A low-power MiniLED backlight control method, system, electronic device, and storage medium

Abstract

This invention discloses a low-power MiniLED backlight control method, system, electronic device, and storage medium. The method includes: mapping a gridded and sparsely sampled RGB image to space, outputting perceived brightness and initial chromaticity; generating a suppressed backlight matrix based on the perceived brightness and a suppression factor, and calculating the actual light intensity of the LCD surface through PSF convolution; calculating the perceived brightness compensation amount using the actual light intensity to generate a target perceived brightness, and outputting corrected chromaticity in conjunction with a color shift function; fine-tuning the suppressed backlight matrix to bring the drive signal back to a safe range when the estimated LCD drive signal exceeds the limit; and inversely mapping the data to XYZ space and generating the final drive signal through a 3D LUT. This eliminates the loss of detail in dark areas and color shift, achieving high dynamic range and low halo display.

Description

Technical Field

[0001] This invention relates to the field of MiniLED backlight control technology, specifically to a low-power MiniLED backlight control method, system, electronic device, and storage medium. Background Technology

[0002] MiniLED backlighting technology has been widely used in the LCD display field due to its high peak brightness, high contrast ratio, and local dimming capabilities. In a typical MiniLED-backlit LCD display system, in order to reduce backlight power consumption and improve display contrast, the backlight decision module dynamically suppresses the backlight LED array according to the brightness distribution of the input image, reducing backlight power in dark areas. At the same time, to compensate for the brightness reduction caused by backlight suppression, the system usually performs digital compensation on the opening of the LCD panel at the image data end, using the increase in liquid crystal opening to maintain the target emitted light intensity, thereby achieving high dynamic range display.

[0003] Existing backlight suppression and liquid crystal compensation algorithms are mainly based on simple energy superposition and product calculations in the linear physical light intensity domain. They ignore the nonlinear subjective perception characteristics of the human eye's response to light intensity, making it difficult to accurately match the amount of liquid crystal compensation with the subjective visual needs of the human eye. This can easily cause problems such as dull and lost details in dark areas and insufficient transparency in highlights in LCD display images after excessive backlight suppression. Summary of the Invention

[0004] The purpose of this invention is to overcome the above-mentioned technical deficiencies and provide a low-power MiniLED backlight control method, system, electronic device and storage medium, which can improve the problems of dull and lost details in dark areas and insufficient transparency of highlights in LCD display images after excessive backlight suppression.

[0005] To achieve the above-mentioned technical objectives, the present invention adopts the following technical solution:

[0006] In a first aspect, the present invention provides a low-power MiniLED backlight control method, comprising:

[0007] S1. Perform salient gridding and sparse sampling on the input RGB image, and map it to a lookup table. Perceive a uniform color space and output perceived brightness data and initial chromaticity coordinate data;

[0008] S2, used to determine the target backlight matrix based on the perceived brightness data, introduce a suppression factor to calculate and generate a suppression backlight matrix, and perform point spread function convolution calculation on the suppression backlight matrix to output the actual light intensity data of the LCD surface.

[0009] S3. Calculate the perceived brightness compensation amount based on the perceived brightness data and the actual light intensity data of the LCD surface, and superimpose the perceived brightness compensation amount onto the perceived brightness data to output the target perceived brightness data; at the same time, correct the initial chromaticity coordinate data based on the actual light intensity data of the LCD surface and the nonlinear color shift function of the liquid crystal panel, and output the corrected chromaticity coordinate data.

[0010] S4. Based on the target perceived brightness data and corrected chromaticity coordinate data, estimate the LCD driving signal. If it is detected that the estimated LCD driving signal corresponding to more than a preset proportion of pixels enters the liquid crystal physical response saturation region, then finely adjust the suppression backlight matrix upward and update the actual light intensity data of the LCD surface, so that the estimated LCD driving signal returns to the linear response safe range.

[0011] S5. Based on the final target perceived brightness data and corrected chromaticity coordinate data, reverse map back to the XYZ space, and generate the final LCD driving signal through a three-dimensional lookup table.

[0012] In some embodiments, in S1, performing salient gridding and sparse sampling on the input RGB image includes:

[0013] The input RGB image is segmented into multiple image grids, and the saliency index of each image grid is calculated.

[0014] The significance index is compared with a preset threshold. For image grids with a significance index less than or equal to the preset threshold, a portion of the pixels are extracted and sparsely sampled. For image grids with a significance index greater than the preset threshold, all pixels are collected. After spatial mapping, interpolation is performed on the sparsely sampled pixels to complete the data.

[0015] In some embodiments, in S1, mapping is performed via a lookup table to... Perceiving a uniform color space includes:

[0016] A one-dimensional logarithmic lookup table is used to perform non-linear perceptual quantization processing on the sampled pixel data to obtain perceptual quantization data.

[0017] Multiply the perceived quantized data by a preset three-dimensional transformation matrix to complete the transformation. Perceive the mapping of a uniform color space.

[0018] In some embodiments, in S2, the calculation of the suppression backlight matrix by introducing a suppression factor includes:

[0019] The suppression factor is calculated based on the local contrast of each region of the input RGB image and the visual masking threshold.

[0020] The initial backlight brightness of each MiniLED zone is determined based on the perceived brightness data to form the target backlight matrix;

[0021] Multiply each element in the target backlight matrix by the corresponding suppression factor to obtain the suppression backlight matrix;

[0022] The point spread function convolution calculation for the suppressed backlight matrix includes:

[0023] Obtain the spatial optical attenuation distribution matrix formed by the emission of a single LED on the LCD surface in the MiniLED backlight module, and use it as the point spread function;

[0024] The suppression backlight matrix and the point spread function are subjected to two-dimensional spatial convolution operation to calculate and generate the actual light intensity data of the LCD surface.

[0025] In some embodiments, in S3, calculating the perceived brightness compensation amount based on the perceived brightness data and the actual light intensity data of the LCD surface includes:

[0026] The actual light intensity data of the LCD surface is input into a preset LCD liquid crystal panel transmittance and electro-optic response model, and the normalized calculation is used to obtain the actual display light intensity of the LCD surface.

[0027] use The perception quantization function in space performs nonlinear compression on the actual displayed light intensity to obtain the actual perceived brightness.

[0028] The target difference is obtained by subtracting the actual perceived brightness from the perceived brightness data, and the target difference is then adjusted by weighted filtering gain to obtain the perceived brightness compensation amount.

[0029] The initial chromaticity coordinate data is corrected based on the actual light intensity data of the LCD surface and the nonlinear color shift function of the LCD panel, including:

[0030] Real-time acquisition of actual light intensity data of the LCD surface and temperature feedback data of the backlight module;

[0031] The actual light intensity data of the LCD surface and the temperature feedback data are simultaneously input into a pre-fitted multivariate polynomial nonlinear color shift model to calculate the color deviation caused by the wavelength drift of the blue light chip and the change in phosphor excitation efficiency.

[0032] The initial chromaticity coordinate data is compensated by offset superposition based on the chromaticity deviation to generate corrected chromaticity coordinate data.

[0033] In some embodiments, in S4, the estimated LCD driving signal corresponding to more than a preset proportion of pixels enters the liquid crystal physical response saturation region, including:

[0034] Based on the perceived brightness compensation amount, the estimated LCD driving signal corresponding to each pixel is calculated and generated.

[0035] The amplitude of each estimated LCD driving signal is compared with the preset liquid crystal molecule transmittance saturation threshold, and the number of pixels exceeding the saturation threshold is counted.

[0036] Calculate the ratio of the number of pixels to the total number of pixels in the current region of the image. When the ratio is greater than a preset ratio threshold, determine that the estimated LCD driving signal has entered the liquid crystal physical response saturation region and trigger a backlight fine-tuning request.

[0037] Fine-tuning the suppressed backlight matrix and updating the actual light intensity data of the LCD surface includes:

[0038] In response to the backlight fine-tuning request, obtain the brightness undercompensation amount of pixels that exceed the saturation threshold;

[0039] Calculate the backlight increment based on the brightness undercompensation amount, and superimpose the backlight increment onto the corresponding suppressed backlight matrix to generate an updated target backlight matrix.

[0040] The updated target backlight matrix is ​​then re-convolved with the point spread function in a second two-dimensional spatial operation to update the actual light intensity data of the LCD surface in real time.

[0041] In some embodiments, in S5, generating the final LCD drive signal using a three-dimensional lookup table includes:

[0042] The initial chromaticity coordinate data corrected by the color shift compensation value is fused with the target perceived brightness after being superimposed by the perceived brightness compensation amount to construct the three-dimensional color space input coordinates.

[0043] The three-dimensional color space input coordinates are input into a pre-fixed hardware-level three-dimensional lookup table, and the liquid crystal transmittance correction coefficient and gamma correction data corresponding to the input coordinates are searched in parallel.

[0044] The correction coefficients and the gamma correction data are combined to generate and output the final LCD driving signal that conforms to the physical response characteristics of the current LCD panel.

[0045] Secondly, the present invention also provides a low-power MiniLED backlight control system, comprising:

[0046] The spatial transformation module performs salient gridding and sparse sampling on the input RGB image, and maps it to a lookup table. Perceive a uniform color space and output perceived brightness data and initial chromaticity coordinate data;

[0047] The backlight decision module is used to determine the target backlight matrix based on the perceived brightness data, introduce a suppression factor to calculate and generate a suppression backlight matrix, perform point spread function convolution calculation on the suppression backlight matrix, and output the actual light intensity data of the LCD surface.

[0048] The brightness and color decoupling compensation module is used to calculate the perceived brightness compensation amount based on the perceived brightness data and the actual light intensity data of the LCD surface, and to superimpose the perceived brightness compensation amount onto the perceived brightness data to output the target perceived brightness data; at the same time, it corrects the initial chromaticity coordinate data based on the actual light intensity data of the LCD surface and the nonlinear color shift function of the liquid crystal panel, and outputs the corrected chromaticity coordinate data.

[0049] The adaptive backoff module is used to estimate the LCD driving signal based on the target perceived brightness data and the corrected chromaticity coordinate data. If it is detected that the estimated LCD driving signal corresponding to more than a preset proportion of pixels enters the liquid crystal physical response saturation region, the suppression backlight matrix is ​​finely adjusted upward and the actual light intensity data of the LCD surface is updated so that the estimated LCD driving signal returns to the linear response safe range.

[0050] The reverse mapping module is used to reverse map back to the XYZ space based on the final target perceived brightness data and corrected chromaticity coordinate data, and generate the final LCD driving signal through a three-dimensional lookup table.

[0051] Thirdly, the present invention also provides an electronic device, comprising: a processor and a memory;

[0052] The memory stores a computer-readable program that can be executed by the processor;

[0053] When the processor executes the computer-readable program, it implements the steps in the low-power MiniLED backlight control method described above.

[0054] Fourthly, the present invention also provides a computer-readable storage medium storing one or more programs that can be executed by one or more processors to implement the steps in the low-power MiniLED backlight control method described above.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] In this invention, by introducing The perceptual color space and perceptual quantization function perform nonlinear compression and brightness decoupling of the actual display light intensity on the LCD surface. Combined with low-pass filtering and dynamic gain to generate perceptual brightness compensation, it achieves accurate alignment and high-fidelity dynamic compensation of display brightness in the dimension of human visual perception. This solves the problems of loss of dark details, dullness and insufficient transparency of highlights in LCD screens caused by excessive suppression of MiniLED backlight in traditional algorithms.

[0057] 2. In this invention, by constructing a backlight fine-tuning closed-loop control mechanism that includes driving signal saturation region detection, saturation pixel ratio threshold determination, brightness undercompensation extraction, and PSF-based double convolution, the linkage adaptive back-off adjustment of liquid crystal driving signal and backlight matrix is ​​realized. This solves the problems of bright detail truncation and edge overshoot halo caused by liquid crystal molecules deflecting into the physical response saturation region in the traditional unidirectional open-loop compensation algorithm, ensuring that liquid crystal molecules always work within the linear response safety range.

[0058] 3. In this invention, by integrating the actual light intensity of the LCD surface and the temperature feedback data of the backlight module into a multivariate polynomial nonlinear color shift model, and using a hardware-level three-dimensional lookup table for parallel addressing interpolation and correction data synthesis, the invention achieves accurate color correction of light source wavelength drift and phosphor thermal quenching, as well as hardware-level real-time processing of complex color space conversion. This solves the display color shift problem caused by MiniLED driving current changes and temperature drift, and overcomes the bottleneck of high algorithm computation complexity and difficulty in real-time implementation under high refresh rate display terminals. Attached Figure Description

[0059] Figure 1 This is an architecture diagram of a low-power MiniLED backlight control method proposed in this invention;

[0060] Figure 2 This is a flowchart of the S1 process of a low-power MiniLED backlight control method proposed in this invention.

[0061] Figure 3 This is a flowchart of the S2 process of a low-power MiniLED backlight control method proposed in this invention.

[0062] Figure 4 This is a flowchart of the S3 process of a low-power MiniLED backlight control method proposed in this invention.

[0063] Figure 5 Here is the S4 closed-loop control flowchart of a low-power MiniLED backlight control method proposed in this invention;

[0064] Figure 6 The above is a flowchart of S5 for a low-power MiniLED backlight control method proposed in this invention.

[0065] Figure 7 This is an architecture diagram of a low-power MiniLED backlight control system proposed in this invention.

[0066] Figure 8 This is a schematic diagram of the operating environment of the electronic device proposed in this invention. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0068] This embodiment provides a low-power MiniLED backlight control method. Please refer to [link / reference]. Figures 1-6 A low-power MiniLED backlight control method includes:

[0069] S1. Perform salient gridding and sparse sampling on the input RGB image, and map it to a lookup table. Perceive a uniform color space and output perceived brightness data and initial chromaticity coordinate data;

[0070] Furthermore, in S1, the input RGB image is saliently meshed and sparsely sampled, including:

[0071] The input RGB image is segmented into multiple image grids, and the saliency index of each image grid is calculated.

[0072] The significance index is compared with a preset threshold. For image grids with a significance index less than or equal to the preset threshold, a portion of the pixels are extracted and sparsely sampled. For image grids with a significance index greater than the preset threshold, all pixels are collected. After spatial mapping, interpolation is performed on the sparsely sampled pixels to complete the data.

[0073] Furthermore, in S1, mapping is performed via a lookup table to... Perceiving a uniform color space includes:

[0074] A one-dimensional logarithmic lookup table is used to perform non-linear perceptual quantization processing on the sampled pixel data to obtain perceptual quantization data.

[0075] Multiply the perceived quantized data by a preset three-dimensional transformation matrix to complete the transformation. Perceive the mapping of a uniform color space.

[0076] Specifically, the system first receives an input RGB image from the previous-level video decoding circuit or image signal processor. The input RGB image is typically stored as a two-dimensional pixel matrix, where each pixel contains the grayscale values ​​of the three basic color channels: red, green, and blue. The input RGB image is then divided into multiple image grids arranged in an array in two-dimensional space; for example, the image is segmented into pixel blocks of fixed size. For each image grid, its pixel grayscale matrix is ​​extracted, and the saliency index corresponding to that grid is calculated to quantify the visual appeal or information richness of the current grid region to the human visual system.

[0077] In the specific calculation of the saliency index, a weighted sum is performed by combining the local contrast and spatial high-frequency energy of the image grid. Specifically, for a given image grid, its saliency index can be calculated using the following formula:

[0078] ;

[0079] In the above formula, S represents the saliency index of the current image grid; C represents the local contrast of the current image grid, which is obtained by calculating the variance or absolute value of the difference between the average brightness of each pixel in the grid and the average brightness of the surrounding image grids; E represents the spatial high-frequency energy of the current image grid, which is obtained by summing the absolute values ​​after applying two-dimensional Laplacian operator filtering to the gray-level matrix of the pixels in the grid. and These represent the weighting coefficients for local contrast and spatial high-frequency energy, respectively, and the sum of the weighting coefficients is 1. For example, in one feasible embodiment, It can be set to 0.6. It can be set to 0.4.

[0080] After calculating the saliency index of each image grid, the saliency index is compared with a preset threshold. In one feasible implementation, the preset threshold can be set to 0.35. If the saliency index of an image grid is less than or equal to the preset threshold, it indicates that the grid area belongs to a flat background or a low-contrast minor area, and sparse sampling logic is activated for it, that is, only a portion of the pixels at specific locations within the image grid are extracted, for example, by downsampling at fixed pixel intervals; if the saliency index of an image grid is greater than the preset threshold, it indicates that the grid belongs to a visually sensitive edge or detail area, and full sampling logic is activated for it, collecting all pixels within the image grid.

[0081] After sampling is complete, the extracted sampled pixel data is submitted to the color space conversion unit. The sampled pixel data first enters a one-dimensional logarithmic lookup table to complete the non-linear perceptual quantization processing from RGB quantization values ​​to the perceptual quantization domain. The one-dimensional logarithmic lookup table is pre-stored in hardware memory, and its mapping relationship simulates the logarithmic response characteristics of the human visual system to changes in light intensity. The pixel data uses its RGB value as an address index to look up the one-dimensional logarithmic lookup table and directly outputs the corresponding perceptual quantization data.

[0082] After acquiring the perceptual quantization data, the data is constructed into a three-dimensional numerical vector and multiplied by a pre-stored three-dimensional transformation matrix to map and transform the data to a perceptually uniform color space. The three-dimensional transformation matrix used is determined based on the standard color space transformation relationship and the human visual perception curve. The specific multiplication operation is shown below:

[0083] ;

[0084] In the above formula, These represent the perceived brightness scalar, the red-green axis chromaticity component scalar, and the yellow-blue axis chromaticity component scalar, respectively, in a perceived uniform color space. These represent the perceptual quantization data of the red, green, and blue channels obtained after processing with a one-dimensional logarithmic lookup table; matrix elements to The constant parameters that constitute the three-dimensional transformation matrix are determined by matching the primary color coordinates of the target color space and the input RGB color space.

[0085] After the 3D transformation matrix multiplication calculation, for pixels in the fully sampled area, the corresponding perceptual luminance scalar and chrominance components are directly generated. For pixels that were previously sparsely sampled, due to pixel gaps in the spatial grid, spatial interpolation is performed to fill them in the perceptual uniform color space. Specifically, based on the perceptual luminance scalar and chrominance components of adjacent collected and transformed pixels, the perceptual coordinates of the missing pixels in the sparse sampling are calculated and filled using a bilinear interpolation algorithm, thereby restoring the complete data distribution aligned with the original image resolution on the spatial grid.

[0086] Through a series of coherent data processing steps, including saliency partitioning, sparse sampling, lookup table nonlinear quantization, matrix multiplication spatial transformation, and receptive domain spatial interpolation completion, the system ultimately outputs globally corresponding perceived brightness data and initial chromaticity coordinate data to subsequent steps. The perceived brightness data consists of a two-dimensional matrix composed of the perceived brightness scalars of each pixel, while the initial chromaticity coordinate data consists of a two-dimensional vector matrix composed of the red-green axis chromaticity component scalars and the yellow-blue axis chromaticity component scalars of each pixel.

[0087] S2, used to determine the target backlight matrix based on the perceived brightness data, introduce a suppression factor to calculate and generate a suppression backlight matrix, and perform point spread function convolution calculation on the suppression backlight matrix to output the actual light intensity data of the LCD surface.

[0088] Furthermore, in S2, the calculation of the suppression factor to generate the suppression backlight matrix includes:

[0089] The suppression factor is calculated based on the local contrast of each region of the input RGB image and the visual masking threshold.

[0090] The initial backlight brightness of each MiniLED zone is determined based on the perceived brightness data to form the target backlight matrix;

[0091] Multiply each element in the target backlight matrix by the corresponding suppression factor to obtain the suppression backlight matrix;

[0092] The point spread function convolution calculation for the suppressed backlight matrix includes:

[0093] Obtain the spatial optical attenuation distribution matrix formed by the emission of a single LED on the LCD surface in the MiniLED backlight module, and use it as the point spread function;

[0094] The suppression backlight matrix and the point spread function are subjected to two-dimensional spatial convolution operation to calculate and generate the actual light intensity data of the LCD surface.

[0095] Specifically, it receives perceived brightness data and the original input RGB image. First, it needs to calculate a suppression backlight matrix based on the input RGB image and perceived brightness data, which can match the image display brightness, minimize power consumption, and suppress halo. Then, it uses optical simulation to deduce the actual light intensity distribution reaching the LCD panel surface.

[0096] Spatial physical partitioning mapping is performed on the perceived brightness data. Since the MiniLED backlight module consists of multiple independent backlight partitions arranged in a matrix, its physical resolution is much lower than the pixel resolution of an LCD panel. Therefore, the backlight generation unit divides the two-dimensional matrix formed by the perceived brightness data into multiple sub-regions according to the physical layout of the MiniLED partitions. For each sub-region corresponding to a backlight partition, the backlight generation unit extracts the perceived brightness scalar of all pixels within that sub-region and takes its maximum value or weighted average as the initial backlight brightness control value for the corresponding MiniLED partition. The initial backlight brightness control values ​​of all MiniLED partitions are arranged in physical array order, thus forming the target backlight matrix.

[0097] To effectively suppress halo effects and reduce backlight power consumption in scenes with bright details against a dark background, a backlight suppression mechanism based on the human visual masking effect is introduced. The input RGB image is extracted, and the local contrast and visual masking threshold of the corresponding regions in each backlight zone are calculated. The local contrast reflects the intensity of the boundary between light and dark areas within the current image region, while the visual masking threshold reflects the human eye's sensitivity to distortion of details in dark areas surrounding highlights or light leakage from halos under the current background brightness. Based on the local contrast and visual masking threshold, the suppression factor for each backlight zone is calculated. Specifically, for the _____, ... Line 1 The suppression factor of the backlight zones in a column can be calculated using the following formula:

[0098] ;

[0099] In the above formula, Indicates the first Line 1 The suppression factor of the backlight zone ranges from 0 to 1; This represents the local contrast of the image region corresponding to the current backlight partition, which is obtained by calculating the standard deviation or the difference between the maximum and minimum values ​​of the pixel gray levels in that region. This represents the visual masking threshold for the current region, which is calculated based on the average background brightness of the current region using the human eye brightness masking curve. This represents a preset, minimal positive real constant used to prevent the denominator from being zero. For example, in one feasible embodiment... It can be set to 0.001; This represents the adjustment gain coefficient that controls the intensity of the suppression, for example, in one feasible embodiment... It can be set to 0.5. When the local contrast of a region is high and the visual masking threshold is low, the suppression factor calculated by the formula decreases, thereby appropriately reducing the backlight brightness of that region; conversely, the suppression factor is close to 1, maintaining the original brightness.

[0100] After calculating the suppression factor covering all backlight zones, each element in the target backlight matrix is ​​multiplied element-wise by the suppression factor at the corresponding position using a dot product operation. The calculation process is as follows:

[0101] ;

[0102] In the above formula, Indicates the suppression of the backlight matrix. Line 1 Backlight brightness elements of the column; This represents the initial backlight brightness control value at the corresponding position in the target backlight matrix; This represents the suppression factor at the corresponding position. After the calculation is completed, a suppression backlight matrix that balances display brightness and power consumption is generated.

[0103] After generating the suppression backlight matrix, optical modeling techniques are used to calculate the physical light intensity distribution of the light emitted by the MiniLED after passing through the diffuser plate, quantum dot film, and optical film assembly to reach the LCD panel surface. First, the point spread function (PSF), pre-mapped and stored in hardware, is obtained. The PSF characterizes the two-dimensional spatial optical intensity attenuation distribution matrix formed on the LCD panel surface after the light emitted by a single LED in the MiniLED backlight module propagates and diffuses in space at a unit reference driving level. The spatial distribution shape of the PSF is usually an axisymmetric Gaussian distribution or a double Gaussian distribution, and its spatial attenuation characteristics can be mathematically described by the following two-dimensional Gaussian model:

[0104] ;

[0105] In the above formula, This represents the relative coordinates of the LCD surface space when the physical center point of a single LED is taken as the origin. The relative light intensity attenuation coefficient at that location; This represents the normalization coefficient of the peak luminous intensity of a single LED; The parameter represents the attenuation standard deviation of light intensity as it diffuses in space. Its value is determined by the physical hardware structure, such as the MiniLED chip packaging angle, lens structure, and backlight module mixing distance.

[0106] The suppressed backlight matrix is ​​used as input to a low-resolution point light source array, and the point spread function is used as the convolution kernel. Both are subjected to a two-dimensional discrete spatial convolution operation. Since the resolution of the suppressed backlight matrix corresponds to the number of backlight zones, and the final calculation requires the actual light intensity at the pixel level of the LCD panel, the suppressed backlight matrix is ​​mapped to a two-dimensional grid space aligned with the LCD pixel resolution according to its physical spatial location during the convolution operation. The calculation process is as follows:

[0107] ;

[0108] In the above formula, The coordinates of the LCD surface are: The actual physical light intensity at the pixel location; Indicates the suppression of the backlight matrix. Line 1 The driving brightness values ​​for each backlight zone; Indicates the first The coordinates of each backlight zone center mapped to the LCD surface Then, for coordinates The light intensity contribution generated by each pixel; and These represent the pixel spacing distance between the centers of adjacent backlight zones on the horizontal and vertical sides of the LCD panel, respectively.

[0109] By performing a two-dimensional spatial convolution operation on the suppressed backlight matrix and the dot spread function as described above, the actual light intensity data of the LCD surface formed at each pixel position of the LCD panel after MiniLED emission and spatial light mixing is accurately calculated. This actual light intensity data of the LCD surface is output in the form of a high-resolution two-dimensional light intensity distribution matrix and is submitted as a key reference physical quantity to subsequent steps for performing decoupling and precise compensation of brightness and chromaticity.

[0110] S3. Calculate the perceived brightness compensation amount based on the perceived brightness data and the actual light intensity data of the LCD surface, and superimpose the perceived brightness compensation amount onto the perceived brightness data to output the target perceived brightness data; at the same time, correct the initial chromaticity coordinate data based on the actual light intensity data of the LCD surface and the nonlinear color shift function of the liquid crystal panel, and output the corrected chromaticity coordinate data.

[0111] Furthermore, in S3, calculating the perceived brightness compensation amount based on the perceived brightness data and the actual light intensity data of the LCD surface includes:

[0112] The actual light intensity data of the LCD surface is input into a preset LCD liquid crystal panel transmittance and electro-optic response model, and the normalized calculation is used to obtain the actual display light intensity of the LCD surface.

[0113] use The perception quantization function in space performs nonlinear compression on the actual displayed light intensity to obtain the actual perceived brightness.

[0114] The target difference is obtained by subtracting the actual perceived brightness from the perceived brightness data, and the target difference is then adjusted by weighted filtering gain to obtain the perceived brightness compensation amount.

[0115] The initial chromaticity coordinate data is corrected based on the actual light intensity data of the LCD surface and the nonlinear color shift function of the LCD panel, including:

[0116] Real-time acquisition of actual light intensity data of the LCD surface and temperature feedback data of the backlight module;

[0117] The actual light intensity data of the LCD surface and the temperature feedback data are simultaneously input into a pre-fitted multivariate polynomial nonlinear color shift model to calculate the color deviation caused by the wavelength drift of the blue light chip and the change in phosphor excitation efficiency.

[0118] The initial chromaticity coordinate data is compensated by offset superposition based on the chromaticity deviation to generate corrected chromaticity coordinate data.

[0119] Specifically, the process receives the perceived brightness data and initial chromaticity coordinate data output from the previous step, as well as the calculated actual light intensity data of the LCD surface. This step aims to address the brightness attenuation caused by insufficient transmitted light intensity of the LCD panel due to MiniLED backlight suppression, and the color shift issues of the LCD panel and backlight caused by changes in MiniLED driving current and temperature rise. This is achieved by decoupling the brightness and chromaticity dimensions within a uniform color space to generate target perceived brightness data and corrected chromaticity coordinate data, respectively.

[0120] First, the calculation logic for the perceived brightness compensation is executed. Since the actual light intensity data of the LCD surface only represents the actual backlight intensity reaching the back of the LCD panel, and the brightness ultimately observed by the human eye depends on the actual emitted light intensity after the backlight passes through the liquid crystal molecules, the actual light intensity data of the LCD surface is input into a preset LCD panel transmittance and electro-optic response model. This model characterizes the physical transmittance of the liquid crystal panel under reference or maximum opening conditions and the electro-optic conversion attenuation characteristics of the liquid crystal molecules. The model performs transmission attenuation and normalization calculations on the input actual light intensity data of the LCD surface to obtain the actual display light intensity of the LCD surface. The normalization calculation process is shown in the following formula:

[0121] ;

[0122] In the above formula, Indicates the LCD panel at coordinates The normalized value of the actual displayed light intensity at the pixel location; This represents the actual light intensity data of the LCD surface at the corresponding pixel location; This represents the reference maximum backlight intensity of the MiniLED backlight module at full power peak emission. The reference peak physical transmittance constant of the LCD panel, for example, in one feasible embodiment, It can be set to 0.06.

[0123] After calculating the actual displayed light intensity, in order to dimensionally align the data of the physical light intensity domain with the perceived color space, the following method is used: The perceptual quantizer (PQ) in the color space performs nonlinear compression on the actual displayed light intensity, converting it into the actual perceived brightness as perceived by the human eye. The nonlinear compression calculation of the perceptual quantizer is shown in the following formula:

[0124] ;

[0125] In the above formula, Represents the coordinates of the LCD surface The actual perceived brightness scalar at the pixel; This represents the normalized actual displayed light intensity obtained from the aforementioned steps; All The standard constant of the spatially perceived quantization curve, for example in one feasible embodiment... It can be set to 0.15930276. It can be set to 78.84375. It can be set to 0.8359375. It can be set to 18.8515625. It can be set to 18.6875.

[0126] After obtaining the actual perceived brightness, the output raw perceived brightness data is subtracted pixel by pixel from the actual perceived brightness to obtain the target difference. This target difference reflects the amount of perceived brightness loss caused by backlight suppression. To eliminate spatial high-frequency noise and avoid overcompensation leading to edge overshoot, the target difference is subjected to weighted filtering and dynamic gain adjustment. The calculation process is shown in the following formula:

[0127] ;

[0128] In the above formula, Represents the final generated coordinates Perceived brightness compensation at each pixel; This represents the perceived brightness scalar at the corresponding position in the original perceived brightness data; This indicates the actual perceived brightness at the corresponding location; This represents a two-dimensional low-pass spatial filtering operation; This represents a preset compensation gain adjustment coefficient used to control the response intensity of brightness compensation. For example, in one feasible embodiment, It can be set to 0.95.

[0129] Generate perceived brightness compensation amount Then, it is directly superimposed back onto the original perceived brightness data, and the calculation formula is as follows:

[0130] ;

[0131] In the above formula, That is, the coordinates in the target perceived brightness data. The scalar value at the pixel.

[0132] While performing brightness compensation, color correction logic is executed in parallel. Real-time data on the actual light intensity of the LCD surface is acquired, and temperature feedback data from the backlight module is read in real-time from a temperature sensor array deployed on the MiniLED backlight board. The actual light intensity data of the LCD surface and the temperature feedback data are simultaneously input into a pre-fitted multivariate polynomial nonlinear color shift model. This model, constructed through prior physical mapping of the MiniLED chip and phosphor characteristics, is used to quantitatively describe the color deviation of the emitted spectrum caused by the blue light chip's dominant wavelength drift due to changes in driving light intensity, and by the combined effects of phosphor excitation efficiency decay and thermal quenching due to increased junction temperature.

[0133] A multivariate polynomial nonlinear color shift model, using light intensity and temperature as independent variables, calculates the wavelength drift of the blue LED chip and the change in phosphor excitation efficiency. The calculation process for the chromaticity deviation along the red-green axis and the chromaticity deviation along the yellow-blue axis caused in space is shown in the following formula:

[0134] ;

[0135] ;

[0136] In the above formula, and Representing coordinates The scalar values ​​of the red-green axis chromaticity deviation and the yellow-blue axis chromaticity deviation at the pixel; This represents the actual light intensity value of the LCD surface at the corresponding pixel location; This represents the temperature feedback data of the backlight module, which is spatially interpolated and aligned to the corresponding pixel position. to as well as to The constant polynomial coefficients are pre-fitted for the multivariate polynomial model.

[0137] After calculating the chromaticity deviation, the chromaticity deviation is constructed into a chromaticity compensation vector, and offset compensation is performed by superimposing it with the initial chromaticity coordinate data. The calculation process is shown in the following formula:

[0138] ;

[0139] ;

[0140] In the above formula, and These represent the coordinates in the output corrected chromaticity coordinate data. Corrected red-green axis chromaticity components and yellow-blue axis chromaticity components at the pixel; and These represent the component values ​​at corresponding positions in the initial chromaticity coordinate data.

[0141] Through the above calculation process, the decoupled and accurate calculation of brightness and chromaticity is completed, and finally the target perceived brightness data and corrected chromaticity coordinate data are output to the subsequent steps of the system.

[0142] S4. Based on the target perceived brightness data and corrected chromaticity coordinate data, estimate the LCD driving signal. If it is detected that the estimated LCD driving signal corresponding to more than a preset proportion of pixels enters the liquid crystal physical response saturation region, then finely adjust the suppression backlight matrix upward and update the actual light intensity data of the LCD surface, so that the estimated LCD driving signal returns to the linear response safe range.

[0143] Furthermore, in S4, the estimated LCD driving signal corresponding to pixels exceeding a preset ratio entering the liquid crystal physical response saturation region includes:

[0144] Based on the perceived brightness compensation amount, the estimated LCD driving signal corresponding to each pixel is calculated and generated.

[0145] The amplitude of each estimated LCD driving signal is compared with the preset liquid crystal molecule transmittance saturation threshold, and the number of pixels exceeding the saturation threshold is counted.

[0146] Calculate the ratio of the number of pixels to the total number of pixels in the current region of the image. When the ratio is greater than a preset ratio threshold, determine that the estimated LCD driving signal has entered the liquid crystal physical response saturation region and trigger a backlight fine-tuning request.

[0147] Fine-tuning the suppressed backlight matrix and updating the actual light intensity data of the LCD surface includes:

[0148] In response to the backlight fine-tuning request, obtain the brightness undercompensation amount of pixels that exceed the saturation threshold;

[0149] Calculate the backlight increment based on the brightness undercompensation amount, and superimpose the backlight increment onto the corresponding suppressed backlight matrix to generate an updated target backlight matrix.

[0150] The updated target backlight matrix is ​​then re-convolved with the point spread function in a second two-dimensional spatial operation to update the actual light intensity data of the LCD surface in real time.

[0151] Specifically, it simultaneously receives the target perceived brightness data and corrected chromaticity coordinate data output from the previous step, as well as the previously calculated perceived brightness compensation amount and dot spread function. This step aims to address the problem of the liquid crystal panel driving signal exceeding the linear region of the liquid crystal molecule's electro-optic response due to excessive backlight suppression. By real-time monitoring and estimating the amplitude distribution of the liquid crystal driving signal, and triggering closed-loop adaptive fine-tuning of the backlight array and liquid crystal driving signal when liquid crystal molecule deflection enters the physical response saturation region, it ensures that the liquid crystal panel always operates within the safe range of linear response.

[0152] The logic for predicting the liquid crystal drive signal is executed. The perceived brightness compensation amount calculated in the previous step is extracted and, combined with the electro-optical transmission characteristic equation of the current liquid crystal panel, the brightness compensation requirement of the perceived domain is inversely calculated into the predicted drive level of each pixel on the liquid crystal panel. For coordinates... The estimated LCD driving signal for a given pixel can be calculated using the following formula:

[0153] ;

[0154] In the above formula, Representing coordinates Estimated LCD drive signal amplitude at each pixel; This represents a scalar value indicating the perceived brightness data of the target at the corresponding pixel location. This represents the actual light intensity value on the LCD surface during the current iteration period; This represents the inverse function of the electro-optic response of the LCD panel, which is pre-mapped and stored, and is used to map the desired target liquid crystal transmittance to the corresponding liquid crystal driving voltage signal.

[0155] After generating the estimated LCD driving signal for each pixel in the entire image, the estimated LCD driving signal for each pixel is... With respect to the preset liquid crystal molecule transmittance saturation threshold Perform pixel-by-pixel amplitude comparison. Liquid crystal molecule transmittance saturation threshold. This is the critical voltage constant at which the liquid crystal molecule deflection angle approaches the physical limit and the transmittance no longer increases linearly with increasing driving voltage. For example, in a feasible embodiment, for an 8-bit grayscale driving system, this saturation threshold can be set to the voltage value corresponding to 240 grayscale levels. When the estimated LCD driving signal for a certain pixel satisfies... When a pixel exceeds the saturation threshold, it is considered to have entered the liquid crystal physical response saturation region. The total number of saturated pixels is obtained by accumulating the number of pixels exceeding the saturation threshold across the entire image or within a specific image region. .

[0156] Next, the number of saturated pixels is calculated as a percentage of the total number of pixels in the current image region. The percentage, and compared with a preset ratio threshold. Numerical comparisons are performed. In one feasible embodiment, a preset ratio threshold is used. It can be set to 0.05 (i.e., 5%). The determination logic is shown in the following formula:

[0157] ;

[0158] In the above formula, This indicates a backlight fine-tuning request flag. When the calculated saturation pixel ratio exceeds a preset ratio threshold, the flag is set to 1, indicating that the current estimated LCD drive signal has entered the liquid crystal physical response saturation region on a large scale, and a backlight fine-tuning request is formally triggered to the backlight control loop; otherwise, the flag is set to 0, and the system directly transmits the current estimated LCD drive signal as a safety signal to the next step.

[0159] In response to the backlight fine-tuning request, the backlight matrix up-adjustment mechanism is activated. All elements exceeding the saturation threshold are extracted. For each pixel, calculate its corresponding brightness undercompensation. Brightness undercompensation represents the brightness gap that cannot be compensated by increasing the liquid crystal aperture due to liquid crystal molecule deflection saturation. For saturated pixels, the brightness undercompensation can be calculated using the following formula:

[0160] ;

[0161] In the above formula, Representing coordinates Physical brightness undercompensation at the pixel; This represents the actual physical transmittance of liquid crystal molecules at the saturation critical voltage.

[0162] The brightness undercompensation amount of each saturated pixel is assigned and mapped to the corresponding MiniLED backlight zone according to its physical spatial location, and the maximum or average value is calculated within the zone to determine the corresponding value. Line 1 Backlight increment of column backlight zones The backlight increment is calculated using the following formula:

[0163] ;

[0164] In the above formula, Indicates the first Line 1 The incremental value of backlight enhancement for each backlight zone; Indicates the first Line 1 The set of pixel coordinates covered by the column backlight partitions on the LCD panel; This represents a preset backlight fine-tuning gain coefficient used to control the closed-loop convergence speed. For example, in one feasible embodiment... It can be set to 0.8.

[0165] After calculating the backlight increment covering all backlight zones, the backlight increment matrix is ​​superimposed on the suppressed backlight matrix generated in the previous steps to generate the updated target backlight matrix. The superposition calculation is shown in the following formula:

[0166] ;

[0167] In the above formula, Indicates the th element in the updated target backlight matrix Line 1 Backlight brightness values ​​for each column partition; This represents the backlight brightness value at the corresponding position in the original suppressed backlight matrix.

[0168] To maintain the continuity of the system's optical topology, the updated target backlight matrix is ​​re-convolved with the pre-stored point spread function in a second two-dimensional spatial operation to update the actual light intensity data on the LCD surface in real time. The calculation process is shown in the following formula:

[0169] ;

[0170] In the above formula, This indicates the updated actual light intensity data of the LCD surface in coordinates. The scalar value at the pixel; and These represent the pixel spacing distances between the centers of adjacent backlight zones on the horizontal and vertical sides of the LCD panel, respectively. This represents the point spread function mentioned earlier.

[0171] After updating the actual light intensity data of the LCD surface, the liquid crystal driving signal is recalculated using the updated actual light intensity. At this time, due to the moderate increase in the light intensity of the backlight base, the liquid crystal transmittance required by the pixels that were previously in a saturated state is correspondingly reduced, thereby forcibly pulling the estimated LCD driving signal of all pixels in the entire image back to the normal level. Within the following linear safe range of liquid crystal physical response.

[0172] Through the closed-loop control process of liquid crystal response saturation detection, threshold ratio determination, undercompensation amount extraction, backlight increment superposition, and secondary optical convolution update, the cutoff of dark highlight details and color distortion caused by excessive backlight suppression are avoided. Finally, the driving parameters within the safe range and the updated target data are passed to the subsequent steps.

[0173] S5. Based on the final target perceived brightness data and corrected chromaticity coordinate data, reverse map back to the XYZ space, and generate the final LCD driving signal through a three-dimensional lookup table.

[0174] Furthermore, in S5, generating the final LCD drive signal through a three-dimensional lookup table includes:

[0175] The initial chromaticity coordinate data corrected by the color shift compensation value is fused with the target perceived brightness after being superimposed by the perceived brightness compensation amount to construct the three-dimensional color space input coordinates.

[0176] The three-dimensional color space input coordinates are input into a pre-fixed hardware-level three-dimensional lookup table, and the liquid crystal transmittance correction coefficient and gamma correction data corresponding to the input coordinates are searched in parallel.

[0177] The correction coefficients and the gamma correction data are combined to directly generate and output the final LCD driving signal that conforms to the physical response characteristics of the current LCD panel.

[0178] Specifically, it receives the final target perceived brightness data, corrected chromaticity coordinate data, and the actual light intensity data of the LCD surface after closed-loop update, all confirmed in the previous step. This step, as the terminal hardware execution node of the entire display processing pipeline, aims to ensure that the data is within a perceptibly uniform color space (such as...). The high-dimensional luminance and color decoupled data (in space) are accurately reverse-mapped back to the XYZ tristimulus value space on which the physical display depends, and the final LCD driving signal (such as RGB grayscale driving level) that conforms to the physical response characteristics of the liquid crystal panel is quickly generated using a hardware-fixed 3D Look-Up Table (3D LUT).

[0179] First, the coordinate fusion and inverse space transformation logic of the color space is executed. The corrected chromaticity coordinate data (scalar components) after color cast compensation is extracted. and ), and the final target perceived brightness data (scalar component) after perception brightness compensation and adaptive backoff confirmation. (This involves) stitching and fusing the data according to the three-dimensional spatial dimensions to construct a system in... 3D color space input coordinate vector .

[0180] After constructing the three-dimensional coordinates, use The inverse transformation mathematical model maps the three-dimensional coordinates of the perceptual domain back to the absolute physical tristimulus value space (such as the CIE XYZ space). The inverse mapping process first de-nonlinearly compresses the brightness using the inverse function of the perceptual quantization function (PQ), recovering the perceptual tristimulus values ​​of the physical domain; then, it multiplies by the inverse linear transformation matrix to calculate the absolute tristimulus coordinates of the corresponding pixel in the XYZ space. The calculation process is shown in the following formula:

[0181] ;

[0182] In the above formula, Representing coordinates The physical tristimulus values ​​of a pixel in the CIE XYZ space; This indicates that the color gamut and white point parameters of the monitor have been calculated and determined in advance. Inverse transformation space matrix; subscript Indicates via The dimensionless sensory tristimulus vector after inverse function transformation.

[0183] The calculated XYZ tristimulus values, or the initial RGB signals obtained from their linear transformation, are used as 3D index addressing coordinates and input into a pre-pre-built hardware-level 3D lookup table (3D LUT). This 3D LUT is a 3D data mapping matrix that is pre-sampled and curve-fitted into the on-chip memory (SRAM / ROM) of the display driver IC (DDIC) after full sampling and curve fitting of the actual display light intensity and color deviation data of the LCD panel under different driving voltages and backlight intensities using a high-precision spectroradiometer.

[0184] Using hardware parallel addressing and a trilinear interpolation algorithm, the liquid crystal transmittance correction coefficients and gamma correction data corresponding to the three-dimensional input coordinates are quickly found within a single clock cycle. The lookup and interpolation process is calculated using the following formulas:

[0185] ;

[0186] ;

[0187] In the above formula, This represents the liquid crystal transmittance correction coefficient obtained from the lookup table; This represents the hardware gamma response correction parameters retrieved from the lookup table; and These represent the three-dimensional lookup mapping functions for transmittance and gamma correction, respectively.

[0188] Finally, the obtained liquid crystal transmittance correction coefficient and gamma correction data are synthesized and calculated, and combined with the closed-loop updated actual light intensity data of the LCD surface at the current pixel position, to directly synthesize and generate the final LCD driving signal that takes into account both dynamic backlight response and panel nonlinear characteristics (e.g., (Digital grayscale signal). The synthesis process is shown in the following formula:

[0189] ;

[0190] In the above formula, This represents the final RGB drive level scalar data output to the LCD panel source driver; This represents the luminance component in the target tristimulus value; This represents the actual physical light intensity data of the LCD surface after the aforementioned steps. and These represent the correction coefficients and gamma correction data obtained from the aforementioned steps, respectively.

[0191] Through the above-mentioned color space inverse conversion, three-dimensional lookup table parallel addressing interpolation, and hardware synthesis operation of correction data, the electro-optic nonlinear distortion of the LCD panel itself and the influence of backlight intensity fluctuation are eliminated, and a high-precision final LCD driving signal is output, thereby perfectly reproducing the image with high contrast, low halo and accurate color on the display terminal.

[0192] Based on the aforementioned low-power MiniLED backlight control method, this invention also provides a corresponding low-power MiniLED backlight control system. Please refer to [link to relevant documentation]. Figure 7 ,include:

[0193] The spatial transformation module performs salient gridding and sparse sampling on the input RGB image, and maps it to a lookup table. Perceive a uniform color space and output perceived brightness data and initial chromaticity coordinate data;

[0194] The backlight decision module is used to determine the target backlight matrix based on the perceived brightness data, introduce a suppression factor to calculate and generate a suppression backlight matrix, perform point spread function convolution calculation on the suppression backlight matrix, and output the actual light intensity data of the LCD surface.

[0195] The brightness and color decoupling compensation module is used to calculate the perceived brightness compensation amount based on the perceived brightness data and the actual light intensity data of the LCD surface, and to superimpose the perceived brightness compensation amount onto the perceived brightness data to output the target perceived brightness data; at the same time, it corrects the initial chromaticity coordinate data based on the actual light intensity data of the LCD surface and the nonlinear color shift function of the liquid crystal panel, and outputs the corrected chromaticity coordinate data.

[0196] The adaptive backoff module is used to estimate the LCD driving signal based on the target perceived brightness data and the corrected chromaticity coordinate data. If it is detected that the estimated LCD driving signal corresponding to more than a preset proportion of pixels enters the liquid crystal physical response saturation region, the backlight suppression matrix is ​​finely adjusted upward and the actual light intensity data of the LCD surface is updated so that the estimated LCD driving signal returns to the linear response safe range.

[0197] The reverse mapping module is used to reverse map back to the XYZ space based on the final target perceived brightness data and corrected chromaticity coordinate data, and generate the final LCD driving signal through a three-dimensional lookup table.

[0198] like Figure 8 As shown, based on the aforementioned low-power MiniLED backlight control method, this invention also provides an electronic device, which can be a mobile terminal, desktop computer, laptop, handheld computer, server, or other computing device. The electronic device includes a processor 910, a memory 920, and a display 930. Figure 8 Only some components of the electronic device are shown; however, it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0199] In some embodiments, the memory 920 may be an internal storage unit of the electronic device, such as a hard disk or memory. In other embodiments, the memory 920 may be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc. Furthermore, the memory 920 may include both internal and external storage units. The memory 920 is used to store application software and various types of data installed on the electronic device, such as program code installed on the electronic device. The memory 920 can also be used to temporarily store data that has been output or will be output. In one embodiment, the memory 920 stores a low-power MiniLED backlight control system 940, which can be executed by the processor 910 to implement a low-power MiniLED backlight control method according to various embodiments of this application.

[0200] In some embodiments, processor 910 may be a central processing unit (CPU), microprocessor, or other data processing chip, used to run program code stored in memory 920 or process data, such as executing a low-power MiniLED backlight control method.

[0201] In some embodiments, display 930 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. Display 930 is used to display information from the intelligent low-power MiniLED backlight control system and to display a visual user interface. Components 910-930 of the electronic device communicate with each other via a system bus.

[0202] Of course, those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.). The program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The storage medium can be a memory, magnetic disk, optical disk, etc.

[0203] The specific embodiments of the present invention described above do not constitute a limitation on the scope of protection of the present invention. Any other corresponding changes and modifications made in accordance with the technical concept of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A low-power MiniLED backlight control method, characterized in that, include: S1. Perform salient gridding and sparse sampling on the input RGB image, and map it to a lookup table. Perceive a uniform color space and output perceived brightness data and initial chromaticity coordinate data; S2, used to determine the target backlight matrix based on the perceived brightness data, introduce a suppression factor to calculate and generate a suppression backlight matrix, and perform point spread function convolution calculation on the suppression backlight matrix to output the actual light intensity data of the LCD surface. S3. Calculate the perceived brightness compensation amount based on the perceived brightness data and the actual light intensity data of the LCD surface, and superimpose the perceived brightness compensation amount onto the perceived brightness data to output the target perceived brightness data; at the same time, correct the initial chromaticity coordinate data based on the actual light intensity data of the LCD surface and the nonlinear color shift function of the liquid crystal panel, and output the corrected chromaticity coordinate data. S4. Based on the target perceived brightness data and corrected chromaticity coordinate data, estimate the LCD driving signal. If it is detected that the estimated LCD driving signal corresponding to more than a preset proportion of pixels enters the liquid crystal physical response saturation region, then finely adjust the suppression backlight matrix upward and update the actual light intensity data of the LCD surface, so that the estimated LCD driving signal returns to the linear response safe range. S5. Based on the final target perceived brightness data and corrected chromaticity coordinate data, reverse map back to the XYZ space, and generate the final LCD driving signal through a three-dimensional lookup table.

2. The low-power MiniLED backlight control method according to claim 1, characterized in that, In S1, the input RGB image is saliently meshed and sparsely sampled, including: The input RGB image is segmented into multiple image grids, and the saliency index of each image grid is calculated. The significance index is compared with a preset threshold. For image grids with a significance index less than or equal to the preset threshold, a portion of the pixels are extracted and sparsely sampled. For image grids with a significance index greater than the preset threshold, all pixels are collected. After spatial mapping, interpolation is performed on the sparsely sampled pixels to complete the data.

3. The low-power MiniLED backlight control method according to claim 1, characterized in that, In S1, mapping is performed via a lookup table. Perceiving a uniform color space includes: A one-dimensional logarithmic lookup table is used to perform non-linear perceptual quantization processing on the sampled pixel data to obtain perceptual quantization data. Multiply the perceived quantized data by a preset three-dimensional transformation matrix to complete the transformation. Perceive the mapping of a uniform color space.

4. The low-power MiniLED backlight control method according to claim 1, characterized in that, In S2, the calculation of the suppression factor to generate the suppression backlight matrix includes: The suppression factor is calculated based on the local contrast of each region of the input RGB image and the visual masking threshold. The initial backlight brightness of each MiniLED zone is determined based on the perceived brightness data to form the target backlight matrix; Multiply each element in the target backlight matrix by the corresponding suppression factor to obtain the suppression backlight matrix; The point spread function convolution calculation for the suppressed backlight matrix includes: Obtain the spatial optical attenuation distribution matrix formed by the emission of a single LED on the LCD surface in the MiniLED backlight module, and use it as the point spread function; The suppression backlight matrix and the point spread function are subjected to two-dimensional spatial convolution operation to calculate and generate the actual light intensity data of the LCD surface.

5. The low-power MiniLED backlight control method according to claim 1, characterized in that, In S3, the calculation of the perceived brightness compensation amount based on the perceived brightness data and the actual light intensity data of the LCD surface includes: The actual light intensity data of the LCD surface is input into a preset LCD liquid crystal panel transmittance and electro-optic response model, and the normalized calculation is used to obtain the actual display light intensity of the LCD surface. use The perception quantization function in space performs nonlinear compression on the actual displayed light intensity to obtain the actual perceived brightness. The target difference is obtained by subtracting the actual perceived brightness from the perceived brightness data, and the target difference is then adjusted by weighted filtering gain to obtain the perceived brightness compensation amount. The initial chromaticity coordinate data is corrected based on the actual light intensity data of the LCD surface and the nonlinear color shift function of the LCD panel, including: Real-time acquisition of actual light intensity data of the LCD surface and temperature feedback data of the backlight module; The actual light intensity data of the LCD surface and the temperature feedback data are simultaneously input into a pre-fitted multivariate polynomial nonlinear color shift model to calculate the color deviation caused by the wavelength drift of the blue light chip and the change in phosphor excitation efficiency. The initial chromaticity coordinate data is compensated by offset superposition based on the chromaticity deviation to generate corrected chromaticity coordinate data.

6. The low-power MiniLED backlight control method according to claim 1, characterized in that, In S4, the estimated LCD driving signal corresponding to pixels exceeding a preset ratio enters the liquid crystal physical response saturation region, including: Based on the perceived brightness compensation amount, the estimated LCD driving signal corresponding to each pixel is calculated and generated. The amplitude of each estimated LCD driving signal is compared with the preset liquid crystal molecule transmittance saturation threshold, and the number of pixels exceeding the saturation threshold is counted. Calculate the ratio of the number of pixels to the total number of pixels in the current region of the image. When the ratio is greater than a preset ratio threshold, determine that the estimated LCD driving signal has entered the liquid crystal physical response saturation region and trigger a backlight fine-tuning request. Fine-tuning the suppressed backlight matrix and updating the actual light intensity data of the LCD surface includes: In response to the backlight fine-tuning request, obtain the brightness undercompensation amount of pixels that exceed the saturation threshold; Calculate the backlight increment based on the brightness undercompensation amount, and superimpose the backlight increment onto the corresponding suppressed backlight matrix to generate an updated target backlight matrix. The updated target backlight matrix is ​​then re-convolved with the point spread function in a second two-dimensional spatial operation to update the actual light intensity data of the LCD surface in real time.

7. The low-power MiniLED backlight control method according to claim 1, characterized in that, In S5, generating the final LCD drive signal using a three-dimensional lookup table includes: The initial chromaticity coordinate data corrected by the color shift compensation value is fused with the target perceived brightness after being superimposed by the perceived brightness compensation amount to construct the three-dimensional color space input coordinates. The three-dimensional color space input coordinates are input into a pre-fixed hardware-level three-dimensional lookup table, and the liquid crystal transmittance correction coefficient and gamma correction data corresponding to the input coordinates are searched in parallel. The correction coefficients and the gamma correction data are combined to generate and output the final LCD driving signal that conforms to the physical response characteristics of the current LCD panel.

8. A low-power MiniLED backlight control system, characterized in that, include: The spatial transformation module performs salient gridding and sparse sampling on the input RGB image, and maps it to a lookup table. Perceive a uniform color space and output perceived brightness data and initial chromaticity coordinate data; The backlight decision module is used to determine the target backlight matrix based on the perceived brightness data, introduce a suppression factor to calculate and generate a suppression backlight matrix, perform point spread function convolution calculation on the suppression backlight matrix, and output the actual light intensity data of the LCD surface. The brightness and color decoupling compensation module is used to calculate the perceived brightness compensation amount based on the perceived brightness data and the actual light intensity data of the LCD surface, and to superimpose the perceived brightness compensation amount onto the perceived brightness data to output the target perceived brightness data; at the same time, it corrects the initial chromaticity coordinate data based on the actual light intensity data of the LCD surface and the nonlinear color shift function of the liquid crystal panel, and outputs the corrected chromaticity coordinate data. The adaptive backoff module is used to estimate the LCD driving signal based on the target perceived brightness data and the corrected chromaticity coordinate data. If it is detected that the estimated LCD driving signal corresponding to more than a preset proportion of pixels enters the liquid crystal physical response saturation region, the backlight suppression matrix is ​​finely adjusted upward and the actual light intensity data of the LCD surface is updated so that the estimated LCD driving signal returns to the linear response safe range. The reverse mapping module is used to reverse map back to the XYZ space based on the final target perceived brightness data and corrected chromaticity coordinate data, and generate the final LCD driving signal through a three-dimensional lookup table.

9. An electronic device, characterized in that, include: Processor and memory; The memory stores a computer-readable program that can be executed by the processor; When the processor executes the computer-readable program, it implements the steps in the low-power MiniLED backlight control method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores one or more programs, which can be executed by one or more processors to implement the steps in the low-power MiniLED backlight control method as described in any one of claims 1-7.

Images