LED backlight integration control system of liquid crystal display

Abstract

The application relates to the technical field of liquid crystal display, and discloses an LED backlight integrated control system of a liquid crystal display screen and an LED backlight integrated control method of the liquid crystal display screen. The LED backlight integrated control system of the liquid crystal display screen comprises a data processing unit, a prediction analysis unit, a weight calculation unit, a backlight optimization unit and a driving control unit. The LED backlight integrated control method of the liquid crystal display screen comprises the following steps: receiving a video frame and extracting data in a partition manner; predicting future visual characteristics in combination with historical and current data; calculating attention weights; then, integrating the weights, prediction and optimization target to determine a target backlight; and finally, generating a driving signal. Through time sequence prediction and space-time attention model and in combination with global optimization, the LED backlight intelligent control is realized. Compared with the prior art, the dynamic response lag is significantly improved, the picture details and focal points are improved, multiple display indexes are balanced, visual defects are reduced, and the picture quality performance and viewing comfort of the liquid crystal screen for various contents are enhanced.

Description

Technical Field

[0001] This invention relates to the field of liquid crystal display technology, specifically to an integrated LED backlight control system for a liquid crystal display screen. Background Technology

[0002] Liquid crystal displays (LCDs) do not emit light themselves and rely on backlights to display images. Traditional global backlighting methods result in uniform brightness, limiting contrast and dark-field performance, and also consume a lot of energy. To improve this situation, using light-emitting diodes (LEDs) as the backlight and combining them with local dimming technology has become mainstream. This technology divides the LED backlight into multiple independently controlled areas, dynamically adjusting the brightness of each area according to the content displayed on the screen, thereby improving contrast and reducing power consumption.

[0003] However, existing LED local dimming control still has shortcomings. Many control strategies adjust only based on the current frame information, which often leads to backlight response lag when dealing with rapid scene changes or high-speed moving objects in video streams, affecting the smoothness and naturalness of dynamic image quality.

[0004] Furthermore, achieving the optimal balance among multiple objectives—such as improving image contrast, accurately reproducing image details, avoiding visual imperfections like halos or block effects, ensuring a smooth transition of brightness over time, and effectively controlling energy consumption—remains a technical challenge. Existing methods may fail to adequately distinguish the visual importance of different areas when analyzing image content, or lack effective prediction of short-term changes in the image, thus requiring improvements in the precision and intelligence of backlight adjustment.

[0005] Therefore, this invention proposes an integrated LED backlight control system for liquid crystal displays to address the shortcomings of existing technologies. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides an integrated LED backlight control system for liquid crystal displays, which solves the problem in liquid crystal display LED backlight control technology that, due to the lack of in-depth analysis and prediction of the spatiotemporal dynamic characteristics of video content, it is difficult to achieve the optimal balance between improving image quality, ensuring visual comfort, and reducing energy consumption.

[0007] To achieve the above objectives, the present invention provides the following technical solution: an integrated LED backlight control system for a liquid crystal display screen, the system comprising the following modules:

[0008] The data processing unit is used to receive the current video frame via the video input interface and perform image partitioning processing on the current video frame to obtain image region data for each of the multiple image regions.

[0009] The predictive analysis unit is used to combine historical video frame information with the image region data of each of the multiple image regions, and apply a time-series prediction model to predict the predictive visual characteristics of each of the multiple image regions in the near future.

[0010] The weight calculation unit is used to combine the image region data of each of the multiple image regions with the predictive visual characteristics, and apply a multi-scale spatiotemporal attention model to calculate the attention weights corresponding to each of the multiple image regions.

[0011] The backlight optimization unit is used to execute an optimization algorithm based on the attention weights, the predictive visual characteristics, and a preset global optimization objective function to determine the target backlight brightness of multiple backlight zones corresponding to the multiple image regions.

[0012] A drive control unit is used to generate and output drive signals for controlling the actual brightness of the plurality of backlight zones based on the target backlight brightness.

[0013] Preferably, the data processing unit is specifically used for:

[0014] Perform color space conversion processing on the received current video frame;

[0015] According to the preset backlight physical partition layout, the current video frame, after color space conversion processing, is divided into the multiple image regions;

[0016] And extract the corresponding image data for each image region as the image region data.

[0017] Preferably, the predictive analysis unit is specifically used for:

[0018] For each image region, the brightness and motion features of that image region are extracted from the historical video frame information and the image region data of that image region;

[0019] Based on the extracted brightness features and motion features, the temporal state corresponding to the image region is updated through the temporal prediction model;

[0020] Based on the updated temporal state, the predictive brightness value and predictive brightness change trend of the image region in the near future are predicted as the predictive visual characteristics.

[0021] Preferably, the image region data for each image region, when the predictive analysis unit updates the temporal state, includes brightness-related features and motion-related features extracted from the image region; the predictive visual characteristics predicted by the temporal state include the predictive brightness value and predictive brightness change trend of the image region.

[0022] Preferably, the weight calculation unit is specifically used for:

[0023] For each of the image regions, multi-scale spatial features of the image region are extracted based on its image region data;

[0024] By combining the multi-scale spatial features with the predictive visual characteristics of the image region, the multi-scale spatiotemporal attention model is applied to comprehensively evaluate the visual importance of the image region in both spatial and temporal dimensions, and the attention weight corresponding to the image region is calculated.

[0025] Preferably, when applying the multi-scale spatiotemporal attention model, the weight calculation unit, for each image region:

[0026] Based on its image region data, the spatial saliency features and spatial detail features of the image region at multiple spatial scales are extracted;

[0027] The temporal attention information of the image region is calculated by combining its predictive visual characteristics.

[0028] The extracted spatial saliency features, spatial detail features, and calculated temporal attention information are then fused to generate the attention weight corresponding to the image region.

[0029] Preferably, the global optimization objective function used by the backlight optimization unit comprehensively evaluates the following aspects:

[0030] The accuracy of image content reproduction is measured based on the image region data of each image region and its corresponding attention weight;

[0031] The smoothness of brightness transition between adjacent backlight zones is measured based on the brightness relationship between adjacent image regions, their corresponding attention weights, and the image edge information between the image regions.

[0032] The continuity of backlight brightness over time is measured based on the relationship between the backlight brightness of the current frame and the previous frame, as well as the predictive brightness change trend contained in the predictive visual characteristics.

[0033] The energy consumption of the overall backlight system is measured based on the relationship between the target backlight brightness of each backlight zone and its preset power consumption model, the attention weight, and the overall visual characteristics of the current scene.

[0034] Preferably, in the global optimization objective function:

[0035] The attention weights are used to dynamically adjust the relative importance of each image region when evaluating the accuracy of the image content reproduction, and to dynamically adjust the constraint strength applied when evaluating the smoothness of the brightness transition between adjacent backlight zones in combination with the image edge information between the image regions.

[0036] The attention weights, combined with the overall visual characteristics of the current scene, are used to dynamically adjust the weighting ratio when evaluating the energy consumption of the overall backlight system.

[0037] The predictive brightness change trend included in the predictive visual characteristics is used to dynamically guide the desired change target when assessing the continuity of the backlight brightness over time.

[0038] Preferably, the drive control unit is specifically used to convert the target backlight brightness of each of the plurality of backlight zones into pulse width modulation parameters or analog drive current parameters for controlling the luminous intensity of the corresponding backlight zones, and output the converted pulse width modulation parameters or analog drive current parameters as the drive signal.

[0039] The present invention also provides an integrated LED backlight control method for a liquid crystal display screen, the method comprising the following steps:

[0040] S1. Receive the current video frame via the video input interface, and perform image partitioning processing on the current video frame to obtain image region data for each of the multiple image regions;

[0041] S2. Combining historical video frame information with the image region data of each of the multiple image regions, a time-series prediction model is applied to predict the predictive visual characteristics of each of the multiple image regions in the near future.

[0042] S3. Combining the image region data of each of the multiple image regions with the predictive visual characteristics, a multi-scale spatiotemporal attention model is applied to calculate the attention weights corresponding to each of the multiple image regions.

[0043] S4. Using the attention weights, the predictive visual characteristics, and the preset global optimization objective function, an optimization algorithm is executed to determine the target backlight brightness of multiple backlight zones corresponding to the multiple image regions.

[0044] S5: Generate and output a drive signal based on the target backlight brightness to control the actual brightness of the multiple backlight zones.

[0045] This invention provides an integrated LED backlight control system for a liquid crystal display screen. It has the following advantages:

[0046] 1. This invention employs a temporal prediction model to pre-determine future visual changes in image regions, enabling the LED backlight system to respond to image dynamics earlier. Compared to existing technologies that rely solely on current frame information for backlight adjustment, this invention effectively reduces the lag in backlight adjustment, especially during rapid scene transitions or high-speed object movement, resulting in a more natural and smooth brightness transition and enhancing the dynamic display quality of the LCD screen.

[0047] 2. This invention introduces a multi-scale spatiotemporal attention model to calculate the visual importance weights of each image region. It can precisely identify key content and details in the image and weight them based on their future dynamic trends. Compared to existing technologies that may employ global dimming or simple local dimming, treat all regions equally, or roughly classify importance, this invention solves the problems of failing to effectively highlight visual focus and potentially losing important details, resulting in a clear distinction between primary and secondary elements in the image and a better viewing experience.

[0048] 3. This invention determines the final backlight brightness by constructing a comprehensive global optimization objective function. This function cleverly balances image reproduction accuracy, backlight spatial transition smoothness, temporal series continuity, and overall energy consumption. Unlike existing technologies that often focus on a single objective at the expense of other performance aspects, this invention overcomes the shortcomings of balancing multiple display indicators and the tendency to compromise on one aspect for another, achieving a synergistic improvement in both display performance and energy efficiency of the liquid crystal display.

[0049] 4. In backlight optimization, this invention pays special attention to the smooth transition of brightness between adjacent backlight zones, and dynamically adjusts the constraint strength by combining attention weights and image edge information. It also guides the continuous change of backlight over time based on predicted brightness change trends. Compared to some existing backlight control strategies that easily produce halos at the boundary between light and dark areas or exhibit brightness flickering in dynamic scenes, this invention significantly reduces visual defects and improves the viewing comfort of the LCD screen.

[0050] 5. The integrated LED backlight control system of this invention, through a complete process from data processing, predictive analysis, weight calculation, backlight optimization to drive control, can intelligently adjust the brightness of each LED backlight zone according to the real-time characteristics and future trends of the video content. Compared with existing technologies that may use fixed control logic or algorithms with poor adaptability, this invention solves the limitation of its inability to continuously provide optimal display effects when facing diverse video content, enabling the LCD screen to exhibit excellent adaptability and image quality performance for different types of images. Attached Figure Description

[0051] Figure 1 This is a system architecture diagram of the present invention;

[0052] Figure 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0053] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] Please see Figure 1 This invention provides an integrated LED backlight control system for a liquid crystal display screen, the system comprising the following modules:

[0055] The data processing unit is used to receive the current video frame via the video input interface and perform image partitioning processing on the current video frame to obtain image region data for each of the multiple image regions.

[0056] In this embodiment, the data processing unit is responsible for performing initial processing on the input video signal and extracting basic analysis data.

[0057] The data processing unit receives the current video frame via a video input interface. The video input interface can be a High-Definition Multimedia Interface (HDMI), a DisplayPort, or a system internal bus, used to receive digital video signals from external video playback devices, graphics processing units, or other image sources. The current video frame is an independent frame that constitutes a dynamic video sequence, carrying image information at that moment.

[0058] Upon receiving the current video frame, the data processing unit performs a color space conversion process on it. This process aims to convert the video frame data from its original color space to a color space more suitable for subsequent brightness analysis and backlight control. Specifically, a commonly used conversion method is to convert the input RGB (Red, Green, Blue) color space video frame data to the YCbCr color space. In this YCbCr color space, the Y component represents the brightness information of the image, while the Cb and Cr components represent the chromaticity information. Since the adjustment of the LED backlight system directly affects the perceived brightness of the displayed image, separating the brightness information (Y component) allows for more focused and efficient subsequent analysis and optimization.

[0059] For example, if the red, green, and blue component values ​​of the input pixel are R, G, and B respectively, its corresponding luminance component Y can be calculated using the following linear transformation formula:

[0060] Y = w R ·R+w G ·G+w B ·B+O offset ;

[0061] Where R, G, and B represent the normalized or non-normalized pixel values ​​of the red, green, and blue channels, respectively; w R ,w G ,w B These are preset weighting coefficients, which vary depending on the color standard used (e.g., ITU-R-BT.601, ITU-R-BT.709, or ITU-R-BT.2020). For example, under the ITU-R-BT.709 standard, for R, G, B values ​​normalized to the [0, 1] interval, the commonly used coefficient is w. R =0.2126,w G =0.7152,w B =0.0722; O offset This is an optional offset used in some standards (such as when processing digital video components) to adjust the black level; for example, for 8-bit video data, the offset might be 16. After color space conversion, the resulting luminance component map will serve as the primary basis for subsequent image partitioning and feature extraction.

[0062] The data processing unit divides the current video frame (especially its luminance component map) into multiple image regions based on a preset backlight physical partition layout, after color space conversion processing. The purpose of this image partitioning process is to spatially divide the continuous image frame into several regions corresponding to independently controllable LED physical partitions in the LCD backlight module. The backlight physical partition layout information describes the number, geometry, size, and spatial arrangement of backlight LED units or groups on the screen. This information is typically predefined by the display device's hardware specifications and stored in the system. For example, the backlight system might be divided into an M×N regular rectangular grid array, where M and N represent the number of backlight partitions along the vertical and horizontal directions of the screen, respectively.

[0063] Preferably, irregular partitioning may also be used to better adapt to specific screen shapes or to differentiate the importance of different display areas. The partitioning operation ensures that each generated image area is spatially closely associated with one or a set of specific physical backlight partitions.

[0064] After image partitioning, the data processing unit extracts corresponding image data for each partitioned image region. These data sets constitute the image region data for each of the multiple image regions. Extracting the image region data is to quantify the image content characteristics within each local region, thereby providing necessary input information for subsequent predictive analysis, attention calculation, and backlight optimization. Specifically, for the k-th image region A... k The extracted image region data may include, but is not limited to, one or more of the following features:

[0065] Average brightness value Calculate the image region A k The arithmetic mean of the brightness values ​​of all pixels within region A. k Contains N k There are pixels, and the brightness value of each pixel p is L(p). Then:

[0066] Brightness histogram H k (l): Statistical analysis of image region A k The number of pixels or frequency of occurrence at each brightness level l.

[0067] Luminance standard deviation σ L,k : Measure the image region A k The dispersion of internal brightness values ​​reflects the contrast or complexity of the content within a region:

[0068] Other statistical characteristics, such as maximum brightness value, minimum brightness value, and median brightness value.

[0069] Texture features: For example, parameters such as energy, contrast, correlation, and entropy, which can be calculated based on the gray-level co-occurrence matrix (GLCM), can be used to describe the texture roughness or uniformity of a region.

[0070] Edge information: For example, by applying edge detection operators (such as Sobel, Prewitt, Canny, etc.), the number of edge pixels, average edge intensity, or edge orientation histogram within the statistical region can be obtained.

[0071] The extracted image region data comprehensively characterizes the static content characteristics of each local image block, serving as the basis for subsequent refined analysis and decision-making in the entire backlight integrated control system.

[0072] The predictive analysis unit is used to combine historical video frame information with the image region data of each of the multiple image regions, and apply a time-series prediction model to predict the predictive visual characteristics of each of the multiple image regions in the near future.

[0073] In this embodiment, the predictive analysis unit aims to endow the system with adjustment capabilities through a deep insight into the temporal dynamics of video content.

[0074] The core task of the predictive analysis unit is to combine historical video frame information with the image region data of each of the multiple image regions currently acquired from the data processing unit, and apply a temporal prediction model to predict the visual characteristics of each of the multiple image regions in the near future. These predicted visual characteristics are called predictive visual characteristics, and they provide crucial temporal dimension information for subsequent attention weight calculation and backlight optimization.

[0075] Specifically, for each individual image region, the predictive analysis unit performs the following operations:

[0076] The predictive analysis unit extracts the brightness and motion features of the image region from the historical video frame information and the image region data of the current image region. The historical video frame information may include several past frames (e.g., the previous N frames). h Frames, where N h The image region data (which is a preset integer) corresponding to the image region includes information such as its historical average brightness and historical brightness variations. As mentioned earlier, the image region data for the current image region already contains rich brightness-related features (such as the average brightness of the current region). At time t, the brightness histogram H k (t) etc.).

[0077] Motion feature extraction is crucial for understanding scene dynamics. In one possible implementation, motion features can be estimated by comparing the content of corresponding image regions in the current frame t with that in one or more immediately adjacent historical frames (e.g., t-1, t-2, ...). For example, image region A can be calculated. k A measure of the inter-frame difference (FD) between the current frame t and the previous frame t-1, such as the mean absolute difference:

[0078]

[0079] Where L(p,t) represents the pixel p in the k-th image region A k The brightness value at time t, N k It is region A k The number of pixels in FD. k The magnitude of (t) can characterize the intensity of motion in that region.

[0080] Alternatively, the Block-Matching Algorithm (BMA) can be used to match image region A. kEstimate its motion vector MV k (t)=(mv x ,mv y ) k (t). This motion vector describes region A. k The main displacement direction and magnitude between frames. Motion features can include the magnitude ||MV of the motion vector. k (t)|| and direction θ k (t). These brightness and motion features together form the basis for analyzing the temporal behavior of this image region.

[0081] Based on the extracted brightness features (e.g., time series) ) and the motion features (e.g., time series {FD k (t),FD k (t-1),...} or {MV k (t),MV k (t-1),...}), the prediction analysis unit updates the temporal state S corresponding to the image region through the temporal prediction model. k (t). The time series prediction model is a model that can learn and characterize the inherent patterns and dynamics of time series data.

[0082] In some embodiments, the time-series prediction model can be a classical statistical model, such as a Kalman filter. A Kalman filter recursively estimates the system state based on a noisy observation sequence (i.e., extracted brightness and motion features) and performs a one-step prediction. Here, the time-series state S... k (t) can be the state vector in the Kalman filter, which contains estimates of latent variables such as the current brightness of the region and the rate of change of brightness.

[0083] In other embodiments, more complex machine learning models can be employed, such as Auto-regressive Integrated Moving Average (ARIMA) models, or models based on Recurrent Neural Networks (RNNs) and their variants such as Long Short-Term Memory (LSTM) networks or Gated Recurrent Units (GRUs). Deep learning models are particularly adept at capturing nonlinear and long-term dependencies. For RNN / LSTM / GRU models, the time-series state S k (t) typically corresponds to the state vector of the hidden layer of the network after processing the input at time t.

[0084] According to the updated timing state S k (t), the prediction and analysis unit further predicts that the image region will be in the near future (e.g., from frame t+1 to frame t+Δt). pred Frame, where Δt pred It is a preset prediction time window (usually a small number of frames) containing predictive brightness values ​​and predictive brightness variation trends. These together constitute the predictive visual characteristics for that image region.

[0085] The predicted brightness value (where 1≤τ≤Δt) pred ) is a direct numerical estimate of the average brightness of the image region at a future time t+τ.

[0086] The predicted brightness change trend can be a qualitative description of the future direction of brightness change (e.g., significant increase, slight increase, stable, slight decrease, significant decrease), or a quantitative estimate of the rate or acceleration of brightness change (e.g., the predicted first derivative of brightness). or second derivative ).

[0087] For example, if the time-series prediction model is an ARIMA model, the model's prediction function can be used directly to output the brightness values ​​for multiple future steps. If it is an RNN / LSTM model, the current hidden state S can be used to predict the brightness values. k (t) and possible future inputs (or zero inputs, or autoregressive inputs based on previous outputs) are fed into the network to generate a predicted sequence of future brightness.

[0088] By predicting the future brightness and dynamic trends of image areas, the system can adjust the backlight earlier and more smoothly, avoiding backlight response lag or overshoot caused by sudden scene changes, thereby improving visual comfort and the display quality of dynamic images. For example, if it is predicted that the brightness of a certain area will increase significantly, the backlight system can start increasing the brightness of the corresponding area in advance.

[0089] It should be clarified that the image region data for each image region, when used by the predictive analysis unit to update the temporal state, does indeed include the brightness-related features and motion-related features extracted from that image region. Furthermore, the predictive visual characteristics predicted through the temporal state do indeed include the predictive brightness value and predictive brightness change trend of that image region.

[0090] The weight calculation unit is used to combine the image region data of each of the multiple image regions with the predictive visual characteristics, and apply a multi-scale spatiotemporal attention model to calculate the attention weights corresponding to each of the multiple image regions.

[0091] In this embodiment, the weight calculation unit is responsible for evaluating the visual importance of each image region in the video frame, so as to provide differentiated guidance for subsequent backlight optimization:

[0092] The core function of the weight calculation unit is to combine the image region data of the multiple image regions provided by the data processing unit and the predictive visual characteristics provided by the predictive analysis unit, and apply a preset multi-scale spatiotemporal attention model to calculate the attention weights corresponding to each of the multiple image regions. These attention weights aim to quantify the importance of each image region to human visual perception.

[0093] Specifically, for each image region A k At the current time t, the weight calculation unit performs the following processing:

[0094] The weight calculation unit first calculates the weight based on the image region A. k Image region data This study extracts the spatial saliency and spatial detail features of the image region at multiple spatial scales. The purpose of multi-scale analysis is to capture the structural information and content characteristics of the image at different granularities, as the human eye has different sensitivities to features at different scales.

[0095] Spatial saliency features At scale s, the aim is to quantize image region A. k The visual saliency of a region is relative to its neighborhood or the entire scene. Generally, the human visual system prioritizes regions that differ significantly from their surroundings in color, brightness, orientation, or texture. In one possible implementation, spatial saliency features can be calculated using known visual saliency detection algorithms, such as those based on local contrast analysis, frequency domain analysis, or graph theory. For example, for each image region A... k A saliency value is calculated at a specific scale s, which reflects the likelihood that the area will become a visual focal point.

[0096] Spatial details At scale s, it is used to characterize image region A. k The richness of internal detail and structural complexity. These features help identify areas containing fine textures, sharp edges, or other high-frequency information, which typically require higher display fidelity. Specifically, spatial detail features can include:

[0097] Edge intensity or density: By analyzing the image region A at scale s k Apply edge detection operators (such as Sobel, Canny, etc.) to count the number of edge pixels or the average value of edge amplitude.

[0098] Texture descriptor: For example, calculating region A at scale s. k The gray-level co-occurrence matrix (GLCM) is obtained, and texture parameters such as energy, contrast, and entropy are extracted from it.

[0099] Local variance or standard deviation: Calculation area A k The variance or standard deviation of luminance values ​​at scale s; a higher variance generally indicates richer detail. To achieve multi-scale extraction, an image pyramid (e.g., a Gaussian pyramid or a Laplacian pyramid) can be constructed from the current video frame (or its luminance components). Subsequently, for the original image region A... k In the region corresponding to each level of the pyramid (i.e., each scale s), the aforementioned spatial saliency features are extracted respectively. and spatial details

[0100] In parallel, the weight calculation unit combines the image region A k Predictive visual characteristics (Provided by the predictive analytics unit, such as predicted future brightness values) and predicted brightness change trend Calculate its temporal attention information I T,k (t). Temporal attention information aims to capture the importance of regions that are expected to undergo significant changes or exhibit specific dynamic behaviors over time. Generally, fast-moving objects, areas of scene about to experience drastic brightness changes, or areas with high predictive uncertainty are more likely to attract the user's attention.

[0101] In one possible implementation, the temporal attention information I T,k (t) can be computed as a function of predictive visual properties. For example:

[0102]

[0103] in, This represents the absolute value of the predicted rate of change in brightness. This represents the absolute value of the predicted acceleration of the change in brightness, uncertainty. k (t) represents a measure of uncertainty (e.g., prediction variance) in the prediction results for this region. The function f temporal It could be a weighted sum or a more complex mapping function, such that when the prediction changes more drastically or the uncertainty is higher, the temporal attention information I... t,k The value of (t) is larger.

[0104] Spatial saliency features at multiple spatial scales were obtained. Spatial details and temporal attention information I t,k After (t), the weight calculation unit fuses this information to generate the image region A. k The corresponding final attention weight W k (t). This fusion process is defined by the multi-scale spatiotemporal attention model.

[0105] The fusion process may include the following steps:

[0106] First, multi-scale spatial features can be integrated to obtain a comprehensive spatial attention score. For example, this can be achieved by weighting and averaging spatial saliency features and spatial detail features at different scales, or by combining them using a small neural network.

[0107]

[0108] Among them, g spatial It is a predefined spatial feature fusion function.

[0109] Subsequently, the obtained comprehensive spatial attention score Compared with the previously calculated temporal attention information I t,k (t) are fused to produce the final attention weight W. k (t). This fusion aims to balance the spatial importance of image content at the current moment with the temporal importance of anticipated future changes. In one possible implementation, the final attention weight W k (t) can be obtained through weighted combination:

[0110]

[0111] Here, α and β are preset weighting coefficients used to adjust the relative contributions of spatial and temporal attention factors.

[0112] In another possible implementation, the fusion process can be achieved through a specially designed neural network model that takes the extracted spatial features and temporal attention information as input and directly outputs attention weights.

[0113] The final attention weight W k (t) is typically normalized, for example, by summing it to 1 across all image regions, or by constraining each weight value to the interval [0, 1]. This normalized attention weight W k (t) will be passed to the subsequent backlight optimization unit to guide the allocation of backlight brightness. Areas with high attention weights indicate that they are more visually important and their display quality should be prioritized.

[0114] The backlight optimization unit is used to execute an optimization algorithm based on the attention weights, the predictive visual characteristics, and a preset global optimization objective function to determine the target backlight brightness of multiple backlight zones corresponding to the multiple image regions.

[0115] In this embodiment, the backlight optimization unit is the core module that performs backlight brightness decisions. Its purpose is to determine the optimal luminous intensity of each backlight zone under the premise of comprehensively considering various visual quality and system performance indicators.

[0116] The backlight optimization unit receives multiple image regions A provided by the weight calculation unit for the current video frame t. k The corresponding attention weights W k (t), and predictive visual characteristics for these image regions provided by the predictive analysis unit. Based on this input information, and combined with a pre-defined global optimization objective function The backlight optimization unit executes an optimization algorithm to calculate and determine the plurality of image regions A. k The target backlight brightness of each of the corresponding backlight zones

[0117] The global optimization objective function The design aims to provide a comprehensive and balanced evaluation of display performance and system power consumption. This objective function is typically designed as a scalar function composed of a weighted combination of multiple sub-objectives. The optimization algorithm aims to find a set of target backlight brightness values ​​{B}. k (t)} k This aims to optimize (e.g., minimize) the function value. Specifically, the global optimization objective function comprehensively evaluates the following aspects:

[0118] Accuracy of image content reproduction (fidelity item) ):

[0119] This is intended to ensure that the adjusted backlight accurately reproduces the visual brightness that the original image content should present. The measurement is based on each image region A. k Image region data (For example, from) The raw average brightness extracted from (or target perceived brightness) and its corresponding attention weight W k (t).

[0120] The attention weight W k (t) Here, the relative importance of each image region is dynamically adjusted. Regions with high attention weights will be penalized more severely for deviations in brightness reproduction. In one possible implementation, the fidelity term can be expressed as:

[0121]

[0122] Among them, B k (t) is the image region A to be optimized. k The brightness of the corresponding backlight zones; At a backlight brightness of B k At time (t), image region A k The actual perceived brightness displayed on the screen after modulation by the LCD panel, this perceived brightness is related to B. k (t) and data from image regions The average transmittance T of the liquid crystal pixel obtained in k (t) and the display's gamma characteristic γ lcd Related, for example, L′ k ≈T k (t)·(B k (t)) γlod L target,k (t) is the region A k The target display brightness is typically derived from the raw image data. d(·,·) is a non-negative error metric function, such as the squared error (xy). 2 .

[0123] Smoothness of brightness transition between adjacent backlight zones (spatial smoothness item) ):

[0124] This design aims to avoid abrupt brightness jumps between different backlight zones, thereby reducing potential halo effects or blocking artifacts and improving the overall uniformity and naturalness of the image. The metric is based on adjacent image regions A. k With A j The brightness relationship between them, and their corresponding attention weights W k (t) and W j (t), and the image edge information E between the image regions. k,j (t).

[0125] The attention weight W k (t) and W j (t) Combining the image edge information E between the image regions k,j (t) Dynamically adjust the constraint strength applied when evaluating the smoothness of the brightness transition between adjacent backlight zones. For example, if there are strong natural image edges (E) between adjacent areas. k,j If (t) is relatively large, then a certain difference in backlight brightness is allowed; conversely, if the content between areas is smooth or is a high-attention area, then stricter brightness consistency is required.

[0126] In one possible implementation, the spatial smoothness term can be expressed as:

[0127]

[0128] in, Representative and Region A k The set of indices of adjacent image regions; ω s,kj It is a dynamic weighting function whose value depends on W. k (t), W j (t) and E k,j (t) is determined. For example, ω s,kj It can be set to (W) k (t)+W j (t) is directly proportional to E k,j The inverse relationship between (t) indicates that stronger smoothing constraints are needed between regions with high attention or low edge intensity.

[0129] Continuity of backlight brightness over time (time smoothness term) ):

[0130] This measure aims to ensure a smooth transition of backlight brightness between video frame sequences, avoiding eye discomfort or screen flickering caused by abrupt brightness fluctuations. It is based on the target backlight brightness B of each backlight zone in the current frame. k (t) Actual backlight brightness B of the corresponding partition of the previous frame k The relationship between (t-1) and the predictive visual characteristics It includes predictive brightness change trends.

[0131] The predictive brightness change trend included in the predictive visual characteristics is used to dynamically guide the desired change target when assessing the continuity of the backlight brightness over time. That is, it not only penalizes rapid changes but also encourages the backlight to evolve smoothly towards the predicted trend.

[0132] In one possible implementation, the time smoothness term can be expressed as:

[0133]

[0134] Among them, B k (t-1) is the image region A k The brightness value of the corresponding backlight zone in the previous frame; ΔB target,k (t) is based on predictive visual characteristics The desired ideal brightness adjustment for the backlight zone in the current frame is calculated using the predicted rate or direction of brightness change (e.g., the predicted brightness change rate or direction). If the predicted area will brighten, ΔB... target,k(t) is positive, and vice versa.

[0135] Energy consumption of the overall backlight system (energy consumption item) ):

[0136] This initiative aims to reduce the overall power consumption of the backlight system, achieving energy efficiency. The metric is based on the target backlight brightness (B) of each backlight zone. k (t) and its preset power consumption model P(B) k (t) (This model describes the relationship between backlight brightness and its actual power consumption, which is usually a non-linear increasing function), and comprehensively considers the attention weight W. k (t) and the overall visual characteristics of the current scene Ω scene (t).

[0137] The attention weight W k (t) and combined with the overall visual characteristics of the current scene Ω scene (t) (e.g., scenes with low average brightness and simple content, indicating high energy-saving potential), used to dynamically adjust the weighting when evaluating the overall energy consumption of the backlight system. For areas with lower visual importance or scenes with high overall energy-saving potential, the backlight brightness can be more aggressively reduced to save energy.

[0138] In one possible implementation, the energy consumption term can be represented as:

[0139]

[0140] Among them, P(B) k (t) represents the backlight brightness B k (t) corresponds to the power consumption; ω e,k It is a dynamic weighting function whose value depends on the attention weights W. k (t)(For example, with 1-W k (t) is directly proportional to (indicating that low-attention areas prioritize energy saving) and the overall visual characteristics of the scene Ω scene (t)(For example, when Ω scene (t) indicates when the scene is dark or simple, ω c,k Adjustments will be made accordingly to strengthen energy-saving constraints.

[0141] Finally, the global optimization objective function It can be the weighted sum of the above items:

[0142]

[0143] Where, λ fid ,λ spat ,λ temp ,λen These are preset non-negative weighting coefficients used to balance the importance of different optimization objectives. These coefficients can be adjusted according to the application scenario or user preferences.

[0144] The backlight optimization unit then employs a suitable optimization algorithm to solve the aforementioned global optimization objective function, in order to find a set of... Optimal target backlight brightness value Possible optimization algorithms include, but are not limited to, gradient-based optimization methods (such as gradient descent and conjugate gradient methods), quadratic programming (QP) solvers (if the objective function and constraints can be transformed into quadratic forms), or various heuristic optimization algorithms (such as simulated annealing and genetic algorithms). During the optimization process, physical constraints must typically be met, namely, the target brightness B for each backlight zone. k (t) must be within its adjustable minimum brightness B min and maximum brightness B max between.

[0145] A drive control unit is used to generate and output a drive signal for controlling the actual brightness of the plurality of backlight zones based on the target backlight brightness.

[0146] In this embodiment, the drive control unit, as the final execution link of the integrated control system for LED backlight of the liquid crystal display, is mainly responsible for converting the abstract target backlight brightness value determined by the upstream backlight optimization unit into specific physical control commands that can be recognized and executed by the underlying hardware.

[0147] The drive control unit receives the target backlight brightness for each of the plurality of backlight zones k for the current video frame t, calculated by the backlight optimization unit. These target backlight brightness values ​​are typically normalized logical values ​​(e.g., in the range [0, 1], where 0 represents backlight off and 1 represents maximum brightness) or represent the desired luminous flux or luminance units.

[0148] The core task is to drive the control unit to adjust the target backlight brightness of each of the multiple backlight zones. This is converted into pulse-width modulation (PWM) parameters or analog drive current parameters to control the luminous intensity of the corresponding backlight zones (typically consisting of one or a group of LEDs). The choice between PWM control and analog current control typically depends on the specific design and capabilities of the backlight driver hardware.

[0149] In one possible implementation, if pulse width modulation (PWM) control is used, the drive control unit will adjust the target backlight brightness. Converted to the corresponding PWM signal duty cycle D k (t). PWM control adjusts the average luminous intensity of the LED by changing the pulse width (i.e., the proportion of the "on" time to the entire cycle) of the driving LED within a fixed high-frequency cycle. Generally, higher target backlight brightness corresponds to a larger duty cycle. This conversion process can be achieved by a predefined mapping function f. PWM To achieve:

[0150]

[0151] Among them, D k The value range of (t) is typically from 0% to 100% (or from 0 to 1). The function f PWM The form may not be a simple linear relationship because it needs to take into account the photoelectric conversion characteristics of the LED itself (i.e., the relationship between light output and PWM duty cycle may not be linear) and possible visual perception corrections (e.g., gamma correction to make the perceived brightness linearly related to the target brightness value). In some embodiments, f PWM This can be achieved using a look-up table (LUT), which pre-stores the optimal PWM duty cycle corresponding to different target brightness values. These correspondences can be obtained by performing detailed calibration and characteristic measurements on a specific LED backlight module.

[0152] As an alternative, if analog drive current control is used, the drive control unit will adjust the target backlight brightness. Convert to the corresponding analog drive current value I k (t). The luminous intensity of an LED can be controlled by directly adjusting the current flowing through it; generally, the higher the current, the higher the brightness. This conversion process can also be achieved using a predefined mapping function f. current To achieve:

[0153]

[0154] Among them, I k The value of (t) must be within the safe operating current range of the LED. The function f current A precise luminous flux-current (LI) characteristic curve of the LED is needed, which describes the non-linear relationship between the LED's luminous intensity and the driving current. Similarly, to achieve precise brightness control, f... current The specific form or parameters can be determined by calibration testing of the LED module.

[0155] Regardless of the conversion method used, the core objective is to ensure that the converted physical control parameters (PWM duty cycle or analog current) enable the corresponding backlight zones to produce the target backlight brightness. Consistent actual light output.

[0156] After completing the conversion from target backlight brightness to specific control parameters, the drive control unit will convert the pulse width modulation parameter D into a pulse width modulation parameter D. k (t) or the analog drive current parameter I k (t) is output as the driving signal. These driving signals are transmitted to the individual LED driver chips or circuits in the display backlight module. Based on the received driving signals, these driver chips or circuits precisely adjust the voltage or current applied to the corresponding LED backlight zones, thereby achieving independent and precise control over the actual luminous brightness of each backlight zone.

[0157] Please see Figure 2 The present invention also provides an integrated LED backlight control method for a liquid crystal display screen, the method comprising the following steps:

[0158] S1. Receive the current video frame via the video input interface, and perform image partitioning processing on the current video frame to obtain image region data for each of the multiple image regions;

[0159] In this step, the system first receives the current video frame in real time through a standard video input interface, such as HDMI or DisplayPort. The received video frame data then undergoes a color space conversion, for example, from the common RGB color space to a color space more conducive to brightness information analysis, thereby separating the brightness component for subsequent processing. Next, based on the physical partition layout information preset by the LCD backlight hardware, the color-space-converted video frame (especially its brightness information) is spatially divided into multiple independent image regions, so that each image region corresponds to one or a group of independently controllable LED backlight partitions. After partitioning, the system extracts key image features for each partitioned image region, such as the average brightness, brightness distribution, texture complexity, or edge strength of the region. These features together constitute the image region data for that region, providing a basis for subsequent analysis and decision-making.

[0160] S2. Combining historical video frame information with the image region data of each of the multiple image regions, a time-series prediction model is applied to predict the predictive visual characteristics of each of the multiple image regions in the near future.

[0161] In this step, the system focuses on the temporal dynamics of the video content. For each image region, the system combines information from several historical video frames (e.g., past brightness change trajectories) with the image region data obtained from step S1 in the current frame to extract brightness and motion features that reflect changes in the content of that region. Based on these extracted temporal features, the system applies a built-in temporal prediction model (e.g., a Kalman filter, an ARIMA model, or a recurrent neural network-based model) to update and maintain an internal temporal state for the image region, which summarizes the current dynamic characteristics of the region. Then, based on this updated temporal state, the temporal prediction model infers and predicts the possible visual characteristics of the image region within a short future time window (e.g., the next few frames), mainly including its predicted brightness value and predicted brightness change trend.

[0162] S3. Combining the image region data of each of the multiple image regions with the predictive visual characteristics, a multi-scale spatiotemporal attention model is applied to calculate the attention weights corresponding to each of the multiple image regions.

[0163] In this step, the system aims to evaluate the relative importance of different image regions in the current video frame to human visual perception. For each image region, the system first extracts its spatial features at multiple different spatial scales or resolution levels based on the image region data obtained in step S1. These spatial features can include spatial saliency features reflecting the region's prominence, as well as spatial detail features describing the region's internal texture details and edge sharpness. Simultaneously, the system combines the predictive visual characteristics of the image region obtained in step S2 (especially predicted brightness changes) to calculate its attention information in the temporal dimension, i.e., the degree to which future dynamic changes in the region might attract human attention. Finally, the system effectively fuses the extracted multi-scale spatial features with the calculated temporal attention information using a multi-scale spatiotemporal attention model, comprehensively evaluating the spatial importance of the image region at the current moment and the temporal importance of expected future changes, and accordingly calculating a quantified attention weight value for each image region.

[0164] S4. Using the attention weights, the predictive visual characteristics, and the preset global optimization objective function, an optimization algorithm is executed to determine the target backlight brightness of multiple backlight zones corresponding to the multiple image regions.

[0165] In this step, the system makes the core backlight brightness decision. The system comprehensively utilizes the attention weights of each image region calculated in the previous step, the predictive visual characteristics of the near future predicted in step S2, and a pre-defined global optimization objective function. This global optimization objective function is a comprehensive evaluation criterion designed to balance multiple interrelated and potentially conflicting performance indicators, such as: ensuring the fidelity of accurate image content brightness reproduction, maintaining the smoothness of brightness transitions between adjacent backlight zones to avoid halos or block effects, ensuring the continuity of backlight brightness changes over time to avoid flicker, and minimizing the energy consumption of the entire backlight system. The system then executes an optimization algorithm (such as gradient descent or quadratic programming) to calculate an optimal target backlight brightness value for the backlight zone corresponding to each image region, guided by the global optimization objective function and considering the upper and lower limits of LED physical brightness constraints.

[0166] S5: Generate and output a drive signal based on the target backlight brightness to control the actual brightness of the multiple backlight zones;

[0167] In this step, the system converts the calculated target backlight brightness value into actual hardware control commands. For each backlight zone and its target backlight brightness determined in step S4, the system converts it into corresponding physical drive parameters. If the backlight hardware uses pulse width modulation (PWM) control, the target backlight brightness is converted into PWM signal parameters with a specific duty cycle; if analog current driving is used, the target backlight brightness is converted into a specific drive current value. This conversion process takes into account the photoelectric characteristics of the LED itself to ensure that the generated drive parameters enable the LED to emit light intensity that matches the target brightness. Finally, these generated PWM parameters or analog drive current parameters are used as drive signals and output to each LED unit or zone in the LCD backlight module through the drive circuit, thereby precisely controlling their respective actual luminous brightness.

[0168] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An integrated LED backlight control system for a liquid crystal display screen, characterized in that, The system includes the following modules: The data processing unit is used to receive the current video frame via the video input interface and perform image partitioning processing on the current video frame to obtain image region data for each of the multiple image regions. The predictive analysis unit is used to combine historical video frame information with the image region data of each of the multiple image regions, and apply a temporal prediction model to predict the predictive visual characteristics of each of the multiple image regions in the near future. Specifically, for each image region, brightness and motion features are extracted from the historical video frame information and the image region data of that image region; based on the extracted brightness and motion features, the temporal state corresponding to that image region is updated using the temporal prediction model; and based on the updated temporal state, the predictive brightness value and predictive brightness change trend of that image region in the near future are predicted as the predictive visual characteristics. The weight calculation unit is used to combine the image region data of each of the multiple image regions with the predictive visual characteristics, and apply a multi-scale spatiotemporal attention model to calculate the attention weights corresponding to each of the multiple image regions. A backlight optimization unit is used to execute an optimization algorithm based on the attention weights, the predictive visual characteristics, and a preset global optimization objective function to determine the target backlight brightness of multiple backlight zones corresponding to the multiple image regions. The global optimization objective function used by the backlight optimization unit comprehensively evaluates the following aspects: the accuracy of image content reproduction, which is measured based on the image region data of each image region and its corresponding attention weights; the smoothness of brightness transition between adjacent backlight zones, which is measured based on the brightness relationship between adjacent image regions, their corresponding attention weights, and image edge information between the image regions; the continuity of backlight brightness in the time series, which is measured based on the backlight brightness relationship between the current frame and the previous frame and the predictive brightness change trend contained in the predictive visual characteristics; and the energy consumption of the overall backlight system, which is measured based on the relationship between the target backlight brightness of each backlight zone and its preset power consumption model, the attention weights, and the overall visual characteristics of the current scene. A drive control unit is used to generate and output drive signals for controlling the actual brightness of the plurality of backlight zones based on the target backlight brightness.

2. The integrated LED backlight control system for a liquid crystal display screen according to claim 1, characterized in that, The data processing unit is specifically used for: Perform color space conversion processing on the received current video frame; According to the preset backlight physical partition layout, the current video frame, after color space conversion processing, is divided into the multiple image regions; And extract the corresponding image data for each image region as the image region data.

3. The integrated LED backlight control system for a liquid crystal display screen according to claim 1, characterized in that, The image region data for each image region, when the predictive analysis unit updates the temporal state, includes brightness-related features and motion-related features extracted from that image region; the predictive visual characteristics predicted through the temporal state include the predictive brightness value and predictive brightness change trend of that image region.

4. The integrated LED backlight control system for a liquid crystal display screen according to claim 1, characterized in that, The weight calculation unit is specifically used for: For each of the image regions, multi-scale spatial features of the image region are extracted based on its image region data; By combining the multi-scale spatial features with the predictive visual characteristics of the image region, the multi-scale spatiotemporal attention model is applied to comprehensively evaluate the visual importance of the image region in both spatial and temporal dimensions, and the attention weight corresponding to the image region is calculated.

5. The integrated LED backlight control system for a liquid crystal display screen according to claim 4, characterized in that, When applying the multi-scale spatiotemporal attention model, the weight calculation unit, for each of the image regions: Based on its image region data, the spatial saliency features and spatial detail features of the image region at multiple spatial scales are extracted; The temporal attention information of the image region is calculated by combining its predictive visual characteristics. The extracted spatial saliency features, spatial detail features, and calculated temporal attention information are then fused to generate the attention weight corresponding to the image region.

6. The integrated LED backlight control system for a liquid crystal display screen according to claim 1, characterized in that, In the global optimization objective function: The attention weights are used to dynamically adjust the relative importance of each image region when evaluating the accuracy of the image content reproduction, and to dynamically adjust the constraint strength applied when evaluating the smoothness of the brightness transition between adjacent backlight zones in combination with the image edge information between the image regions. The attention weights, combined with the overall visual characteristics of the current scene, are used to dynamically adjust the weighting ratio when evaluating the energy consumption of the overall backlight system. The predictive brightness change trend included in the predictive visual characteristics is used to dynamically guide the desired change target when assessing the continuity of the backlight brightness over time.

7. The integrated LED backlight control system for a liquid crystal display screen according to claim 1, characterized in that, The drive control unit is specifically used to convert the target backlight brightness of each of the plurality of backlight zones into pulse width modulation parameters or analog drive current parameters for controlling the luminous intensity of the corresponding backlight zones, and output the converted pulse width modulation parameters or analog drive current parameters as the drive signal.

8. A method for integrated LED backlight control of a liquid crystal display screen, applied to the system described in any one of claims 1-7, characterized in that, The method includes the following steps: S1. Receive the current video frame via the video input interface, and perform image partitioning processing on the current video frame to obtain image region data for each of the multiple image regions; S2. Combining historical video frame information with the image region data of each of the multiple image regions, a time-series prediction model is applied to predict the predictive visual characteristics of each of the multiple image regions in the near future. S3. Combining the image region data of each of the multiple image regions with the predictive visual characteristics, a multi-scale spatiotemporal attention model is applied to calculate the attention weights corresponding to each of the multiple image regions. S4. Using the attention weights, the predictive visual characteristics, and the preset global optimization objective function, an optimization algorithm is executed to determine the target backlight brightness of multiple backlight zones corresponding to the multiple image regions. S5: Generate and output a drive signal based on the target backlight brightness to control the actual brightness of the multiple backlight zones.

Images