High-contrast Mini LED control method and system and medium

Abstract

The invention discloses a control method and system of a high-contrast Mini LED and a medium, and relates to the technical field of Mini LED display control, and the control method of the high-contrast Mini LED ensures uniform distribution of luminosity on a screen by subdividing luminosity data and adjusting luminosity data on an adjacent boundary; meanwhile, the accuracy of the luminosity and the chromatic aberration is adjusted, so that the problem of unnatural color transition or obvious chromatic aberration is avoided; the luminosity regulation and control model is finally constructed, so that more accurate luminosity coordination transition can be realized, the display effect is more real and fine, and the method is suitable for various display scenes; the accurate control and optimization of the luminosity of the Mini LED backlight area are integrally realized; each link solves the problems of uneven luminosity, color distortion and unnatural brightness transition possibly occurring in display in different dimensions, and particularly plays a key role in reducing the halo effect.

Description

Technical Field

[0001] The present invention relates to the field of MiniLED display control technology, and in particular to a control method, system and medium for high-contrast MiniLED. Background Art

[0002] MiniLED technology has made continuous breakthroughs in recent years and has become one of the important representatives of new display technologies in the 21st century. As this technology continues to mature, MiniLED displays have gradually become widely used information products around the world due to their long life and excellent brightness. From home appliances, mobile phones, televisions, computer monitors to automobiles, smart wearable devices and other fields, MiniLED displays have demonstrated their superior technical advantages and have been widely recognized and praised by consumers and industry insiders. The brightness of MiniLED display screen is particularly outstanding. It has higher brightness and more delicate display effect than traditional LED display technology, and can provide richer colors and stronger visual impact, which makes it an increasingly popular choice in high-end monitors, TVs and other devices that require high-quality images. In addition, the long service life of MiniLED is also one of its important advantages. Compared with traditional display screens, it can effectively extend the service life of the equipment, reduce the replacement frequency, and improve the overall cost performance; With the rapid development of 5G technology at home and abroad, many emerging technologies have been accelerated, and display technologies including MiniLED have also ushered in new development opportunities. The high-speed transmission and low-latency characteristics of 5G have greatly improved the speed and stability of image transmission, providing better support and guarantee for high-resolution and high-contrast display effects. However, despite this, MiniLED direct display technology still faces certain challenges in terms of high-contrast and ultra-high-resolution display effects. To achieve truly breakthrough display effects, greater efforts need to be made in technological innovation and process optimization, especially in improving screen contrast, resolution, and color accuracy. There is still a lot of room for improvement. Therefore, although the application prospects of MiniLED technology are broad, the realization of the ultimate display effect still requires continuous research and development and exploration within the industry.

[0003] In summary, how to regulate the brightness of MiniLED display to coordinate the transition between brightness and darkness and reduce the halo effect is an issue that needs to be urgently solved and optimized in the control system of high-contrast MiniLED. Summary of the invention

[0004] The present invention provides a high-contrast MiniLED control method, system and medium to solve the technical problem of how to regulate the MiniLED display luminosity in the related art to coordinate the transition between brightness and darkness and reduce the halo effect.

[0005] In order to solve the above technical problems, the present invention provides a control method, system and medium for high-contrast MiniLED. The specific technical solutions are as follows: In a first aspect, a high contrast MiniLED control method comprises the following steps: Collecting multiple time-frame MiniLED screen display photometric data to obtain a photometric data set; based on the photometric data set, dividing the MiniLED backlight area into multiple sub-areas to obtain a photometric data subset; based on the photometric data subset, subdividing the photometric data of the multiple sub-areas into multiple small-sized backlight blocks one by one, marking the multiple backlight blocks to obtain a photometric marking data set; Acquire the brightness intensity data of each data item in the photometric label data set to obtain a brightness intensity data set; smooth the neighboring boundaries between adjacent data items in the photometric label data set according to the differences between the brightness intensity data items to obtain a first control data set; Acquire photometric color scale data of each data item in the photometric label data set to obtain a photometric color scale data set; based on the photometric label data set after smoothing, adjust the accuracy of the neighboring boundaries between adjacent data items in the photometric label data set according to the color difference between the photometric color scale data to obtain a second control data set; The photometric labeling data set, the first control data set and the second control data set are trained to construct a photometric control model to output a photometric control standard recognition result between adjacent data items in the photometric labeling data set; at least two other data items in the photometric labeling data set are input into the photometric control model to output a photometric control prediction result between other adjacent photometric labeling data items; Based on the prediction results of the photometric control model, the photometric transition between the data items in the photometric labeling data set is coordinated to reduce the halo effect of the MiniLED screen display.

[0006] As a further optimization solution of the present invention, based on the photometric data set, the MiniLED backlight area is divided into a plurality of sub-areas to obtain a photometric data subset, including: Preprocessing the photometric data set to obtain a preprocessed photometric data set; the photometric data set includes an average brightness value of each pixel in the backlight area; Divide the areas of the photometric data items in the preprocessed photometric data set into regions according to spatial distribution by setting photometric thresholds, thereby dividing the backlight area into a plurality of sub-areas; the sub-areas are further divided into bright areas and dark areas according to the high and low photometric values; The bright area and the dark area select K initial center points for division according to the luminance value distribution of each sub-area; each pixel point of the sub-area is assigned to the initial center point closest to obtain the division center point of each cluster; and the classification results of the bright area and the dark area are updated based on the division center point until the classification results converge; Based on the classification convergence result, the photometric data set is finally classified into the bright area and the dark area to obtain the photometric data subset.

[0007] As a further optimization scheme of the present invention, based on the photometric data subset, the photometric data of multiple sub-regions are subdivided into multiple small-sized backlight blocks one by one, and the multiple backlight blocks are marked to obtain a photometric marked data set, including: Dividing the photometric data subset into a plurality of square areas of equal size by a 1x1 sliding window with a step size of 1; Each of the divided square areas is a backlight block; the backlight block covers different photometric data parts of the photometric data subset; the backlight block represents a block area of ​​uniform size in the bright area or the dark area in the sub-area; Extracting the average illumination intensity of each backlight block to measure the degree of brightness change of the backlight block and capturing the texture features of the brightness change; marking each backlight block based on the extracted multiple features to obtain a brightness marking data set; The photometrically labeled dataset is assigned multiple labels according to the labeled target features; the multiple labels include highlight, low light, shadow or normal lighting.

[0008] As a further optimization scheme of the present invention, the neighboring boundaries between adjacent data items in the photometric labeling data set are smoothed according to the difference between the light intensity data items to obtain a first control data set, including: Based on the difference in photometric intensity between adjacent data items in the photometric labeling data set, the transition anomaly at the photometric intersection between the adjacent data items is identified to obtain the neighboring boundary between the adjacent data items; The adjacent data items represent adjacent backlight blocks marked in the photometric marking data set; the adjacent boundary line represents the common boundary line between adjacent backlight blocks, that is, the backlight block cutting line in the photometric data subset; The neighboring boundary includes two dimensions, vertical or horizontal, to obtain the transition of the luminance intensity of multiple neighboring boundaries. When the luminance intensity transition between two adjacent luminance mark data items changes sharply and suddenly; the luminance intensity transition of the sudden change is not smooth, resulting in a halo effect on the screen display; The luminous intensity with drastic transition mutations is smoothed to obtain the first regulation data set.

[0009] As a further optimization scheme of the present invention, the luminous intensity with a sharp transition mutation is smoothed to obtain the first control data set, including: Set the adjacent data items I(x, y) and I(x', y') to represent the light intensity data on both sides of the adjacent boundary respectively; To obtain the light intensity difference value; where, Indicates the difference in luminosity intensity; Set the intensity threshold for drastic changes ,when , to determine that the difference in the photometric intensity values ​​will cause a drastic transition mutation; by To make the luminous intensity of the sudden change smooth; where, represents the luminous intensity value after smoothing, G(i,j,σ) represents the smoothing weight coefficient, σ controls the standard deviation of the smoothing degree, and I(x+i,y+j) represents the luminous intensity value in the neighborhood (i,j); After the smoothing process is completed, the drastic changes in the photometric intensity in the photometric labeling data set have been effectively suppressed, and the generated smoothed data set is the first regulated data set; by comparing the two adjacent photometric labeling data items before and after the process, whether a halo effect is generated, if not, execute the next step, if so, return to the previous step and reprocess; to evaluate the smoothing effect.

[0010] As a further optimization scheme of the present invention, the neighboring boundaries between adjacent data items in the photometric labeling data set are adjusted according to the color difference between the photometric color scale data to obtain a second control data set, including: There are three tones according to the color scale of the color space; based on the three tones, two adjacent data items in the photometric labeling data set are represented as (R, G, B) and (r, g, b) respectively; According to the three said hues, by To obtain the hue difference measurement value of two adjacent photometric labeled data items; preset the color difference threshold ,when Indicates that the color difference is too large; Where ΔE represents the measure of hue difference, R and r represent the brightness of two adjacent photometric labeled data items, G and g represent the green-red, B and b represent the blue-yellow components of the color respectively; By analyzing the hue difference measurement value between two adjacent photometrically marked data items, the light scheduling of the adjacent boundary position is finely adjusted to eliminate the sudden transition of the color difference between the two adjacent photometrically marked data items.

[0011] As a further optimization scheme of the present invention, by analyzing the hue difference measurement value between two adjacent photometric mark data items, the light scheduling of the adjacent boundary position is finely adjusted to eliminate the sudden transition of the color difference between the two adjacent photometric mark data items, including: when , α is close to 1, indicating that the transition of the hue difference measure is natural; , α is close to 0, indicating that the hue difference measurement transition mutation is drastic and needs to be adjusted by precision; where α represents the precision adjustment weight factor, Based on the brightness intensity between two adjacent photometric labeled data items, To adjust the accuracy of the hue difference measurement value between two adjacent photometrically labeled data items; where, Indicates the adjustment accuracy value of the hue difference measurement value. α is adjusted according to the size of the color difference ΔE to eliminate the transitional color difference mutation; After precision adjustment, the hue difference between adjacent data items in the photometrically labeled data set is naturally transitioned, and finally a second regulated data set is generated; To evaluate the accuracy adjustment effect; where w a represents the precision adjustment evaluation value, β represents the evaluation factor, which is used to evaluate the sensitivity of weight adjustment. The larger ΔE is, the greater the sensitivity of w a The smaller it is, the stronger the precision adjustment processing is in this area.

[0012] As a further optimization solution of the present invention, the light intensity control model includes: Integrate the first control data set and the second control data set into a data set, generate structure data from the historically acquired data set, and encode the structure data into sequence data to train a photometric control model; Inputting the sequence data into the photometric control model; the photometric control model comprises an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer, transmitting the intermediate representation data of multiple hidden layers to the output layer, and the output layer outputting the photometric control standard recognition result between two adjacent data items in the photometric labeling data set; Two adjacent data items in the photometrically labeled data set are input into the photometric control model to output a newly acquired output layer output representing a photometric control prediction result between a next pair of adjacent photometrically labeled data items.

[0013] In a second aspect, a high contrast MiniLED control system is provided, wherein an electronic device is provided, including a memory, a processor, and a high contrast MiniLED control method program stored in the memory and executable on the processor, wherein the high contrast MiniLED control method program implements steps of a high contrast MiniLED control method when executed by the processor, and the system includes: Data acquisition module: used to collect the MiniLED screen display photometric data of multiple time frames to obtain a photometric data set; based on the photometric data set, the MiniLED backlight area is divided into multiple sub-areas to obtain a photometric data subset; based on the photometric data subset, the photometric data of the multiple sub-areas are subdivided into multiple small-sized backlight blocks one by one, and the multiple backlight blocks are marked to obtain a photometric marked data set; Data processing module: used to obtain the brightness intensity data of each data item in the photometric labeling data set to obtain a brightness intensity data set; smooth the neighboring boundaries between adjacent data items in the photometric labeling data set according to the difference between the brightness intensity data items to obtain a first control data set; obtain the photometric color scale data of each data item in the photometric labeling data set to obtain a photometric color scale data set; based on the photometric labeling data set after the smoothing process, adjust the accuracy of the neighboring boundaries between adjacent data items in the photometric labeling data set according to the color difference between the photometric color scale data to obtain a second control data set; A photometric control module: used to train the photometric labeling data set, the first control data set and the second control data set to construct a photometric control model, so as to output a photometric control standard recognition result between adjacent data items in the photometric labeling data set; input at least two other data items in the photometric labeling data set into the photometric control model, so as to output a photometric control prediction result between other adjacent photometric labeling data items; A control and evaluation module; it is used to coordinate the luminance transition between the data items in the luminance marking data set based on the prediction results of the luminance control model to reduce the halo effect of the MiniLED screen.

[0014] In a third aspect, a computer-readable storage medium stores a program, which, when executed by a processor, implements the steps of a control method for a high-contrast MiniLED.

[0015] The present invention has at least the following beneficial effects: the present invention collects the MiniLED screen display luminosity data in multiple time frames; thus, it can accurately reflect the luminosity changes of the screen at different time points. This provides a more comprehensive luminosity data set for subsequent luminosity control, ensuring the accuracy and timeliness of luminosity control; the entire backlight area is divided into multiple sub-areas, which can locally handle luminosity problems instead of making uniform adjustments to the entire area. Such subdivision can effectively identify luminosity changes in different areas, thereby making more precise adjustments and reducing display problems caused by local unevenness.

[0016] By subdividing the sub-region luminance data into multiple small-sized backlight blocks and marking them; further subdividing the backlight blocks and marking each small-sized backlight block, the control accuracy of luminance changes can be improved, the differences between different regions can be reduced, and the consistency of display effects can be improved. This step provides the necessary basic data for subsequent precision adjustments.

[0017] Obtain the light intensity data set and perform boundary smoothing; by obtaining the light intensity data set of each backlight block, you can more intuitively understand the brightness changes in each area. Smoothing the boundary is to reduce the brightness difference between backlight blocks, making the transition of brightness changes more natural, thereby avoiding halo effects and abrupt brightness differences and improving the visual experience.

[0018] Acquire the photometric color scale data set and adjust the accuracy of the neighboring boundaries; by obtaining the photometric color scale data set and adjusting the accuracy according to the color difference, it helps to ensure that the photometric control not only considers the brightness difference, but also finely adjusts the color transition. This process avoids color unevenness or obvious color difference, so that the image maintains high-quality visual performance during the display process.

[0019] Training the photometric control model, constructing the photometric control standard recognition and prediction results; by training the photometric control model, combining the photometric labeling data set, the first control data set and the second control data set, a control model with predictive capabilities can be constructed. This model can identify the differences between adjacent photometric data items and predict how to control the photometric to minimize the halo effect and uneven brightness.

[0020] Based on the prediction results of the photometric control model, the photometric transition is coordinated; through the prediction results of the photometric control model, the photometric between each data item in the photometric labeling data set can be coordinated to ensure a smoother transition between photometric transitions. This step helps reduce unnatural transitions between backlight blocks, optimizes display effects, and effectively reduces the halo effect common in MiniLED displays, improving visual quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1is a schematic flow chart of a high-contrast MiniLED control method provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of a control system of a high-contrast MiniLED provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The present application is further described in detail below in conjunction with the accompanying drawings. It is necessary to point out here that the following specific implementation methods are only used to further illustrate the present application and cannot be understood as limiting the scope of protection of the present application. Technical personnel in this field can make some non-essential improvements and adjustments to the present application based on the above application content.

[0023] This embodiment provides a high-contrast MiniLED control method, system and medium, and the specific implementation methods are as follows: like Figure 1 As shown, a high contrast MiniLED control method includes the following steps: Step 11, collecting the MiniLED screen display photometric data of multiple time frames to obtain a photometric data set; based on the photometric data set, dividing the MiniLED backlight area into multiple sub-areas to obtain a photometric data subset; based on the photometric data subset, subdividing the photometric data of the multiple sub-areas into multiple small-sized backlight blocks one by one, marking the multiple backlight blocks to obtain a photometric marking data set; Step 12, obtaining the brightness intensity data of each data item in the photometric labeling data set to obtain a brightness intensity data set; smoothing the neighboring boundaries between adjacent data items in the photometric labeling data set according to the differences between the brightness intensity data items to obtain a first control data set; obtaining the photometric color scale data of each data item in the photometric labeling data set to obtain a photometric color scale data set; based on the photometric labeling data set after the smoothing process, adjusting the accuracy of the neighboring boundaries between adjacent data items in the photometric labeling data set according to the color difference between the photometric color scale data to obtain a second control data set; Step 13, training the photometric labeling data set, the first control data set and the second control data set to construct a photometric control model, so as to output a photometric control standard recognition result between adjacent data items in the photometric labeling data set; inputting at least two other data items in the photometric labeling data set into the photometric control model, so as to output a photometric control prediction result between other adjacent photometric labeling data items; Step 14, based on the prediction result of the photometric control model, coordinate the photometric transition between the data items in the photometric labeling data set to reduce the halo effect of the MiniLED screen display.

[0024] In an embodiment of the present invention, step 11 obtains different brightness (luminosity) of each MiniLED display screen in different time frames on the MiniLED display screen, and collects luminosity data of the display screen through multiple time frames (time periods); these data are usually collected through a luminosity sensor or an image sensor, and these data sets contain brightness information of different areas of the entire screen at different time points; through multi-time frame data collection, the brightness performance of the screen in different time periods can be fully understood, so as to perform luminosity analysis more accurately; capturing luminosity data of multiple time frames helps to discover potential brightness unevenness or dynamic changes, thereby providing a reliable data basis for subsequent luminosity control models.

[0025] Based on the photometric data set, the MiniLED backlight area is divided into multiple sub-areas to obtain a subset of photometric data; based on the collected photometric data set, the backlight area of ​​the entire display screen will be divided into multiple sub-areas, each sub-area corresponding to a different subset of photometric data. This division can be performed according to the physical area of ​​the screen (for example, a rectangle or a grid), or it can be divided non-uniformly according to the characteristics of the photometric distribution; dividing the backlight area into multiple sub-areas can improve the accuracy of photometric control. The photometric performance of different areas may be different, and after subdivision, the brightness of each area can be adjusted more finely; through this division, it can be ensured that local photometric problems (for example, the brightness of a certain area is too high or too low) can be accurately identified and optimized.

[0026] Each sub-area is further subdivided into multiple small-sized backlight blocks, which represent smaller areas, and each backlight block has corresponding photometric data. By adding a tag (for example, number or position) to each backlight block, a labeled data set containing the photometric data of each backlight block is formed; by subdividing the backlight blocks, the brightness of each small block can be adjusted more finely, reducing the brightness unevenness of the display; the labeling of the backlight blocks provides structured information for subsequent data processing and model training, making the photometric control in subsequent steps more systematic and efficient.

[0027] In step 12, a light intensity data set is formed based on the light intensity information of each data item in the photometric labeling data set obtained in step 11. The light intensity represents the brightness value of each backlight block. Then, according to the changing trend of the brightness value, the neighboring boundaries between adjacent data items are smoothed. The neighboring boundaries refer to the brightness transition areas between adjacent backlight blocks, and the smoothing process adjusts these transition areas to be more natural and smooth according to the brightness differences; after the smoothing process, the sudden change of the brightness of the display screen is avoided, the visually unnatural brightness jump is reduced, and the viewing comfort is improved; through the generated first control data set, the brightness change of each backlight block can be accurately controlled, making the brightness transition of the display screen softer and more natural.

[0028] Step 12 extracts the photometric color scale (i.e., the range of color temperature or color gamut) of each data item from the photometric labeling data set to form a photometric color scale data set. According to the difference between the color scales, the accuracy of the neighboring boundaries is adjusted. This adjustment is made based on the color scale difference of adjacent data items, aiming to improve the accuracy of color transition and avoid the abruptness of color transition; through accuracy adjustment, the color transition on the display screen can be smoother and more accurate, avoiding obvious jumps or deviations in color and improving color continuity; this adjustment helps to achieve a high-quality visual experience, especially when displaying high dynamic range (HDR) content.

[0029] Step 13 uses a machine learning method to train a photometric control model based on the photometric labeling data set, the first control data set, and the second control data set obtained in step 11 and step 12. During the training process, the model will learn the relationship between the photometric data and generate photometric control standards between adjacent data items. Then, the photometric control model is used to predict other data items in the photometric labeling data set to obtain the control prediction results between adjacent photometric data items; the trained photometric control model can automatically identify and predict photometric control needs, greatly improving the efficiency and accuracy of control; through this model, intelligent processing of photometric control can be achieved, reducing manual intervention and adapting to different display needs.

[0030] Step 14 In this step, based on the prediction results of the luminosity control model constructed in step 13, the luminosity between the data items in the luminosity labeling data set is coordinated and transitioned. Specifically, according to the output of the model, the brightness of adjacent backlight blocks is fine-tuned to eliminate the "halo effect" caused by uneven brightness or uneven boundary transition. The halo effect usually manifests itself as an unnatural transition of brightness or light overflow, which affects the quality of the display effect; by finely controlling the luminosity transition, the halo effect can be effectively reduced or eliminated, making the brightness distribution of the display more uniform; this coordinated transition can improve the overall visual effect of the display and enhance the user's viewing experience, especially in high-contrast displays or dark field displays, the reduction of the halo effect is particularly important.

[0031] In a preferred embodiment of the present invention, in the above step 11, based on the photometric data set, the MiniLED backlight area is divided into a plurality of sub-areas to obtain a photometric data subset, including: Step 111, preprocessing the photometric data set to obtain a preprocessed photometric data set; the photometric data set includes an average brightness value of each pixel in the backlight area; Step 112, dividing the areas of the luminance data items in the preprocessed luminance data set according to spatial distribution by setting luminance thresholds, thereby dividing the backlight area into a plurality of sub-areas; the sub-areas are further divided into bright areas and dark areas according to the luminance values; Step 113, the bright area and the dark area select K initial center points for division according to the luminance value distribution of each sub-area; assign each sub-area pixel point to the initial center point with the closest distance to obtain the division center point of each cluster; update the classification results of the bright area and the dark area based on the division center point until the classification results converge; Step 114 , based on the classification convergence result, the photometric data set is finally classified into the bright area and the dark area to obtain the photometric data subset.

[0032] In the embodiment of the present invention, in step 111, the photometric data in the image (i.e., the brightness information of each pixel) is first collected and sorted. These data may represent the average brightness value of each pixel in the backlight area of ​​the image. The preprocessing process may include operations such as denoising, normalization, and filtering, which are intended to reduce errors and noise in the original data and improve the accuracy of subsequent analysis; reduce noise in the data and enhance effective information; provide clearer and more accurate photometric data for subsequent steps; and ensure that the data is balanced in different photometric ranges so that the processing of subsequent steps is more accurate.

[0033] Step 112 sets a threshold for the photometric data items in the photometric data set preprocessed in step 111 according to the spatial distribution and photometric value, and uses this threshold to divide the backlight area. In this way, the backlight area is divided into multiple sub-areas, and each sub-area is further divided into "bright area" and "dark area" according to the level of its photometric value; the setting of the photometric threshold can be based on a certain standard, such as average brightness, median or distribution characteristics; the divided "bright area" and "dark area" represent different light intensity areas, which can more accurately express different brightness areas in the image; make the backlight area more refined, and can handle more complex light distribution conditions; dividing into multiple sub-areas can facilitate subsequent analysis, especially distinguishing different brightness areas; it helps to process or optimize the high brightness area and the low brightness area separately, which is very important in tasks such as image enhancement.

[0034] In step 113, photometric data category analysis is performed by selecting K initial center points for division. These initial center points are usually selected by analyzing the distribution of photometric values ​​of each sub-region in the bright and dark regions, with the goal of representing the center points of regions with different brightness. Then, the pixels of each sub-region are assigned to the nearest initial center point, so that each cluster forms a cluster, called a "cluster" or "category"; based on these initial division center points, the pixels of each sub-region are iteratively updated until the classification results converge; used to find the optimal classification results for each sub-region; can accurately identify and divide the bright and dark regions in the image, and perform efficient image segmentation; thereby updating to help distinguish similar regions, thereby accurately reflecting the spatial distribution of regions with different brightness; through multiple iterations, ensure that the final classification results are stable and reduce errors.

[0035] Step 114 further determines the final regional classification result based on the convergence result of classification based on step 111, step 112 and step 113. At this stage, the photometric data set of the backlight area is accurately divided into different bright and dark areas, and the final photometric data subset is formed; the final clear classification result is obtained, which provides a basis for subsequent image processing or display optimization; the accurate division of bright and dark areas can help more complex image processing, such as contrast enhancement, brightness adjustment, etc.; further improve the visual effect of the image, making it more in line with the visual habits of the human eye, especially in environments with different brightness.

[0036] The above steps work together to provide a clean and standardized data foundation for subsequent regional division and cluster analysis through step 111. Without preprocessing, the subsequent classification and regional division may be affected by noise, resulting in reduced accuracy; step 112 relies on the preprocessing results of step 111, and further subdivides the region by setting the photometric threshold, providing a more refined image region to facilitate subsequent classification processing; step 113 selects a suitable initial center point for the divided region of step 112 by combining the photometric value distribution, and classifies the pixels according to these points, providing a key step for the final image segmentation; step 114 completes the final regional division based on the results of the previous steps 111, 112 and 113, ensuring a clear definition of bright and dark areas, and providing a useful photometric data subset for practical applications.

[0037] In a preferred embodiment of the present invention, in the above step 11, based on the photometric data subset, the photometric data of the plurality of sub-regions are subdivided into a plurality of small-sized backlight blocks one by one, and the plurality of backlight blocks are marked to obtain a photometric marked data set, including: Step 115, dividing the photometric data subset into a plurality of square areas of equal size by a 1x1 sliding window with a step size of 1; Step 116, each of the divided square areas is a backlight block; the backlight block covers different luminance data parts of the luminance data subset; the backlight block represents a block area of ​​uniform size of the bright area or the dark area in the sub-area; Step 117, extracting the average illumination intensity of each backlight block to measure the degree of brightness change of the backlight block and capture the texture features of the brightness change; based on the extracted multiple features, marking each backlight block to obtain a brightness marking data set; In step 118, the photometrically labeled dataset is assigned a plurality of labels according to the labeled target features; the plurality of labels include highlight, low light, shadow or normal light.

[0038] In the embodiment of the present invention, step 115 traverses the image or the photometric data subset by using a sliding window of size 1x1 and a step size of 1 (i.e., sliding one pixel at a time); the photometric data subset is divided into a number of small square areas of equal size by the sliding window method. These small square areas are called "backlight blocks"; the step size of the sliding window is 1, which means that only one pixel is processed each time the window is moved, which helps to maintain the details of the photometric data and avoid over-smoothing; it helps to accurately divide the image area and ensure that each backlight block can cover different parts of the photometric data subset in detail; the fine division can provide more information in the subsequent feature extraction process, which helps to more accurately capture the photometric changes in the image.

[0039] Based on step 115, each square area (backlight block) divided by step 116 covers a different part of the luminance data subset; each backlight block has a uniform size, regardless of the light intensity in the area. This means that even in an area with drastic changes in light intensity, the area of ​​each backlight block remains the same, maintaining structural consistency; this uniform size division method simplifies subsequent processing because the luminance data features of each block are relatively independent and consistent; thereby avoiding complex regional deformations, which helps to extract and compare changes in light intensity in subsequent analysis.

[0040] Based on step 116, step 117 calculates the average light intensity of the pixels in each backlight block, that is, the light intensity of all pixels in the area is averaged; by obtaining the average light intensity of the backlight block, the lighting information of the area can be obtained, and then the degree of change in brightness can be measured. At the same time, by analyzing the texture of these backlight blocks, the lighting change pattern in the image can be captured, such as shadows, reflections and other features; the extracted average light intensity provides a way to quantify the brightness change, which helps to further analyze the lighting characteristics of the image; capturing texture features helps to reveal the details of lighting changes, such as the existence of highlights, low lights and other areas, providing a basis for subsequent marking and classification.

[0041] Step 118 labels each backlight block according to the photometric features (such as average light intensity and texture features) extracted in step 117; assigning different labels to each backlight block, and common labels include "highlight", "low light", "shadow" and "normal light". Highlights: Usually refer to areas with strong light.

[0042] Low light: refers to areas with low light intensity.

[0043] Shadow: refers to the area where light is blocked, which may have lower light intensity or special texture.

[0044] Normal lighting: refers to the regular light intensity in most areas without excessive light variations.

[0045] Through this labeling method, the illumination data can be converted into information with clear classification, providing valuable input for subsequent image analysis (such as enhancement, correction or generation); labeled illumination data can help the automation system understand the lighting conditions in the image, thereby providing useful feature support for the visual tasks of MiniLED screen display (such as image enhancement, 3D reconstruction, etc.).

[0046] In a preferred embodiment of the present invention, in the above step 12, the boundary lines between adjacent data items in the photometric labeling data set are smoothed according to the difference between the light intensity data items to obtain the first control data set, including: Step 121, based on the difference in photometric intensity between adjacent data items in the photometric labeling data set, the transition anomaly at the photometric intersection between the adjacent data items is identified to obtain the neighboring boundary between the adjacent data items; Step 122, the adjacent data items represent adjacent backlight blocks marked in the photometric marking data set; the adjacent boundary line represents a common boundary line between adjacent backlight blocks, that is, a backlight block dividing line in the photometric data subset; Step 123, the neighboring boundary includes two dimensions, vertical or horizontal, to obtain the transition of the luminance intensity of multiple neighboring boundaries. When the luminance intensity transition between two adjacent luminance mark data items changes suddenly; the sudden change in luminance intensity transition is not smooth, resulting in a halo effect on the screen display; Step 124, smoothing the luminous intensity with a sharp transition mutation to obtain the first control data set.

[0047] In the embodiment of the present invention, in step 121, it is first necessary to calculate the photometric difference between adjacent data items in the photometric labeling data set; adjacent data items refer to backlight blocks that have been marked with lighting features. The photometric data subset has been divided into many small backlight blocks through the previous steps, and each backlight block has its corresponding light intensity label (such as highlight, low light, shadow, etc.); when the photometric difference between adjacent backlight blocks is too large (that is, the light intensity changes abnormally), this photometric difference is considered to be a "transition anomaly"; the detection of transition anomalies can be achieved by calculating the photometric difference between adjacent backlight blocks, and marking the place where the photometric change suddenly changes as a "neighboring boundary"; help identify transition abnormal areas in the image caused by sudden changes in photometric changes, especially those boundaries that are visually prone to unnatural transitions; in image processing, identifying photometric transition anomalies helps subsequent photometric smoothing operations, thereby reducing inappropriate visual effects (such as unnatural brightness jumps).

[0048] In step 122, the adjacent data items refer to the backlight blocks that have been marked in the photometric labeling data set obtained in step 121. These backlight blocks are adjacent and have obvious differences in photometric characteristics; by comparing the photometric changes between adjacent backlight blocks, their boundaries are determined, and these boundaries are defined as "neighboring boundaries"; neighboring boundaries are the dividing lines between adjacent backlight blocks, which mark the intersection of two areas with different light intensities. Usually these boundary lines appear as dividing lines in the image; marking neighboring boundaries helps to further analyze the boundary areas of lighting changes, especially those places where the photometric changes suddenly change; determining the position of neighboring boundaries provides a clear reference for subsequent photometric smoothing and visual effect optimization, which helps to reduce hard boundaries and sudden changes in the image due to lighting differences.

[0049] Step 123, based on step 122, further analyze the performance of neighboring boundaries in different dimensions. Neighboring boundaries may appear in the horizontal direction (left-right direction) or the vertical direction (up-down direction), which means that the direction of the brightness mutation can be horizontal or vertical; analyzing the brightness intensity changes of neighboring boundaries in different dimensions (vertical or horizontal) can help identify areas with more drastic brightness changes. In particular, when the brightness changes between adjacent backlight blocks are very drastic, it may cause non-smooth visual effects (such as halo effects on the screen); this step can detect those areas with too drastic brightness changes by quantifying the transition of brightness changes; by considering the brightness transition in different dimensions (vertical and horizontal), the illumination mutation areas in the image can be more comprehensively analyzed; if a drastic brightness mutation is detected, it can help determine whether there are unnatural visual effects (such as halo effects, edge blur, etc.), providing a basis for subsequent processing.

[0050] Step 124 performs smoothing on the region with drastic sudden changes in luminance transition detected in step 123. The goal of smoothing is to reduce the illumination difference at the sudden change, so that the illumination transition in the image is more natural and soft; thereby smoothing out the places where the luminance value transition is too drastic, and reducing the visual defects (such as halo effect) caused by hard boundaries; after smoothing, the illumination intensity of the image changes more smoothly and naturally, avoiding overly abrupt transitions; the final result is the "first control data set", which represents the luminance data after smoothing, which can be used for subsequent image optimization or display; through smoothing, the visual effect problems (such as hard boundaries, halo effects, etc.) caused by overly drastic sudden changes in luminance are effectively reduced; this step improves the visual quality of the image, making the illumination transition in the image smoother and natural, avoiding abrupt and uncomfortable visual performance; after processing the transition anomaly in the image, it can better adapt to the display device and provide a better user experience, especially in the display effect of high brightness or low light areas.

[0051] Steps 121 to 124 describe the process of detecting the brightness difference between adjacent backlight blocks and identifying the adjacent boundaries, and then processing the brightness mutation area to reduce the transition anomaly in the image. These steps detect and correct the brightness transition anomaly (such as the halo effect) to improve the image display effect; specifically, by analyzing the brightness transition of the adjacent boundaries, identifying and processing the area with drastic brightness changes, and finally improving the lighting transition in the image through smoothing, thereby improving the visual effect.

[0052] In a preferred embodiment of the present invention, the above step 124 further includes: Step 1241, set the adjacent data items I(x, y) and I(x', y') to represent the light intensity data on both sides of the adjacent boundary respectively; To obtain the light intensity difference value; where, Indicates the difference in luminosity intensity; Step 1242, setting the intensity threshold of the sudden change ,when , to determine that the difference in the luminous intensity values ​​will cause a drastic transition mutation; by To make the luminous intensity of the sudden change smooth; where, represents the luminous intensity value after smoothing, G(i,j,σ) represents the smoothing weight coefficient, σ controls the standard deviation of the smoothing degree, and I(x+i,y+j) represents the luminous intensity value in the neighborhood (i,j); Step 1243, after the smoothing process is completed, the drastic changes in the photometric intensity in the photometric labeling data set have been effectively suppressed, and the generated smoothed data set is the first regulated data set; by comparing the two adjacent photometric labeling data items before and after the process, whether a halo effect is generated, if not, execute the next step, if yes, return to the previous step and reprocess; to evaluate the smoothing effect.

[0053] In the embodiment of the present invention, in step 1241, adjacent data items I(x,y) and I(x',y') are set, which respectively represent the luminance intensity data on both sides of the neighboring boundary in the image. It can be understood that the pixel values ​​of two adjacent positions are selected as references for luminance changes; then the luminance changes of the image are quantified by calculating their luminance intensity difference values; by calculating the luminance intensity difference value, it can be determined whether there is a large luminance mutation in the image. This difference value serves as a standard for measuring the severity of luminance changes, helping us identify possible abnormal changes, such as image noise or edge mutations in the image; when the difference value is too large, it may indicate that the luminance mutation is too drastic, and subsequent smoothing processing is required.

[0054] In step 1242, a threshold is set to determine whether the sudden change in luminosity intensity is "too drastic". When the difference value is greater than the threshold, the sudden change is considered to be excessively drastic; Specifically, based on the sensitivity of the human eye to changes in luminosity, people are more sensitive to larger changes (such as changes of 30% or more) when perceiving changes in luminosity. Therefore, setting a threshold of 20%-30% can better avoid unnatural mutations.

[0055] In order to smooth the luminosity intensity, the contribution of each neighborhood point to luminosity smoothing can be controlled by the smoothing weight coefficient. σ represents the standard deviation, which determines the degree of smoothing. I(x+i,y+j) represents the luminosity intensity value in the neighborhood. By weighted averaging the luminosity values ​​in the neighborhood, the drastic changes in luminosity in the local area are smoothed, reducing the sudden changes in transitions. By setting the threshold and performing smoothing, the drastic luminosity changes in the image can be effectively suppressed, making the image look smoother and more natural. This is especially important for eliminating noise and reducing unnecessary halo effects. Appropriate smoothing can avoid overly abrupt luminosity changes, which is very beneficial for tasks such as image processing and edge detection, and can improve the effect of subsequent processing, such as image segmentation and object detection.

[0056] In step 1243, firstly, by comparing the two adjacent luminosity data items before and after the processing, it is determined whether the halo effect occurs. The halo effect usually refers to the abnormal halo in the transition area of ​​the image due to excessive smoothing or inappropriate smoothing, which affects the visual effect; when the halo effect is detected, it may indicate that the smoothing process is excessive or inappropriate, and the program will return to the previous step for adjustment to optimize the smoothing effect; when the halo effect is not detected, it is considered that the smoothing process is successful and the luminosity change of the image has been effectively controlled; by evaluating the halo effect, it is possible to avoid damage to the image details caused by excessive smoothing, and ensure that the smoothing process is only performed on the violent mutation area without affecting the smoother part of the image; this feedback mechanism makes the luminosity adjustment process more accurate, and can adaptively adjust the smoothing intensity to maintain the natural transition of image details and edges.

[0057] In summary, the various steps work together, and the core goal is to identify and eliminate drastic brightness changes or anomalies in the image, and smooth the brightness intensity to avoid unnatural transitions. The specific working principle can be summarized as follows: by calculating the brightness difference of adjacent pixels, determine whether there is a drastic brightness change; by setting a threshold and using a smoothing algorithm to alleviate excessive brightness changes; by evaluating the processing effect, avoid over-smoothing or halo effects, and ensure the visual naturalness and consistency of the image.

[0058] In a preferred embodiment of the present invention, in the above step 12, the neighboring boundaries between adjacent data items in the photometric labeling data set are adjusted according to the color difference between the photometric color scale data to obtain a second control data set, including: Step 125, there are three tones according to the color scale of the color space; based on the three tones, two adjacent data items in the photometric label data set are represented as (R, G, B) and (r, g, b) respectively; Step 126, according to the three tones, by To obtain the hue difference measurement value of two adjacent photometric labeled data items; preset the color difference threshold ,when Indicates that the color difference is too large; Where ΔE represents the measure of hue difference, R and r represent the brightness of two adjacent photometric labeled data items, G and g represent the green-red, B and b represent the blue-yellow components of the color respectively; Step 127, by analyzing the hue difference measurement value between two adjacent photometric mark data items, the light scheduling of the adjacent boundary position is finely adjusted to eliminate the sudden color difference transition between the two adjacent photometric mark data items.

[0059] In the embodiment of the present invention, in step 125, three tones are first defined to represent the colors in the image. These three tones are usually divided based on the RGB (red, green, and blue) color model, specifically: R and r: The "brightness" component of adjacent photometrically labeled data items (usually related to brightness, changes in brightness).

[0060] G and g: Green-red component, indicating the relative intensity of green and red.

[0061] B and b: Blue-yellow components, indicating the relative intensities of blue and yellow.

[0062] Two adjacent photometric label data items are represented as (R, G, B) and (r, g, b), respectively, where lowercase letters represent the corresponding hue values ​​of adjacent data items. Through this hue division, the color information of the image can be decomposed into brightness and two color components (green-red, blue-yellow), thereby achieving a more detailed color analysis. This provides more accurate color information, which is helpful for the subsequent detection and adjustment of color mutations.

[0063] In step 126, the hue difference between two adjacent photometrically marked data items is obtained, and ΔE is usually used to represent the color difference in the color space. ΔE is a standardized color difference measurement, which is often used to quantify the difference between two colors; ΔE is used to measure the color difference between (R, G, B) and (r, g, b); where R, G, B represent the brightness and color components of a photometrically marked data item, and r, g, b represent the corresponding values ​​of another adjacent data item; a preset color difference threshold is set, and when the color difference between two adjacent photometrically marked data items exceeds this threshold, it means that the hue difference is too large, which may cause abrupt color changes or unnatural transitions; Specifically, ΔE<1: Usually indistinguishable to the naked eye, with almost no difference.

[0064] ΔE=1-2: Usually considered a very subtle difference, noticeable to the naked eye, but the change is very small.

[0065] ΔE=2-5: This is a change that is easier to detect with the naked eye and is suitable for general visual applications.

[0066] ΔE=5-10: Obvious color difference, very easy to perceive with the naked eye.

[0067] ΔE>10: The color change is very significant, usually giving people an unnatural or abrupt feeling.

[0068] By measuring color difference, color mutations or unnatural transitions in images can be effectively detected, especially in areas with large color differences (such as the junction of light and shadow, color block dividing lines, etc.); the color difference threshold provides an adjustable standard that allows the system to flexibly decide when to make further tonal adjustments, thereby avoiding unnecessary transitions or color distortions.

[0069] In step 127, when the hue difference measurement value (ΔE) of two adjacent photometrically marked data items exceeds the preset color difference threshold, the system will reduce the color difference at the transition by finely adjusting the position of the neighboring boundary (i.e., the transition area between adjacent data items); by adjusting the neighboring boundary, the transition area can be smoothed, so that the color difference mutation between adjacent photometrically marked data items is alleviated. The purpose of this is to eliminate the transition of color difference mutation, so that the image is more natural in color transition; thereby eliminating the hue mutation caused by the excessive color difference between the photometrically marked data items, making the color transition of the image smoother and more natural, avoiding visual discomfort; fine light scheduling adjustment can improve the detail performance of the color transition area in the image, especially at the edge or in the area with complex color transition. By adjusting the position of the neighboring boundary, the color transition is softer, avoiding the sense of color discontinuity or unnatural light and shadow changes; by controlling the color difference between adjacent data items, the overall visual effect of the image can be optimized, especially in image processing (such as image fusion, image enhancement), to achieve better color balance and beauty.

[0070] The common goal of the above three steps is to fine-tune the color transition between photometric marker data items in the image by analyzing the hue differences in the color space to avoid overly drastic color difference mutations.

[0071] In a preferred embodiment of the present invention, the above step 127 further includes: Step 1271, when , α is close to 1, indicating that the transition of the hue difference measure is natural; , α is close to 0, indicating that the hue difference measurement transition mutation is drastic and needs to be adjusted by precision; where α represents the precision adjustment weight factor, Step 1272, based on the brightness intensity between two adjacent photometric mark data items, To adjust the accuracy of the hue difference measurement value between two adjacent photometrically labeled data items; where, Indicates the adjustment accuracy value of the hue difference measurement value. α is adjusted according to the size of the color difference ΔE to eliminate the transitional color difference mutation; Step 1273, after precision adjustment, the hue difference between adjacent data items in the photometrically labeled data set is naturally transitioned, and finally a second regulated data set is generated; To evaluate the accuracy adjustment effect; where w a represents the precision adjustment evaluation value, β represents the evaluation factor, which is used to evaluate the sensitivity of weight adjustment. The larger ΔE is, the greater the sensitivity of w a The smaller it is, the stronger the precision adjustment processing is in this area.

[0072] In the embodiment of the present invention, in step 1271, α is a precision adjustment weight factor, which is used to adjust the transition effect of the hue difference metric. When α is close to 1, it means that the hue transition process is very natural, that is, the difference between the hues is small and the transition is smooth. When α is close to 0, the transition of the hue difference metric becomes abrupt and violent, which usually means that the difference between the hues is large and the transition is abrupt.

[0073] By adjusting the value of α, you can achieve flexible control of the hue difference measurement, especially the smoothness of the transition. Specifically, the larger the α value, the more natural the transition; the smaller the α value, the more drastic the transition; by adjusting the value of α, the transition of the hue difference measurement can be made more natural or more refined as needed. For areas with large hue differences, you can adjust α to a smaller value to refine the color difference, make the transition more in line with visual perception, and thus improve the accuracy of hue transition.

[0074] In step 1272, the accuracy is adjusted based on the hue difference measurement value between two adjacent photometrically labeled data items in combination with the α value. The color difference ΔE is a standard for measuring the difference between two colors. Generally, the larger the ΔE value, the more obvious the hue difference. The purpose of the adjustment is to modify the accuracy of the hue difference measurement so that the hue difference between adjacent data items is more in line with the expected visual effect. For example, when the ΔE value is large, the α value will be adjusted to be smaller, thereby reducing the sudden change during the color difference transition and making the transition smoother. By adjusting the accuracy of the hue difference measurement, the color difference transition can be effectively controlled, making the originally abrupt color difference more natural and reducing visual discomfort. This helps to obtain a smoother and more coherent hue transition effect in applications such as data processing or image processing, thereby improving the user experience.

[0075] In step 1273, a second control data set is generated after precision adjustment. In this process, the hue difference between adjacent data items is effectively adjusted, thereby achieving a natural hue transition; at the same time, the adjustment strategy can be further optimized by evaluating the effect of precision adjustment. The evaluation process depends on the precision adjustment evaluation value (wa) and the evaluation factor (β). The precision adjustment evaluation value wa is inversely proportional to the color difference ΔE, that is, the larger the ΔE, the smaller the wa, indicating that the precision adjustment processing of the area requires stronger intervention; β is a factor used to evaluate the sensitivity of weight adjustment, which helps the system determine how the intensity of precision adjustment should be adjusted under different color difference conditions; through precision evaluation and sensitivity adjustment, the effect of the precision adjustment process is maximized. By making stronger adjustments to areas with larger ΔE, the visual effect of mutations is reduced, and the overall visual smoothness of the image or data set is improved; in addition, by generating a second control data set, the precision adjustment strategy can be continuously adjusted and optimized to further improve the quality of hue transition and avoid any unnaturalness or incoordination in the transition.

[0076] Through the flexible adjustment of the precision adjustment weight factor α, the transition of the hue difference measurement is changed from abrupt to natural and smooth, especially in the area with large color difference; the precision is automatically adjusted according to the size of the color difference ΔE to ensure that the precision adjustment process can achieve the best visual effect; through the precision adjustment evaluation and sensitivity factor β, the adjustment process is dynamically evaluated and optimized to ensure that the ideal hue transition effect can be obtained under various conditions.

[0077] On the whole, steps 1271 to 1273 make the hue transition between adjacent photometric mark data items more natural and smooth by precisely controlling the adjustment of hue differences, and effectively avoid the problems of sudden color difference or abrupt transition, thereby improving the naturalness and accuracy of the visual effect.

[0078] In a preferred embodiment of the present invention, the light intensity control model in step 13 includes: Step 131, integrating the first control data set and the second control data set into a data set, generating structure data from the data set acquired historically, encoding the structure data into sequence data, so as to train and obtain a photometric control model; Step 132, inputting the sequence data into the photometric control model; the photometric control model includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer, transmitting the intermediate representation data of multiple hidden layers to the output layer, and the output layer output represents the photometric control standard recognition result between two adjacent data items in the photometric labeling data set; Step 133, input two adjacent data items in the photometrically labeled data set into the photometric control model to output a newly acquired output layer output representing the photometric control prediction result between the next pair of adjacent photometrically labeled data items.

[0079] In an embodiment of the present invention, in step 131, two control data sets (a first control data set and a second control data set) are first merged to form a more complete data set. These data sets may contain data from different aspects of photometric control or at different time points. By integrating these data sets, more comprehensive photometric control information can be obtained; after the data sets are integrated, the data are converted into structured data. Structured data generally refers to a form of data with a clear format or relationship, which may include information such as photometric values, timestamps, and photometric change rates. This step is mainly to structure the data so that subsequent processing and modeling are more efficient and systematic.

[0080] Encoding as sequence data: Convert structured data into sequence data. Since the photometric control model may rely on sequence information (such as time series or dependencies), it is necessary to encode the data into sequence data. Sequence data can reflect the order and interdependence between data items and is suitable for training time series models (such as recurrent neural networks, etc.).

[0081] By integrating the two datasets, we can obtain richer information, which helps to improve the generalization ability of the model; structuring the data and encoding it into a sequence can be conveniently provided to the model for processing, especially for processing sequence or time series tasks, such as time dependency in photometric control.

[0082] In step 132, the encoded sequence data is input into a photometric control model. The model may be a multi-layer neural network, including multiple hidden layers (such as a first hidden layer, a second hidden layer, a third hidden layer) and an output layer.

[0083] Network structure: The input layer of the photometric control model receives sequence data. The data is then passed through multiple hidden layers. The role of each hidden layer is to extract features or patterns in the input data for deeper processing and understanding. Usually, the hidden layers of the neural network use nonlinear transformations to enable the model to learn complex patterns.

[0084] Output layer: Finally, the data processed by multiple hidden layers will be passed to the output layer. The output layer will generate the final prediction result - the standard recognition result of photometric control, that is, the photometric control standard between two adjacent data items.

[0085] Multi-level transmission: The intermediate representation data of the hidden layer is transmitted to the output layer through full connection or other means. The model extracts and transmits information at multiple levels, gradually optimizes and updates the weights, thereby improving the accuracy of the prediction.

[0086] Multiple hidden layers can extract high-order features of the data and improve the learning ability of the model; Time dependency modeling: Through the input of sequence data, the model can capture the temporal dependencies in luminosity changes and make more accurate predictions.

[0087] In step 133, the model receives two adjacent data items from the photometric labeled data set. Each pair of adjacent data items input represents the relationship between two photometric points or brightness points, and the model uses these data items as input to make predictions.

[0088] Output prediction results: After the input adjacent data items are processed by each layer (including hidden layers) of the photometric control model, the output layer will give the predicted value of the next photometric point. In other words, the model can not only learn the standard control relationship between photometric points, but also make predictions about future photometric values.

[0089] Photometric control prediction: The output of the model is the predicted photometric control standard, which reflects the adjustment standard or expected change between two adjacent photometric points.

[0090] By inputting adjacent data items for prediction, real-time photometric control prediction can be achieved. In practical applications, this is crucial for photometric control systems (such as display brightness, intelligent lighting control, etc.); by learning the relationship between adjacent data items through the model, more precise photometric control can be provided, avoiding sudden or overly drastic brightness changes, and improving user experience.

[0091] like Figure 2 As shown, a control system of a high contrast MiniLED is provided, wherein an electronic device is provided, including a memory, a processor, and a high contrast MiniLED control method program stored in the memory and executable on the processor, wherein the high contrast MiniLED control method program implements steps of a high contrast MiniLED control method when executed by the processor, and the system includes: Data acquisition module: used to collect the MiniLED screen display photometric data of multiple time frames to obtain a photometric data set; based on the photometric data set, the MiniLED backlight area is divided into multiple sub-areas to obtain a photometric data subset; based on the photometric data subset, the photometric data of the multiple sub-areas are subdivided into multiple small-sized backlight blocks one by one, and the multiple backlight blocks are marked to obtain a photometric marked data set; Data processing module: used to obtain the brightness intensity data of each data item in the photometric labeling data set to obtain a brightness intensity data set; smooth the neighboring boundaries between adjacent data items in the photometric labeling data set according to the difference between the brightness intensity data items to obtain a first control data set; obtain the photometric color scale data of each data item in the photometric labeling data set to obtain a photometric color scale data set; based on the photometric labeling data set after the smoothing process, adjust the accuracy of the neighboring boundaries between adjacent data items in the photometric labeling data set according to the color difference between the photometric color scale data to obtain a second control data set; A photometric control module: used to train the photometric labeling data set, the first control data set and the second control data set to construct a photometric control model, so as to output a photometric control standard recognition result between adjacent data items in the photometric labeling data set; input at least two other data items in the photometric labeling data set into the photometric control model, so as to output a photometric control prediction result between other adjacent photometric labeling data items; A control and evaluation module; it is used to coordinate the luminance transition between the data items in the luminance marking data set based on the prediction results of the luminance control model to reduce the halo effect of the MiniLED screen.

[0092] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0093] When the functions of the above modules are implemented in the form of software functional units and used as independent products, they can be stored in a storage device. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, ROM, RAM, disk or optical disk, etc., various media that can store program codes.

[0094] Therefore, the purpose of the present invention can also be achieved by running a program or a group of programs on any computing device. The computing device can be a well-known general intelligent device. Therefore, the purpose of the present invention can also be achieved by simply providing a program product containing a program code that implements the method or system. That is to say, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any known storage medium or any storage medium developed in the future. It should also be pointed out that in the device and method of the present invention, it is obvious that each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. In addition, the steps of performing the above-mentioned series of processing can naturally be performed in chronological order according to the order of description, but it is not necessary to perform them in chronological order. Some steps can be performed in parallel or independently of each other.

Claims

1. A high contrast MiniLED control method, characterized in that: include: Collect the MiniLED screen display photometric data in multiple time frames to obtain a photometric data set; Based on the photometric data set, dividing the MiniLED backlight area into a plurality of sub-areas to obtain a photometric data subset; Based on the photometric data subset, subdividing the photometric data of the plurality of sub-regions into a plurality of small-sized backlight blocks one by one, marking the plurality of backlight blocks to obtain a photometric marking data set; Acquire the light intensity data of each data item in the photometric label data set to obtain a light intensity data set; Smoothing the neighboring boundaries between adjacent data items in the photometric labeling data set according to the difference between the light intensity data items to obtain a first control data set; Acquire photometric color scale data of each data item in the photometric label data set to obtain a photometric color scale data set; based on the photometric label data set after smoothing, adjust the accuracy of the neighboring boundaries between adjacent data items in the photometric label data set according to the color difference between the photometric color scale data to obtain a second control data set; The photometric labeling data set, the first control data set and the second control data set are trained to construct a photometric control model to output a photometric control standard recognition result between adjacent data items in the photometric labeling data set; Inputting at least two other data items in the photometrically labeled data set into the photometric control model to output a photometric control prediction result representing the other adjacent photometrically labeled data items; Based on the prediction results of the photometric control model, the photometric transition between the data items in the photometric labeling data set is coordinated to reduce the halo effect of the MiniLED screen display.

2. The high contrast MiniLED control method according to claim 1, characterized in that: Based on the photometric data set, the MiniLED backlight area is divided into a plurality of sub-areas to obtain a photometric data subset, including: Preprocessing the photometric data set to obtain a preprocessed photometric data set; the photometric data set includes an average brightness value of each pixel in the backlight area; Divide the areas of the photometric data items in the preprocessed photometric data set into regions according to spatial distribution by setting photometric thresholds, thereby dividing the backlight area into a plurality of sub-areas; the sub-areas are further divided into bright areas and dark areas according to the high and low photometric values; The bright area and the dark area select K initial center points for division according to the luminance value distribution of each sub-area; each pixel point of the sub-area is assigned to the initial center point closest to obtain the division center point of each cluster; and the classification results of the bright area and the dark area are updated based on the division center point until the classification results converge; Based on the classification convergence result, the photometric data set is finally classified into the bright area and the dark area to obtain the photometric data subset.

3. The high contrast MiniLED control method according to claim 2, characterized in that: Based on the photometric data subset, the photometric data of the plurality of sub-regions are subdivided into a plurality of small-sized backlight blocks one by one, and the plurality of backlight blocks are marked to obtain a photometric marked data set, including: Dividing the photometric data subset into a plurality of square areas of equal size by a 1x1 sliding window with a step size of 1; Each of the divided square areas is a backlight block; the backlight block covers different photometric data parts of the photometric data subset; the backlight block represents a block area of ​​uniform size in the bright area or the dark area in the sub-area; Extracting the average illumination intensity of each backlight block to measure the degree of brightness change of the backlight block and capturing the texture features of the brightness change; marking each backlight block based on the extracted multiple features to obtain a brightness marking data set; The photometrically labeled dataset is assigned multiple labels according to the labeled target features; the multiple labels include highlight, low light, shadow or normal light.

4. The high contrast MiniLED control method according to claim 3, characterized in that: Smoothing the boundary lines between adjacent data items in the photometric labeling data set according to the difference between the light intensity data items to obtain a first control data set includes: Based on the difference in photometric intensity between adjacent data items in the photometric labeling data set, the transition anomaly at the photometric intersection between the adjacent data items is identified to obtain the neighboring boundary between the adjacent data items; The adjacent data items represent adjacent backlight blocks marked in the photometric marking data set; the adjacent boundary line represents the common boundary line between adjacent backlight blocks, that is, the backlight block cutting line in the photometric data subset; The neighboring boundary includes two dimensions, vertical or horizontal, to obtain the transition of the luminance intensity of multiple neighboring boundaries. When the luminance intensity transition between two adjacent luminance mark data items changes sharply and suddenly; the luminance intensity transition of the sudden change is not smooth, resulting in a halo effect on the screen display; The luminous intensity with drastic transition mutations is smoothed to obtain the first regulation data set.

5. The high contrast MiniLED control method according to claim 4, characterized in that: Smoothing the luminous intensity with a sudden change in transition to obtain the first control data set includes: Set the adjacent data items I(x, y) and I(x', y') to represent the light intensity data on both sides of the adjacent boundary respectively; To obtain the light intensity difference value; where, Indicates the difference in luminosity intensity; Set the intensity threshold for drastic changes ,when , to determine that the difference in the luminous intensity values ​​will cause a drastic transition mutation; by To make the luminous intensity of the sudden change smooth; where, represents the luminous intensity value after smoothing, G(i,j,σ) represents the smoothing weight coefficient, σ controls the standard deviation of the smoothing degree, and I(x+i,y+j) represents the luminous intensity value in the neighborhood (i,j); After the smoothing process is completed, the drastic changes in the photometric intensity in the photometric labeling data set have been effectively suppressed, and the generated smoothed data set is the first regulated data set; by comparing the two adjacent photometric labeling data items before and after the process, whether a halo effect is generated, if not, execute the next step, if so, return to the previous step and reprocess; to evaluate the smoothing effect.

6. The high contrast MiniLED control method according to claim 5, characterized in that: The step of adjusting the precision of the neighboring boundaries between the adjacent data items in the photometric labeling data set according to the color difference between the photometric color scale data to obtain a second control data set includes: There are three tones according to the color scale of the color space; based on the three tones, two adjacent data items in the photometric labeling data set are represented as (R, G, B) and (r, g, b) respectively; According to the three said hues, by To obtain the hue difference measurement value of two adjacent photometric labeled data items; preset the color difference threshold ,when Indicates that the color difference is too large; Where ΔE represents the measure of hue difference, R and r represent the brightness of two adjacent photometric labeled data items, G and g represent the green-red, B and b represent the blue-yellow components of the color respectively; By analyzing the hue difference measurement value between two adjacent photometrically marked data items, the light scheduling of the adjacent boundary position is finely adjusted to eliminate the sudden transition of the color difference between the two adjacent photometrically marked data items.

7. The high contrast MiniLED control method according to claim 6, characterized in that: By analyzing the hue difference measurement value between two adjacent photometric mark data items, the light scheduling of the adjacent boundary position is finely adjusted to eliminate the sudden transition of the color difference between the two adjacent photometric mark data items, including: when , α is close to 1, indicating that the transition of the hue difference measure is natural; , α is close to 0, indicating that the hue difference measurement transition mutation is drastic and needs to be adjusted by precision; where α represents the precision adjustment weight factor, Based on the brightness intensity between two adjacent photometric labeled data items, To adjust the accuracy of the hue difference measurement value between two adjacent photometrically labeled data items; where, Indicates the adjustment accuracy value of the hue difference measurement value. α is adjusted according to the size of the color difference ΔE to eliminate the transitional color difference mutation; After precision adjustment, the hue difference between adjacent data items in the photometrically labeled data set is naturally transitioned, and finally a second regulated data set is generated; To evaluate the accuracy adjustment effect; where w a represents the precision adjustment evaluation value, β represents the evaluation factor, which is used to evaluate the sensitivity of weight adjustment. The larger ΔE is, the greater the sensitivity of w a The smaller it is, the stronger the precision adjustment processing is in this area.

8. The high contrast MiniLED control method according to claim 7, characterized in that: The photometric control model includes: Integrate the first control data set and the second control data set into a data set, generate structure data from the historically acquired data set, and encode the structure data into sequence data to train a photometric control model; Inputting the sequence data into the photometric control model; the photometric control model comprises an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer, transmitting the intermediate representation data of multiple hidden layers to the output layer, and the output layer outputting the photometric control standard recognition result between two adjacent data items in the photometric labeling data set; Two adjacent data items in the photometrically labeled data set are input into the photometric control model to output a newly acquired output layer output representing a photometric control prediction result between a next pair of adjacent photometrically labeled data items.

9. A high contrast MiniLED control system, characterized in that: The system is provided with an electronic device including a memory, a processor, and a high-contrast MiniLED control method program stored in the memory and executable on the processor, wherein the high-contrast MiniLED control method program is executed by the processor to implement the steps of a high-contrast MiniLED control method as claimed in any one of claims 1 to 8, and the system includes: Data acquisition module: used to collect the MiniLED screen display photometric data of multiple time frames to obtain a photometric data set; based on the photometric data set, the MiniLED backlight area is divided into multiple sub-areas to obtain a photometric data subset; based on the photometric data subset, the photometric data of the multiple sub-areas are subdivided into multiple small-sized backlight blocks one by one, and the multiple backlight blocks are marked to obtain a photometric marked data set; Data processing module: used to obtain the brightness intensity data of each data item in the photometric labeling data set to obtain a brightness intensity data set; smooth the neighboring boundaries between adjacent data items in the photometric labeling data set according to the difference between the brightness intensity data items to obtain a first control data set; obtain the photometric color scale data of each data item in the photometric labeling data set to obtain a photometric color scale data set; based on the photometric labeling data set after the smoothing process, adjust the accuracy of the neighboring boundaries between adjacent data items in the photometric labeling data set according to the color difference between the photometric color scale data to obtain a second control data set; A photometric control module: used to train the photometric labeling data set, the first control data set and the second control data set to construct a photometric control model, so as to output a photometric control standard recognition result between adjacent data items in the photometric labeling data set; input at least two other data items in the photometric labeling data set into the photometric control model, so as to output a photometric control prediction result between other adjacent photometric labeling data items; A control and evaluation module; it is used to coordinate the luminance transition between the data items in the luminance marking data set based on the prediction results of the luminance control model to reduce the halo effect of the MiniLED screen.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which, when executed by a processor, implements the steps of the high-contrast MiniLED control method as described in any one of claims 1 to 8.

Images