles.

[0017] In one specific embodiment, the diffusion layer thickness parameters, scattering particle size parameters, and concentration parameters of the mini LED backlight module are pre-collected. The specific method is as follows: the diffusion layer thickness parameters of the mini LED backlight module are collected at the factory through non-contact optical thickness measurement; the scattering particle size parameters of the mini LED backlight module are collected through SEM (scanning electron microscope); and the concentration parameters of the mini LED backlight module are collected through thermogravimetric analysis.

[0018] In a specific embodiment of the present invention, a model of the spatial distribution characteristics of the light field and the halo transmission attenuation of the mini LED backlight module under various viewing angles is pre-constructed. The specific method is as follows: based on the diffusion layer thickness parameters, scattering particle size parameters and concentration parameters of the mini LED backlight module, a Monte Carlo ray tracing simulation model of the mini LED backlight module is constructed.

[0019] In one specific embodiment, a Monte Carlo ray tracing simulation model of the mini LED backlight module is constructed by: inputting the diffusion layer thickness parameters, scattering particle size parameters, and concentration parameters of the mini LED backlight module into the existing LightTools software, and obtaining the Monte Carlo ray tracing simulation model through simulation.

[0020] For each viewing angle, the spatial position of the detector array corresponding to the viewing angle is set in the Monte Carlo ray tracing simulation model, and photon transmission path tracing calculation is performed to obtain the spatial distribution data of luminous flux of mini LED light source under each viewing angle, and use it as the spatial distribution characteristics of light field of mini LED backlight module under each viewing angle.

[0021] In one specific embodiment, photon propagation path tracing calculations are performed to obtain the spatial distribution data of luminous flux of the mini LED light source under various viewing angles. The specific method is as follows: For each viewing angle, a virtual detector array is set up in the Monte Carlo ray tracing simulation model. The detectors are arranged on a sphere or plane with a specific viewing angle to the light source. A large number of photons are emitted from the LED light source position. Each photon carries initial energy and propagation direction. When the photon propagates in the diffusion layer, the free path length is calculated according to the scattering cross section and concentration parameters of the scattering particles to determine the collision position between the photon and the scattering particles. When the photon is scattered, a new propagation direction and energy loss are randomly determined according to the scattering phase function. The complete path of each photon is traced until it is absorbed or leaves the simulation area. The number of photons reaching each detector position is counted and used as the spatial distribution data of luminous flux under each viewing angle.

[0022] Based on the spatial distribution data of luminous flux of the mini LED backlight module under various viewing angles, an exponential function is fitted to obtain the halo attenuation index and effective halo distance of the mini LED backlight module under various viewing angles, and these are used as the halo transmission attenuation model of the mini LED backlight module under various viewing angles.

[0023] In one specific embodiment, an exponential function fitting is performed to obtain the halo attenuation index and effective halo distance of the mini LED backlight module under various viewing angles. The specific method is as follows: extract the luminous flux values ​​at different radial distances from the center of the light source from the luminous flux spatial distribution data under various viewing angles, normalize all luminous flux values ​​by dividing them by the peak luminous flux to obtain normalized luminous intensity distribution data, fit the data points using the least squares method, adjust the attenuation coefficient parameter to minimize the sum of squared errors between the fitted curve and the actual data points, and the attenuation coefficient obtained after fitting optimization is the halo attenuation index. Calculate the radial distance at which the normalized luminous intensity attenuates to a preset threshold based on the fitted curve. This distance is the effective halo distance, representing the effective range of the halo's influence.

[0024] Step 2: While each viewer is watching the mini LED, the high-definition image signal to be displayed is acquired in real time, and the spatial position and eye angle of each viewer relative to the display screen are identified by the visual acquisition device, thereby obtaining the visual reception area characteristic parameters of each viewer.

[0025] In one specific embodiment, the high-definition image signal to be displayed is acquired in real time, and the specific method is as follows: the high-definition image signal to be displayed is acquired through a miniLED terminal.

[0026] In a specific embodiment of the present invention, the spatial position and eye angle of each viewer relative to the display screen are identified to obtain the visual reception area feature parameters of each viewer. The specific method is as follows: acquire the facial image and depth image of each viewer, use the facial key point detection algorithm to extract the center coordinates of the pupils of each viewer's eyes, and calculate the horizontal offset distance and vertical offset distance of each viewer relative to the center of the mini LED display screen, and use them as the spatial position of each viewer relative to the display screen.

[0027] In one specific embodiment, facial and depth images of each viewer are acquired, and the center coordinates of the pupils of each viewer are extracted using a facial landmark detection algorithm. The specific method is as follows: facial and depth images of each viewer are acquired by an ultra-high-definition camera installed on a mini LED display screen. Existing facial landmark detection algorithms are relatively mature, and the center coordinates of the pupils of each viewer can be extracted using this algorithm.

[0028] In one specific embodiment, the horizontal and vertical offset distances of each viewer relative to the center of the mini LED display screen are calculated. The specific method is as follows: the coordinates of the midpoints of the viewer's eyes are transformed into a coordinate system with the center of the display screen as the origin. The difference between the horizontal coordinates of the midpoints of the viewer's eyes and the horizontal coordinates of the center of the display screen is the horizontal offset distance, and the difference between the vertical coordinates and the vertical coordinates of the center of the display screen is the vertical offset distance. A positive horizontal offset distance indicates that the viewer is located on the right side of the screen, and a negative horizontal offset distance indicates that the viewer is located on the left side. A positive vertical offset distance indicates that the viewer is located above the screen, and a negative vertical offset distance indicates that the viewer is located below the screen.

[0029] The physical size parameters of the mini LED display screen are obtained from the local database. Based on the coordinates of the center of each viewer's pupils and the spatial position of each viewer relative to the display screen, the angle between the midpoint of the line connecting each viewer's eyes and the normal direction of the display screen is calculated and used as the viewing angle of each viewer.

[0030] In one specific embodiment, the angle between the midpoint of the line connecting each viewer's eyes and the normal direction of the display screen is calculated. The specific method is as follows: based on the spatial position of each viewer relative to the display screen and the center coordinates of the mini LED display screen, the direction vector from the midpoint of the line connecting the eyes to the center of the display screen is calculated. The dot product of the direction vector from the midpoint of the line connecting the eyes to the center of the screen and the normal direction vector of the display screen is calculated. After normalizing the two vectors to unit vectors, the angle value is obtained by calculating the inverse cosine function. This angle value reflects the degree to which the viewer's line of sight deviates from the front of the screen. An angle of zero degrees indicates a frontal view, and the larger the angle, the more oblique the viewing angle.

[0031] In one specific embodiment, the azimuth angle between the viewing direction of each viewer and the horizontal central axis of the display screen is calculated as follows: Based on the spatial coordinates of the midpoint of the line connecting each viewer's eyes and the coordinates of the center of the display screen, a spatial vector is calculated pointing from the center of the screen to the midpoint of the line connecting the eyes. This spatial vector is then projected onto the display screen plane to obtain a projection vector. The angle between the projection vector and the positive direction of the horizontal central axis is calculated. The angle value obtained is the azimuth angle. An azimuth angle of zero degrees indicates that the viewer is directly in front of the screen, a positive value indicates that the viewer is biased to one side, and a negative value indicates that the viewer is biased to the other side. The specific direction is determined according to the coordinate system definition.

[0032] The eye angle of each viewer and the spatial position of each viewer relative to the display screen are used as the visual reception area feature parameters of each viewer.

[0033] Step 3: Perform content AI feature analysis on the high-definition image signal to extract the brightness level distribution features and color saturation distribution features of the high-definition image signal. Based on the light field spatial distribution features and halo transmission attenuation model of the mini LED backlight module under various viewing angles, and combined with the visual reception area feature parameters of each viewer, perform fusion inference calculation to obtain the peak brightness requirement value, color reproduction accuracy requirement coefficient, viewing angle compensation coefficient, and halo suppression parameters of each backlight zone of the mini LED.

[0034] In a specific embodiment of the present invention, the brightness level distribution features and color saturation distribution features of the high-definition image signal are extracted. The specific method is as follows: obtain a pre-trained image semantic segmentation AI model from a local database, input the high-definition image signal into the image semantic segmentation AI model, identify the range interval and category label of each semantic region in the high-definition image signal, obtain the visual attention weight value corresponding to each semantic category from the local database, and thus generate a visual attention spatial distribution map of the high-definition image signal.

[0035] It should be noted that the pre-trained image semantic segmentation AI model is based on a deep convolutional neural network architecture and is trained on a large-scale labeled dataset. By learning the mapping relationship between low-level image features and high-level semantics, the model can perform pixel-by-pixel semantic classification prediction of the input image. The model output is a semantic segmentation map with the same size as the input image, and each pixel position is labeled with the corresponding category label.

[0036] In one specific embodiment, the method for generating a visual attention spatial distribution map of a high-definition image signal is as follows: obtain the visual attention weight values ​​corresponding to each semantic category from a local database, traverse each pixel position in the image, find the corresponding visual attention weight value according to the semantic category label of the position, assign the visual attention weight value to the pixel position, and combine the visual attention weight values ​​of all pixels into a two-dimensional matrix with the same size as the high-definition image signal. This matrix is ​​the visual attention spatial distribution map.

[0037] It should be noted that the visual attention weight values ​​are preset based on human visual attention habits. For example, the weight value for the human face category is 0.9, the weight value for the text area category is 0.8, and the weight value for the building category is 0.6. These values ​​can be set by staff.

[0038] The high-definition image signal is converted to YUV color space to extract the luminance component value of each pixel position. Combined with the visual attention spatial distribution map of the high-definition image signal, the luminance component value of each pixel position is evaluated and used as the luminance hierarchical distribution feature of the high-definition image signal.

[0039] In one specific embodiment, the method for evaluating the adjusted brightness component value of the high-definition image signal at each pixel position is as follows: based on the visual attention spatial distribution map of the high-definition image signal, the visual attention weight value of each pixel position in the high-definition image signal is extracted, and multiplied by the brightness component value to obtain the adjusted brightness component value of the high-definition image signal at each pixel position.

[0040] The high-definition image signal is converted to HSV color space to extract the saturation component values ​​of each pixel position in the high-definition image signal. Based on this, the mean saturation value and saturation variance value of each semantic region in the high-definition image signal are statistically analyzed and used as the color saturation distribution characteristics of the high-definition image signal.

[0041] In one specific embodiment, the mean and variance of saturation of each semantic region in the high-definition image signal are statistically analyzed. The specific method is as follows: the mean and variance of saturation of each semantic region in the high-definition image signal can be obtained by using existing methods for calculating the mean and variance.

[0042] In a specific embodiment of the present invention, fusion inference calculation is performed to obtain the peak brightness requirement value, color reproduction accuracy requirement coefficient, viewing angle compensation coefficient and halo suppression parameter of each backlight zone of the mini LED. The specific method is to obtain the zone boundary coordinates of each backlight zone of the mini LED from the local database.

[0043] Based on the visual reception area characteristic parameters of each viewer, the spatial distribution characteristics of the light field of the mini LED backlight module under each viewing angle and the halo transmission attenuation model are mapped to obtain the spatial distribution data of luminous flux, halo attenuation index and effective halo distance of each viewer under the corresponding viewing angle.

[0044] In one specific embodiment, the spatial distribution data of luminous flux, halo attenuation index, and effective halo range for each viewer under the corresponding viewing angle are mapped. The specific method is as follows: based on the viewing angle of each viewer and the spatial position of each viewer relative to the display screen, the data are substituted into the pre-constructed model of the spatial distribution characteristics of the light field and the halo transmission attenuation of the mini LED backlight module under each viewing angle. If the viewing angle of the viewer is completely matched with a certain viewing angle in the model, the spatial distribution data of luminous flux, halo attenuation index, and effective halo range corresponding to that viewing angle are directly extracted. If they are not completely matched, the two viewing angles that are closest to the viewing angle of the viewer are found, and the spatial distribution data of luminous flux, halo attenuation index, and effective halo range are calculated using a linear interpolation method, thereby obtaining the spatial distribution data of luminous flux, halo attenuation index, and effective halo range for each viewer under the corresponding viewing angle.

[0045] Based on the brightness level distribution characteristics and color saturation distribution characteristics of high-definition image signals, and combined with the boundary coordinates of each backlight zone of mini LED, the peak brightness requirement and color reproduction accuracy requirement coefficient of high-definition image signals in each backlight zone of mini LED are calculated.

[0046] Based on a comprehensive analysis of the spatial distribution data of luminous flux, halo attenuation index, and effective halo distance for each viewer at their respective viewing angle, the viewing angle compensation coefficients for each backlight zone of the mini LED are obtained.

[0047] In one specific embodiment, a comprehensive analysis is performed to obtain the viewing angle compensation coefficients for each backlight zone of the mini LED. The specific method is as follows: The halo attenuation index of all viewers is arithmetically averaged to obtain the average halo attenuation index; the effective halo range of all viewers is arithmetically averaged to obtain the average effective halo range; the average halo attenuation index is divided by the average effective halo range to obtain the viewing angle attenuation ratio. Based on the spatial distribution data of luminous flux for each viewer, the number of photons reaching each viewer is extracted and arithmetically averaged to obtain the average photon count; after normalizing the average photon count to eliminate units, it is multiplied by the viewing angle attenuation ratio to obtain the viewing angle compensation coefficient.

[0048] Based on the halo attenuation index and effective halo distance of each viewer at their respective viewing angle, and combined with the boundary coordinates of each backlight zone of the miniLED, the target halo transmission coefficient between each backlight zone of the miniLED and its surrounding backlight zones is evaluated. The halo interference estimate of each backlight zone of the miniLED affected by the surrounding backlight zones is accumulated and used as the halo suppression parameter of each backlight zone of the miniLED.

[0049] In a specific embodiment of the present invention, the peak brightness requirement value and color reproduction accuracy requirement coefficient of the high-definition image signal in each backlight zone of the mini LED are calculated. The specific method is as follows: based on the brightness level distribution characteristics of the high-definition image signal, the adjusted brightness component value of each pixel position within the partition boundary coordinate range of each backlight zone of the mini LED is extracted, the maximum value of the adjusted brightness component value of the pixel position within the partition boundary coordinate range of each backlight zone of the mini LED is selected, and it is used as the peak brightness requirement value of the high-definition image signal in each backlight zone of the mini LED.

[0050] In one specific embodiment, the adjustable brightness component value of each pixel position within the partition boundary coordinate range of each backlight partition of the mini LED is extracted. The specific method is as follows: the high-definition image signal is placed in the mini LED proportionally to obtain the display screen of the high-definition image signal in each backlight partition of the mini LED, thereby obtaining the adjustable brightness component value of each pixel position within the partition boundary coordinate range of each backlight partition of the mini LED.

[0051] Based on the color saturation distribution characteristics of high-definition image signals, the mean and variance values ​​of saturation in each semantic region within the boundary coordinate range of each backlight zone of the mini LED are extracted. The mapping relationship curve between the mean and variance values ​​of saturation and the color reproduction accuracy requirement coefficient is obtained from the local database. The color reproduction accuracy requirement coefficient of each semantic region in each backlight zone of the mini LED is obtained by mapping. The maximum value of the color reproduction accuracy requirement coefficient of the semantic region in the backlight zone is taken as the color reproduction accuracy requirement coefficient of the high-definition image signal in each backlight zone of the mini LED.

[0052] In one specific embodiment, the mean saturation and variance of saturation of each semantic region within the partition boundary coordinate range of each backlight partition of the mini LED are extracted. The specific method is as follows: based on the method of extracting the adjusted brightness component value of each pixel position within the partition boundary coordinate range of each backlight partition of the mini LED, the mean saturation and variance of saturation of each semantic region within the partition boundary coordinate range of each backlight partition of the mini LED are obtained in the same way.

[0053] It should be noted that the mapping curve between the mean saturation value, the variance saturation value, and the color reproduction accuracy requirement coefficient is an empirical curve established based on color display theory and a large amount of display effect test data. When the mean saturation value is high, it means that the color in this area is vivid, requiring higher backlight brightness to accurately reproduce the color, and the color reproduction accuracy requirement coefficient is larger. When the variance saturation value is large, it means that the color variation in this area is rich, requiring more precise backlight control, and the color reproduction accuracy requirement coefficient increases.

[0054] In a specific embodiment of the present invention, the target halo transmission coefficient between each backlight zone of the mini LED and its surrounding backlight zones is evaluated by means of: calculating the center coordinates of each backlight zone of the mini LED based on the zone boundary coordinates of each backlight zone of the mini LED, and calculating the Euclidean distance between each backlight zone of the mini LED and its surrounding backlight zones accordingly.

[0055] In one specific embodiment, the center coordinates of each backlight zone of the mini LED are calculated, and the Euclidean distance between each backlight zone of the mini LED and its surrounding backlight zones is calculated accordingly. The specific method is as follows: the average value of the boundary coordinates of each backlight zone of the mini LED is calculated to obtain the center coordinates of each backlight zone of the mini LED. The absolute value of the difference between the center coordinates of each backlight zone of the mini LED and its surrounding backlight zones is then calculated to obtain the Euclidean distance between each backlight zone of the mini LED and its surrounding backlight zones.

[0056] Based on the halo attenuation index and effective halo distance of each viewer at their respective viewing angles, and combined with the Euclidean distance between each backlight zone of the mini LED and its surrounding backlight zones, the halo transmission coefficient between each backlight zone of the mini LED and its surrounding backlight zones at their respective viewing angles is analyzed.

[0057] In one specific embodiment, the halo transmission coefficient between each backlight zone of the mini LED and its surrounding backlight zones under each viewer's corresponding viewing angle is analyzed. The specific method is as follows: based on the halo attenuation index of each viewer under the corresponding viewing angle... And the effective range of the halo Where x represents the number of each viewer. y is a positive integer greater than 2, and is combined with the Euclidean distance between each backlight zone of the mini LED and its surrounding backlight zones. Where n represents the number of each backlight zone, m is a positive integer greater than 2, and i represents the number of each surrounding backlight zone. j is a positive integer greater than 2.

[0058] Calculate the halo transmission coefficient between each backlight zone of the mini LED and its surrounding backlight zones at the corresponding viewing angle for each viewer. , where e is the natural constant.

[0059] It should be noted that the halo transmission coefficient describes the optical crosstalk phenomenon between backlight zones. That is, when a backlight zone emits light, due to the scattering effect of the diffusion layer, some light will diffuse into the display area of ​​the surrounding adjacent backlight zones, causing halo interference. Based on the Euclidean distance between each backlight zone of the mini LED and its surrounding backlight zones, this distance represents the physical distance between the centers of two backlight zones. When observing the same backlight module from different viewing angles, the diffusion characteristics of the light field differ due to the different viewing angles. Based on the halo attenuation index and the effective halo distance at each viewer's corresponding viewing angle, these two parameters describe the law of halo attenuation with physical distance at that viewing angle. Using an exponential attenuation model, the halo transmission coefficient is equal to the negative exponent of the natural constant. The exponent is the product of the halo attenuation index and the ratio of the Euclidean distance value to the effective halo distance. For each viewer, the halo transmission coefficient between each backlight zone and its surrounding backlight zones at that viewer's viewing angle is calculated using the attenuation parameters at their corresponding viewing angle.

[0060] The halo transmission coefficients between each backlight zone of the mini LED and its surrounding backlight zones are averaged for each viewer at their respective viewing angles, thereby obtaining the target halo transmission coefficients between each backlight zone of the mini LED and its surrounding backlight zones.

[0061] Step 4: Generate a mini LED backlight control matrix based on the peak brightness requirement, color reproduction accuracy requirement coefficient, viewing angle compensation coefficient, and halo suppression parameters of each backlight zone of the mini LED.

[0062] In a specific embodiment of the present invention, the method for generating a mini LED backlight control matrix is ​​as follows: based on the peak brightness requirement, color reproduction accuracy requirement coefficient, and viewing angle compensation coefficient of each backlight zone of the mini LED, the initial target brightness value and initial color saturation of each backlight zone of the mini LED are calculated.

[0063] In one specific embodiment, the initial target brightness value and initial color saturation of each backlight zone of the mini LED are calculated as follows: based on the peak brightness requirement value of each backlight zone of the mini LED, the peak brightness requirement value is multiplied by the viewing angle compensation coefficient of each backlight zone of the mini LED to obtain the initial target brightness value of each backlight zone of the mini LED; based on the high-definition image signal, the ideal saturation of each pixel of the high-definition image signal in each backlight zone of the mini LED is obtained; this is multiplied by the color reproduction accuracy requirement coefficient to obtain the reproduction saturation of each pixel of the high-definition image signal in each backlight zone of the mini LED; and the initial color saturation is obtained by averaging.

[0064] Based on the halo suppression parameters of each backlight zone of the mini LED and the target halo transmission coefficient between each backlight zone of the mini LED and its surrounding backlight zones, the initial target brightness value of each backlight zone of the mini LED is iteratively corrected for halo suppression, thereby obtaining the viewing angle brightness correction value of each backlight zone of the mini LED.

[0065] Similarly, the viewing angle saturation correction values ​​for each backlight zone of the mini LED are obtained.

[0066] The viewing angle brightness correction value and viewing angle saturation correction value of each backlight zone of the mini LED are arranged and combined in a matrix according to the zone boundary coordinates of each backlight zone of the mini LED to generate the mini LED backlight control matrix.

[0067] In a specific embodiment of the present invention, halo suppression iterative correction is performed to obtain the viewing angle brightness correction value of each backlight zone of the mini LED. The specific method is as follows: obtain the halo suppression iteration number threshold and halo error convergence threshold from the local database, initialize the current iteration brightness value of each backlight zone of the mini LED to the initial target brightness value of each backlight zone of the mini LED, and initialize the current iteration number to 0.

[0068] Based on the target halo transmission coefficient between each backlight zone of the mini LED and its surrounding backlight zones, and the current iteration brightness value of each backlight zone of the mini LED, the total halo superposition received by each backlight zone of the mini LED in the current iteration state is calculated, and the difference is calculated with the halo suppression parameter of each backlight zone of the mini LED, so as to obtain the current halo estimated reduction value of each backlight zone of the mini LED.

[0069] In one specific embodiment, the total halo superposition received by each backlight partition of the mini LED in the current iteration state is calculated as follows: For a certain backlight partition, all its surrounding adjacent backlight partitions are traversed, and the current iteration brightness value of each surrounding partition is multiplied by the target halo transfer coefficient between the surrounding partition and the target partition to obtain the halo superposition component generated by the surrounding partition on the target partition in the current iteration state. The halo superposition components of all surrounding partitions on the target partition are summed to obtain the total halo superposition received by the backlight partition in the current iteration state. The total halo superposition received by each backlight partition of the mini LED in the current iteration state is then obtained.

[0070] In one specific embodiment, difference calculation is performed to obtain the current estimated halo reduction value of each backlight zone of the mini LED. The specific method is as follows: subtract the estimated halo interference value of each backlight zone of the mini LED affected by the surrounding backlight zones from the total halo superposition amount received in the current iteration state to obtain the current estimated halo reduction value of each backlight zone of the mini LED.

[0071] If the current iteration number exceeds the halo suppression iteration number threshold or the current halo prediction reduction value of each backlight zone of the mini LED is less than the halo error convergence threshold, the iteration is terminated, and the current iteration brightness value of each backlight zone of the mini LED is used as the viewing angle brightness correction value of each backlight zone of the mini LED. Otherwise, based on the current halo prediction reduction value of each backlight zone of the mini LED, the current iteration brightness value of each backlight zone of the mini LED is reversed and corrected, the current iteration brightness value of each backlight zone of the mini LED is updated, the current iteration number is incremented by 1, and the step of calculating the total halo superposition amount received by each backlight zone of the mini LED in the current iteration state is repeated until the termination condition is met.

[0072] In one specific embodiment, reverse compensation correction is performed to update the current iterative brightness value of each backlight zone of the mini LED. The specific method is as follows: divide the estimated halo interference value of each backlight zone of the mini LED affected by the surrounding backlight zones by the current estimated halo reduction value, and multiply it by the current iterative brightness value of each backlight zone of the mini LED to obtain the updated current iterative brightness value of each backlight zone of the mini LED.

[0073] Step 5: Convert the mini LED backlight control matrix into a driving signal to drive the mini LED display to complete the brightness output in real time.

[0074] In one specific embodiment, the mini LED display is driven to complete the brightness output in real time. The specific method is as follows: the output of the display is controlled according to the viewing angle brightness correction value and viewing angle saturation correction value of each backlight zone in the mini LED backlight control matrix.

[0075] A second aspect of the present invention provides a high-definition signal adaptation mini LED backlight control display screen based on AI analysis, including a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the high-definition signal adaptation mini LED backlight control method based on AI analysis as described above.

[0076] All formulas in this manual are dimensionless and calculated numerically. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0077] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, and all such modifications and additions should fall within the protection scope of the present invention.

Claims

1. A method for high-definition signal adaptive mini LED backlight regulation based on AI analysis, characterized in that, include: Step 1: Before the mini LED leaves the factory, pre-collect the diffusion layer thickness parameters, scattering particle size parameters and concentration parameters of the mini LED backlight module, and perform multi-scene optical simulation modeling based on these parameters, thereby pre-constructing the light field spatial distribution characteristics and halo transmission attenuation model of the mini LED backlight module under various viewing angles. Step 2: While each viewer is watching the mini LED, the high-definition image signal to be displayed is acquired in real time, and the spatial position and eye angle of each viewer relative to the display screen are identified by the visual acquisition device, thereby obtaining the visual reception area characteristic parameters of each viewer. Step 3: Perform content AI feature analysis on the high-definition image signal to extract the brightness level distribution features and color saturation distribution features of the high-definition image signal. Based on the light field spatial distribution features and halo transmission attenuation model of the mini LED backlight module under various viewing angles, and combined with the visual reception area feature parameters of each viewer, perform fusion inference calculation to obtain the peak brightness requirement value, color reproduction accuracy requirement coefficient, viewing angle compensation coefficient and halo suppression parameters of each backlight zone of the mini LED. Step 4: Generate a mini LED backlight control matrix based on the peak brightness requirement, color reproduction accuracy requirement coefficient, viewing angle compensation coefficient, and halo suppression parameters for each backlight zone of the mini LED. Step 5: Convert the mini LED backlight control matrix into a driving signal to drive the mini LED display to complete the brightness output in real time. 2.The AI analysis-based high-definition signal adaptation mini LED backlight regulation method of claim 1, wherein, The specific method for the pre-constructed mini LED backlight module's light field spatial distribution characteristics and halo transmission attenuation model under various viewing angles is as follows: Based on the diffusion layer thickness parameters, scattering particle size parameters, and concentration parameters of the mini LED backlight module, a Monte Carlo ray tracing simulation model of the mini LED backlight module is constructed. For each viewing angle, the spatial position of the detector array corresponding to the viewing angle is set in the Monte Carlo ray tracing simulation model, and photon transmission path tracing calculation is performed to obtain the spatial distribution data of luminous flux of mini LED light source under each viewing angle, and use it as the spatial distribution characteristics of light field of mini LED backlight module under each viewing angle. Based on the spatial distribution data of luminous flux of the mini LED backlight module under various viewing angles, an exponential function is fitted to obtain the halo attenuation index and effective halo distance of the mini LED backlight module under various viewing angles, and these are used as the halo transmission attenuation model of the mini LED backlight module under various viewing angles. 3.The AI analysis-based high-definition signal adaptation mini LED backlight regulation method of claim 1, wherein, The method for identifying the spatial position and eye angle of each viewer relative to the display screen, thereby obtaining the visual reception area feature parameters of each viewer, is as follows: Acquire facial and depth images of each viewer, use facial landmark detection algorithm to extract the center coordinates of the pupils of each viewer, and calculate the horizontal and vertical offset distances of each viewer relative to the center of the mini LED display screen, and use them as the spatial position of each viewer relative to the display screen. The physical size parameters of the mini LED display screen are obtained from the local database. Based on the coordinates of the center of each viewer's pupils and the spatial position of each viewer relative to the display screen, the angle between the midpoint of the line connecting each viewer's eyes and the normal direction of the display screen is calculated and used as the viewing angle of each viewer. The eye angle of each viewer and the spatial position of each viewer relative to the display screen are used as the visual reception area feature parameters of each viewer. 4.The AI analysis-based high-definition signal adaptation mini LED backlight regulation method of claim 1, wherein, The specific method for extracting the brightness level distribution features and color saturation distribution features of the high-definition image signal is as follows: The pre-trained image semantic segmentation AI model is obtained from the local database. The high-definition image signal is input into the image semantic segmentation AI model to identify the range and category label of each semantic region in the high-definition image signal. The visual attention weight value corresponding to each semantic category is obtained from the local database, thereby generating a visual attention spatial distribution map of the high-definition image signal. The high-definition image signal is converted to YUV color space, the luminance component value of each pixel position in the high-definition image signal is extracted, and combined with the visual attention spatial distribution map of the high-definition image signal, the regulated luminance component value of each pixel position in the high-definition image signal is evaluated, and it is used as the luminance hierarchical distribution feature of the high-definition image signal. The high-definition image signal is converted to HSV color space to extract the saturation component values ​​of each pixel position in the high-definition image signal. Based on this, the mean saturation value and saturation variance value of each semantic region in the high-definition image signal are statistically analyzed and used as the color saturation distribution characteristics of the high-definition image signal.

5. The method for high-definition signal adaptation and mini LED backlight control based on AI analysis according to claim 4, characterized in that, The method for performing fusion inference calculations to obtain the peak brightness requirement, color reproduction accuracy requirement coefficient, viewing angle compensation coefficient, and halo suppression parameters for each backlight zone of the mini LED is as follows: Retrieve the boundary coordinates of each backlight zone of the mini LED from the local database; Based on the visual reception area characteristic parameters of each viewer, the spatial distribution characteristics of the light field of the mini LED backlight module under each viewing angle and the halo transmission attenuation model are mapped to obtain the spatial distribution data of light flux, halo attenuation index and effective halo distance of each viewer under the corresponding viewing angle. Based on the brightness level distribution characteristics and color saturation distribution characteristics of high-definition image signals, and combined with the partition boundary coordinates of each backlight partition of mini LED, the peak brightness requirement value and color reproduction accuracy requirement coefficient of high-definition image signals in each backlight partition of mini LED are calculated. Based on the spatial distribution data of luminous flux, halo attenuation index and effective halo distance of each viewer at the corresponding viewing angle, a comprehensive analysis is conducted to obtain the viewing angle compensation coefficient of each backlight zone of the mini LED. Based on the halo attenuation index and effective halo distance of each viewer at their respective viewing angles, and combined with the boundary coordinates of each backlight zone of the mini LED, the target halo transmission coefficient between each backlight zone of the mini LED and its surrounding backlight zones is evaluated. The halo interference estimate of each backlight zone of the mini LED affected by the surrounding backlight zones is accumulated and used as the halo suppression parameter of each backlight zone of the mini LED.

6. The method for high-definition signal adaptation and mini LED backlight control based on AI analysis according to claim 5, characterized in that, The specific method for calculating the peak brightness requirement and color reproduction accuracy requirement coefficient of the high-definition image signal in each backlight zone of the mini LED is as follows: Based on the brightness level distribution characteristics of high-definition image signals, the adjustable brightness component values ​​of each pixel position within the partition boundary coordinate range of each backlight zone of mini LED are extracted. The maximum value of the adjustable brightness component values ​​of each pixel position within the partition boundary coordinate range of each backlight zone of mini LED is selected and used as the peak brightness requirement value of high-definition image signals in each backlight zone of mini LED. Based on the color saturation distribution characteristics of high-definition image signals, the mean and variance values ​​of saturation in each semantic region within the boundary coordinate range of each backlight zone of the mini LED are extracted. The mapping relationship curve between the mean and variance values ​​of saturation and the color reproduction accuracy requirement coefficient is obtained from the local database. The color reproduction accuracy requirement coefficient of each semantic region in each backlight zone of the mini LED is obtained by mapping. The maximum value of the color reproduction accuracy requirement coefficient of the semantic region in the backlight zone is taken as the color reproduction accuracy requirement coefficient of the high-definition image signal in each backlight zone of the mini LED. 7.The AI analysis based high definition signal adaptation mini LED backlight regulation method of claim 5, wherein, The specific method for evaluating the target halo transmission coefficient between each backlight zone of the mini LED and its surrounding backlight zones is as follows: Based on the boundary coordinates of each backlight zone of the mini LED, calculate the center coordinates of each backlight zone of the mini LED, and then calculate the Euclidean distance between each backlight zone of the mini LED and its surrounding backlight zones. Based on the halo attenuation index and effective halo distance of each viewer at the corresponding viewing angle, and combined with the Euclidean distance between each backlight zone of the miniLED and its surrounding backlight zones, the halo transmission coefficient between each backlight zone of the miniLED and its surrounding backlight zones at the corresponding viewing angle of each viewer is analyzed. The halo transmission coefficients between each backlight zone of the mini LED and its surrounding backlight zones are averaged for each viewer at their respective viewing angles, thereby obtaining the target halo transmission coefficients between each backlight zone of the mini LED and its surrounding backlight zones. 8.The AI analysis based high definition signal adaptation mini LED backlight regulation method of claim 1, wherein, The specific method for generating the mini LED backlight control matrix is ​​as follows: Based on the peak brightness requirement, color reproduction accuracy requirement coefficient, and viewing angle compensation coefficient of each backlight zone of the mini LED, the initial target brightness value and initial color saturation of each backlight zone of the mini LED are calculated. Based on the halo suppression parameters of each backlight zone of the mini LED and the target halo transmission coefficient between each backlight zone of the mini LED and its surrounding backlight zones, the initial target brightness value of each backlight zone of the mini LED is iteratively corrected for halo suppression, thereby obtaining the viewing angle brightness correction value of each backlight zone of the mini LED. The viewing angle brightness correction values ​​of each backlight zone of the mini LED are arranged and combined in a matrix according to the boundary coordinates of each backlight zone of the mini LED to generate the mini LED backlight control matrix. 9.The AI analysis based high definition signal adaptation mini LED backlight regulation method of claim 8, wherein, The method for performing halo suppression iterative correction to obtain the viewing angle brightness correction value of each backlight zone of the mini LED is as follows: Obtain the halo suppression iteration threshold and halo error convergence threshold from the local database, initialize the current iteration brightness value of each backlight zone of the mini LED to the initial target brightness value of each backlight zone of the mini LED, and initialize the current iteration number to 0; Based on the target halo transmission coefficient between each backlight zone of the mini LED and its surrounding backlight zones and the current iteration brightness value of each backlight zone of the mini LED, the total halo superposition received by each backlight zone of the mini LED in the current iteration state is calculated, and the difference is calculated with the halo suppression parameter of each backlight zone of the mini LED to obtain the current halo estimated reduction value of each backlight zone of the mini LED. If the current iteration number exceeds the halo suppression iteration number threshold or the current halo prediction reduction value of each backlight zone of the mini LED is less than the halo error convergence threshold, the iteration is terminated, and the current iteration brightness value of each backlight zone of the mini LED is used as the viewing angle brightness correction value of each backlight zone of the mini LED. Otherwise, based on the current halo prediction reduction value of each backlight zone of the mini LED, the current iteration brightness value of each backlight zone of the mini LED is reversed and corrected, the current iteration brightness value of each backlight zone of the mini LED is updated, the current iteration number is incremented by 1, and the step of calculating the total halo superposition amount received by each backlight zone of the mini LED in the current iteration state is repeated until the termination condition is met.

10. A high-definition signal adaptation mini LED backlight control display screen based on AI analysis, characterized in that, The device includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor causes the processor to perform the steps of the AI-based high-definition signal adaptation mini LED backlight control method as described in any one of claims 1-9.

Images