Electronic screen partition brightness self-adaptive energy-saving control algorithm

Abstract

The invention discloses an electronic screen partition brightness self-adaptive energy-saving control algorithm, which relates to the technical field of video signal processing, and comprises the following steps: performing brightness domain unified processing on an input video frame, dividing partitions according to a screen, and constructing a partition logarithmic brightness histogram and cumulative distribution; extracting a partition high quantile and a full-screen reference quantile to generate a partition bright peak traction amount, and determining a dark part inflection point to generate a partition dark fringe conservation amount; based on the global brightness reference, the bright peak traction amount and the dark fringe conservation amount, generating a partition candidate light-emitting control amount, and performing neighborhood difference limitation to obtain a partition target light-emitting control amount; pixel compensation gains are generated during subarea brightness reduction; and constructing a time domain flicker hysteresis bandwidth amount, and executing hysteresis judgment and speed-limiting update to output partition light-emitting control amount and power consumption backspacing. High-brightness small target pressure darkness and bright part detail loss are restrained through bright peak quantile traction, the dark part level and the local contrast ratio are kept through dark fringe inflection point conservation, brightness jitter is reduced through adaptive hysteresis flash suppression, and responsiveness and the energy-saving effect are both considered.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of video signal processing, in particular to an electronic screen partition luminance adaptive energy-saving control algorithm. BACKGROUND

[0002] With the popularization of local dimming and partition light-emitting technology of self-luminous display panels, the display system usually needs to reduce the screen light-emitting power consumption as much as possible while ensuring the picture brightness, contrast and detail level, and to suppress the flicker phenomenon caused by the partition luminance change. The existing partition luminance energy-saving control generally determines the partition backlight / luminous intensity based on the full-screen average luminance, maximum luminance or simple histogram statistics, and then realizes time domain smoothing by limiting amplitude or low-pass filtering with fixed threshold. However, in actual video content, there may be high-brightness small targets and large-area dark textures in different partitions. If the partition control is only based on the average luminance or maximum value, it is easy to appear that the high-brightness target is too dark, the dark texture is smoothed or the halo of discontinuous partition boundary appears; if strong filtering or fixed hysteresis is used to suppress flicker, it will cause response lag, and the tailing or luminance jump will appear in the luminance change scene.

[0003] In addition, the partition light-emitting control has neighborhood coupling in space: the difference between the luminous intensities of adjacent partitions is too large, which easily causes obvious boundary fault and halo, so it is necessary to introduce spatial consistency constraint after the generation of partition control quantity; in time, the inter-frame fluctuation of partition control quantity will induce flicker, and the flicker sensitivity is related to the content change amplitude, so it is difficult to simultaneously consider stability and responsiveness with fixed hysteresis bandwidth or fixed update rate. On the other hand, the display system is often limited by the overall power budget, especially in high-brightness scenes or when multiple partitions simultaneously increase the brightness, if there is no partition priority fallback mechanism based on content characteristics, the power consumption may exceed the limit, the overall brightness may suddenly drop, or the local brightness may be abnormal, which affects the viewing experience and energy-saving control effect.

[0004] Therefore, there is an urgent need for a luminance adaptive energy-saving control algorithm for electronic screen partition light-emitting, which can simultaneously depict the bright peak and dark texture characteristics at the partition statistical level, consider the spatial neighborhood consistency and time domain flicker suppression, and realize controllable partition rollback under the power budget constraint, so as to reduce the power consumption while maintaining the picture contrast and dark details, and reducing the flicker and visual discomfort caused by the partition luminance fluctuation. SUMMARY

[0005] Based on the above-mentioned shortcomings of the prior art, the purpose of the present application is to provide an electronic screen partition luminance adaptive energy-saving control algorithm to solve the above technical problems.

[0006] To achieve the above-mentioned purpose, the present application provides the following technical scheme: an electronic screen partition luminance adaptive energy-saving control algorithm, comprising: The input video frame is uniformly processed in the brightness domain and divided into multiple partitions according to the screen, and a partition logarithmic brightness histogram and its cumulative distribution are constructed for each partition; The partition high quantile point and the full-screen reference quantile point are extracted from the cumulative distribution, and the partition bright peak traction quantity is generated based on the quantile difference in the logarithmic domain through saturation gate mapping; The discrete second-order difference of the cumulative distribution is calculated in the preset dark interval to determine the dark inflection point, and the partition dark pattern conservation quantity is generated based on the density difference above and below the inflection point through saturation gate mapping; The partition candidate luminous control quantity is generated based on the global brightness reference, the partition bright peak traction quantity and the partition dark pattern conservation quantity, and the partition target luminous control quantity is obtained through neighborhood difference limitation; The pixel compensation gain is generated when the partition target luminous control quantity decreases, and the highlight ceiling is set; The time-domain flicker hysteresis bandwidth quantity is constructed based on the inter-frame changes of the partition bright peak traction quantity and the partition dark pattern conservation quantity, the output partition luminous control quantity is obtained by performing hysteresis judgment and speed limiting update on the partition target luminous control quantity through the time-domain flicker hysteresis bandwidth quantity, and the output partition luminous control quantity is rolled back when the power consumption exceeds the budget.

[0007] The application further sets that the partition high quantile point and the full-screen reference quantile point are preset cumulative distribution quantile positions, the quantile positions are obtained by threshold query on the cumulative distribution, and the partition bright peak traction quantity monotonically changes with the logarithmic domain quantile difference and saturates at a preset upper limit.

[0008] The application further sets that the partition logarithmic brightness histogram is obtained by down-sampling statistics on the partition pixels, and the cumulative distribution is obtained by accumulating and normalizing the histogram bin by bin.

[0009] The application further sets that the dark interval is a continuous bin set corresponding to the low brightness range in the logarithmic brightness histogram, and the dark inflection point is a bin position that makes the absolute value of the discrete second-order difference of the cumulative distribution reach an extreme value in the dark interval.

[0010] The application further sets that the calculation logic of the partition dark pattern conservation quantity includes: determining the histogram count cumulative value of the bins below the dark inflection point as the density below the inflection point, and determining the histogram count cumulative value of the bins within a preset bandwidth range above the dark inflection point as the density above the inflection point; taking the logarithmic difference between the density above the inflection point and the density below the inflection point as the input, and obtaining the partition dark pattern conservation quantity through saturation gate mapping.

[0011] The application is further configured that the generation of the partition candidate luminous control quantity comprises: taking a complementary quantity of the global luminance reference, the partition light peak pulling quantity and the partition dark line conservation quantity, multiplying the complementary quantities to obtain a joint complementary quantity, and taking a complementary quantity of the joint complementary quantity to obtain a synthesis quantity; limiting the synthesis quantity through clamping processing between a preset lower limit and an upper limit of the luminous control quantity to obtain the partition candidate luminous control quantity.

[0012] The application is further configured that the neighborhood difference restriction determines a set of adjacent partitions based on a preset neighborhood structure, and determines a neighborhood upper limit and a neighborhood lower limit based on the candidate luminous control quantities of the adjacent partitions, wherein the neighborhood upper limit is determined by the maximum value of the candidate luminous control quantities of the adjacent partitions combined with a preset neighborhood difference threshold, and the neighborhood lower limit is determined by the minimum value of the candidate luminous control quantities of the adjacent partitions combined with the preset neighborhood difference threshold. The candidate luminous control quantity of any partition is clamped into a range defined by the neighborhood upper limit and the neighborhood lower limit, and the execution is repeated until the neighborhood difference constraint is met or a convergence condition is reached.

[0013] The application is further configured that the time domain flicker hysteresis bandwidth quantity is obtained by power mapping the inter-frame change amplitude of the partition light peak pulling quantity and the partition dark line conservation quantity, and the smaller the change amplitude is, the larger the hysteresis bandwidth quantity is, wherein the inter-frame change amplitude is the larger one of the inter-frame difference amplitude of the partition light peak pulling quantity and the inter-frame difference amplitude of the partition dark line conservation quantity.

[0014] The application is further configured that when the difference between the partition target luminous control quantity and the last frame output partition luminous control quantity is less than the hysteresis bandwidth quantity, the last frame output is maintained without updating, and when the foregoing condition is not met, a speed limiting update is performed on the partition target luminous control quantity according to an upper limit of the update step size related to the hysteresis bandwidth quantity to obtain the output partition luminous control quantity.

[0015] The application is further configured that when the estimated power consumption exceeds the power consumption budget, the output partition luminous control quantity is gradually lowered according to a partition priority determined by the partition light peak pulling quantity and the partition dark line conservation quantity, and the estimation is repeated until the power consumption budget is met, wherein the partition priority is determined by the larger one of the partition light peak pulling quantity and the partition dark line conservation quantity, and the gradual lowering is performed from low to high according to the partition priority.

[0016] The application provides an electronic screen partition brightness adaptive energy-saving control algorithm, which uniformly processes input video frames in the brightness domain, divides the screen into multiple partitions, constructs a partition logarithmic brightness histogram and its cumulative distribution for each partition, extracts a partition high quantile point and a full-screen reference quantile point from the cumulative distribution, generates a partition bright peak traction based on the quantile difference in the logarithmic domain, calculates a discrete second-order difference of the cumulative distribution in a preset dark interval to determine a dark inflection point, generates a partition dark pattern conservation based on the density difference of the inflection point, generates a partition candidate luminous control quantity based on the global brightness reference, the partition bright peak traction and the partition dark pattern conservation, and obtains a partition target luminous control quantity through neighborhood difference limitation, generates a pixel compensation gain when the partition target luminous control quantity decreases and sets a highlight ceiling, constructs a time-domain flicker hysteresis bandwidth based on the inter-frame changes of the partition bright peak traction and the partition dark pattern conservation, performs hysteresis judgment and speed limit update on the partition target luminous control quantity through the time-domain flicker hysteresis bandwidth to obtain an output partition luminous control quantity, and retreats the output partition luminous control quantity when the power consumption exceeds the budget, and the beneficial effects include: 1. Bright peak quantile traction: the partition cumulative distribution is constructed in the logarithmic brightness domain, the partition high quantile point and the full-screen reference quantile point are extracted, and the partition bright peak traction is generated based on the quantile difference in the logarithmic domain through saturation gating, so that the partition luminous control forms a traction type retention for high-light small targets and bright peak patterns, avoiding the high-light small targets being darkened and the bright part details being lost in the energy-saving and brightness-reducing process caused by the traditional mean, low quantile or single threshold compression strategy; at the same time, compared with the strategy of directly using the maximum value to pull the partition luminous, the saturation gating suppresses the excessive driving of the control quantity by the abnormal bright peak, reduces the risk of local glare, energy consumption fluctuation and brightness jump, and improves the stability and controllability of the bright part; 2. Dark pattern inflection point conservation: the discrete second-order difference of the cumulative distribution is calculated in the preset dark interval to determine the dark inflection point, and the partition dark pattern conservation is generated based on the histogram count accumulation difference of the dark inflection point, so that the control decision explicitly perceives the change of the dark level distribution and the fidelity demand of the visible details in the dark part, avoids the visible details in the dark part being excessively compressed, the level collapsing or the black sticking when the partition luminous decreases, and thus maintains the dark level and local contrast while saving energy; 3. Adaptive hysteresis flicker suppression: the time-domain flicker hysteresis bandwidth is constructed based on the inter-frame changes of the partition bright peak traction and the partition dark pattern conservation, and the hysteresis judgment and speed limit update are performed accordingly, so that the upper limit of the hysteresis bandwidth and the update step size is self-adaptively adjusted according to the content change amplitude: when the change amplitude is small, the hysteresis bandwidth is increased and the update is tightened to suppress small dithering flicker, and when the change amplitude is large, the hysteresis bandwidth is reduced and the update is relaxed to improve the response speed, avoiding the tailing and sudden jump introduced by fixed hysteresis or fixed filtering, so as to balance the display stability and responsiveness.

[0017] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 The flowchart illustrates an adaptive energy-saving control algorithm for partitioned brightness of an electronic screen, as shown in an exemplary embodiment of the present invention. Detailed Implementation

[0019] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0020] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0021] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0022] Electronic screen zone brightness adaptive energy-saving control algorithm, such as Figure 1 As shown, it includes: The input video frames are processed in a uniform brightness domain and divided into multiple partitions according to the screen. For each partition, a partition logarithmic brightness histogram and its cumulative distribution are constructed. The high quantiles and full-screen reference quantiles of the partitions are extracted from the cumulative distribution, and the quantile difference between the two in the logarithmic domain is used to generate the traction of the bright peaks of the partitions through saturation gating mapping. Within a preset dark area interval, the discrete second-order difference of the cumulative distribution is calculated to determine the dark area inflection point, and the partition dark fringe conservation quantity is generated based on the density difference above and below the inflection point through saturation gating mapping. Based on the global brightness reference, the regional brightness peak traction amount and the regional dark fringe conservation amount, the regional candidate luminescence control amount is generated, and the regional target luminescence control amount is obtained by the neighborhood difference constraint. When the target luminance control amount in the partition is reduced, pixel compensation gain is generated and a high brightness capping is set; The temporal hysteresis bandwidth is constructed based on the inter-frame variation of the partitioned bright peak traction and partitioned dark fringe conservation. The output partitioned luminous control quantity is obtained by performing hysteresis judgment and rate limiting update on the partitioned target luminous control quantity through the temporal hysteresis bandwidth. When the power consumption exceeds the budget, the output partitioned luminous control quantity is rolled back.

[0023] Specifically, after receiving the input video frame, the display controller first performs luminance domain unification processing on the input video frame to make the pixel brightness of different video sources comparable under the same statistical scale. Then, it divides the screen into multiple partitions according to preset partitioning rules, with each partition corresponding to a set of pixels. For each partition, the controller maps the brightness of the pixels in that partition to the logarithmic luminance domain to highlight relative brightness differences and weaken the impact of absolute brightness scale differences on the statistical results. This makes the subsequent quantile extraction more sensitive to small bright targets and bright peak shapes, and more stable to overall grayscale drift. Furthermore, the partitioned logarithmic brightness histogram is obtained by downsampling and statistically analyzing the pixels in each partition. Downsampling employs a preset spatial sampling rule to ensure uniform coverage of sampling points within the partition, reducing computational load and avoiding statistical bias caused by sampling only local areas. The logarithmic brightness of each sampled pixel is counted within its corresponding bin, forming the partitioned logarithmic brightness histogram. Subsequently, the controller performs bin-by-bin accumulation on the histogram to obtain a cumulative count sequence, and then normalizes the cumulative count sequence to obtain a cumulative distribution. Bin-by-bin accumulation makes the cumulative distribution naturally monotonic, facilitating stable threshold lookup, while normalization eliminates the scale effect caused by differences in the number of samples from different partitions, making the cumulative distributions of different partitions comparable on the same probability scale. After obtaining the cumulative distribution, the high quantile points of the partition and the full-screen reference quantile points are set as preset cumulative distribution quantile positions, and the corresponding logarithmic brightness values ​​are obtained by performing a threshold query on the cumulative distribution. The threshold query is as follows: locate the bin position in the cumulative distribution where the cumulative probability first reaches or exceeds the target quantile position, and use the center value of the bin or the interpolated value based on the adjacent bins as the quantile brightness; wherein the high quantile points of the partition are used to characterize the statistical position of the bright peak group in the partition, and the full-screen reference quantile points are used to provide a consistent reference standard across partitions, thereby avoiding the inter-partition brightness scale drift caused by relying solely on the statistics within the partition.

[0024] Subsequently, the controller calculates the quantile difference between the high quantile point of the partition and the full-screen reference quantile point in the logarithmic brightness domain, and inputs the quantile difference into a saturation gating map to generate the partition brightness peak traction amount. The saturation gating map satisfies the following: the brightness peak traction amount monotonically increases as the quantile difference increases, thereby enhancing the traction-based maintenance capability for bright small targets and brightness peak shapes; when the quantile difference exceeds the gate interval corresponding to the preset upper limit, the brightness peak traction amount enters the saturation region, thereby suppressing the over-driving of the luminous emission control amount by abnormal brightness peaks or transient strong light points, and reducing brightness jumps and control instability caused by extreme pixels. Through the above mechanism, the brightness peak traction amount can reflect the relative brightness peak prominence of the partition, and can also achieve robust suppression of abnormal brightness peaks through saturation characteristics, providing a stable and controllable driving basis for subsequent partition luminous emission control.

[0025] After constructing the logarithmic brightness histograms and their cumulative distributions for each zone, the display controller enters the decision-making process for preserving dark area structure. The controller first determines a preset dark area interval in the logarithmic brightness histogram of each zone. This dark area interval is a continuous set of bins corresponding to the low-brightness range in the logarithmic brightness histogram, used to limit the statistical focus to the dark pixel cluster area, avoiding interference from the distribution of mid-to-high brightness areas on dark feature localization. Through this limitation, dark area analysis relies only on the brightness morphology changes within the dark area, thus better aligning with the control objective of preserving dark area hierarchy. Furthermore, the controller calculates the discrete second-order difference of the cumulative distribution within the dark area, and determines the dark inflection point by the bin position where the absolute value of the discrete second-order difference reaches its extreme value. The principle is that the cumulative distribution in the dark area often exhibits a pattern of low-brightness accumulation—transition—entering visible detail distribution. The position with the most significant curvature change corresponds to the interface where the distribution pattern transitions from concentrated accumulation to unfolding. The discrete second-order difference is used to characterize the bending intensity of the cumulative distribution, and taking its extreme value allows for stable positioning of this interface at the discrete bin scale, making the dark inflection point a key anchor point for the dark area's hierarchical distribution pattern. By using the cumulative distribution instead of a single bin count for positioning, the impact of noise and sporadic pixels on inflection point positioning can be reduced, improving the stability of dark feature extraction. After determining the dark inflection point, the controller calculates the partitioned dark fringing conservation quantity. Specifically, the accumulated histogram count of the bins at and below the dark inflection point is determined as the density below the inflection point, which characterizes the pixel aggregation intensity of darker areas; the accumulated histogram count of the bins within a preset bandwidth range above the dark inflection point is determined as the density above the inflection point, which characterizes the pixel distribution intensity of the transition zone of visible details in the dark area. The preset bandwidth range is used to limit the coverage width of the statistics above the inflection point, so that the density is more focused on the dark detail-bearing area and is not diluted by higher brightness levels. Subsequently, the partitioned dark fringing conservation quantity is obtained by saturation gating mapping using the logarithmic difference between the density above and below the inflection point as input. The principle is as follows: the logarithmic difference can reflect the changing trend of the relative proportion above and below the inflection point, thus characterizing the degree of dark layer development or the intensity of detail carrying capacity; when the density above the inflection point increases relatively, it indicates that the dark transition zone is richer and the demand for detail carrying capacity is stronger, and the dark fringes conservation amount is enhanced to suppress the dark collapse caused by brightness reduction; when the logarithmic difference abnormally increases, the dark fringes conservation amount is saturated at the upper limit through saturation gating, avoiding excessive dark preservation caused by noise, quantization steps or local abnormal dark fringes, thus forming a controllable balance between dark detail preservation and energy saving brightness reduction.

[0026] After obtaining the global brightness reference, the zonal brightness peak pull, and the zonal dark fringe conservation, the display controller enters the stage of synthesizing and spatially consistent constraining the zonal luminance control quantities. The controller first uses the global brightness reference as a unified base for the overall screen brightness, ensuring that the control quantities of each zonal operate in tandem under the same reference scale when ambient brightness, user-set brightness, or system strategy changes. Simultaneously, it considers the zonal brightness peak pull as an enhanced signal for bright area fidelity requirements and the zonal dark fringe conservation as an enhanced signal for dark area gradation requirements, thus ensuring that the candidate luminance control quantities not only reflect how bright it should be but also which zonal areas should not be suppressed.

[0027] Furthermore, a complementary product is used to synthesize candidate luminance control quantities for different regions. Specifically, the controller takes complementary values ​​for the global luminance reference, the regional bright peak pulling amount, and the regional dark fringe conservation amount, respectively. The complementary values ​​are used to characterize the space that allows for a reduction in luminance. The three complementary values ​​are multiplied to obtain a joint complementary value, which is used to characterize the degree to which luminance can still be reduced when all three types of constraints (global, bright, and dark) are satisfied simultaneously. The combined value is obtained by taking the complementarity of the joint complementary value. The principle is that by performing a product operation in the complementary domain, the joint complementary value remains large only when all three types of constraints are weak, thus the combined value is small to allow for more energy saving. As long as either the bright peak pulling amount or the dark fringe conservation amount is strong, or the global luminance reference is high, the corresponding complementary value tends to decrease, causing the joint complementary value to decrease accordingly, and the combined value to increase accordingly. This raises the candidate luminance control quantity to meet the fidelity requirements of the bright or dark regions and avoids mutual cancellation under linear weighting between different constraints. Subsequently, the controller clamps the synthesized quantity to a preset lower and upper limit of the light emission control quantity, so as to constrain the output to fall within the driveable and controllable light emission control range, thus obtaining the partitioned candidate light emission control quantity. The principle of clamping is to avoid extreme inputs causing the candidate control quantity to go out of bounds, thereby ensuring that the subsequent driving curve, power consumption model and panel electro-optic response mapping can be executed stably.

[0028] After obtaining the candidate luminance control values ​​for each zone, the controller applies neighborhood difference constraints to obtain the target luminance control value for that zone. Specifically, the controller determines the set of adjacent zones based on a preset neighborhood structure, and determines the upper and lower boundaries of the neighborhood based on the candidate luminance control values ​​of the adjacent zones. The upper boundary of the neighborhood is determined by combining the maximum value of the candidate luminance control values ​​of the adjacent zones with a preset neighborhood difference threshold, and the lower boundary of the neighborhood is determined by combining the minimum value of the candidate luminance control values ​​of the adjacent zones with a preset neighborhood difference threshold. The controller clamps the candidate luminance control values ​​of any zone within the range defined by the upper and lower boundaries of the neighborhood, and repeats this process until the neighborhood difference constraint is met or the convergence condition is reached. The principle is that zone dimming is a spatial discrete control. If the difference between the control values ​​of adjacent zones is too large, it is easy to produce brightness discontinuities, halos, or blocky appearances at the zone boundaries. By constraining the neighborhood difference, the control value of each zone is constrained within the allowable range of the neighborhood, achieving spatial smoothing without relying on global blur filtering, thereby suppressing boundary artifacts while retaining the local differences required for bright peaks and dark fringes. The iterative convergence setting allows the local clamping constraint to gradually propagate across the entire screen, avoiding overall brightness drift or local over-correction caused by a one-time forced clamping.

[0029] After obtaining the target luminance control value for each zone, a pixel compensation gain is generated and a brightness cap is set when the target luminance control value for each zone decreases. The principle is as follows: when the zone luminance decreases, the usable brightness range of the pixel is compressed. To avoid further compression of mid-to-low brightness details, the controller generates a pixel compensation gain based on the decrease in the target luminance control value for each zone relative to the reference luminance control value, and controls the increase in pixel values ​​to maintain subjective brightness and local contrast. At the same time, a brightness cap is set to limit the overshoot of bright pixels after compensation, avoiding brightness overflow, saturation clipping, or color drift caused by compensation. This achieves a synergy between reducing zone luminance to save energy and maintaining visual appeal through pixel compensation.

[0030] After obtaining the target luminous emission control values ​​for each zone, the display controller enters the temporal stabilization control and power consumption constraint stage. The controller first obtains the zone brightness peak pulling amount and zone dark ripple conservation amount for each zone in the current frame and the previous frame, and calculates the inter-frame difference amplitude between the two. In order to make the constraint signal, which is sensitive to "visual changes", dominate the temporal stabilization strategy, the inter-frame change amplitude is defined as the larger of the inter-frame difference amplitude of the zone brightness peak pulling amount and the inter-frame difference amplitude of the zone dark ripple conservation amount. Thus, when a significant content change occurs on either the bright or dark side, it can be captured by this change amplitude and drive the subsequent update strategy. Subsequently, the controller maps the inter-frame variation amplitude input to a power order to obtain the temporal flicker hysteresis bandwidth, satisfying a monotonically inverse relationship where the smaller the variation amplitude, the larger the hysteresis bandwidth. The principle is as follows: when the content change is small, the bright peak pulling amount and the dark fringing conservation amount only fluctuate slightly between frames. At this time, increasing the hysteresis bandwidth can raise the update trigger threshold, so that the partitioned illumination control amount remains stable under small disturbances, suppressing brightness jitter and flicker caused by noise, quantization steps, or small statistical fluctuations. When the content change is large, the hysteresis bandwidth decreases accordingly, so that the control system reduces the inertia of remaining unchanged, so as to respond more quickly to scene switching or sudden changes in brightness and darkness structure, and avoid the generation of trailing and inconsistent brightness lag.

[0031] Furthermore, the temporal flicker hysteresis bandwidth is used to perform hysteresis determination and rate-limited update on the partition target luminescence control quantity to obtain the output partition luminescence control quantity. Specifically, the controller compares the difference between the partition target luminescence control quantity and the output partition luminescence control quantity of the previous frame. When the difference is less than the hysteresis bandwidth, the output of the previous frame is not updated, so that the control quantity has the ability to resist perturbations within the hysteresis interval. When the aforementioned condition is not met, the controller performs rate-limited update on the partition target luminescence control quantity according to the update step size upper limit related to the hysteresis bandwidth to obtain the output partition luminescence control quantity of the current frame. The principle is that: hysteresis determination is used to determine whether an update is needed, and rate-limited update is used to constrain the update speed. The two work together to avoid back-and-forth jumps caused by frequent crossings near the threshold, while limiting the abrupt jump caused by excessive changes in a single frame. Moreover, since the update step size upper limit is related to the hysteresis bandwidth, a stricter rate-limit constraint is applied when the content change is small to enhance stability, and a looser rate-limit constraint is applied when the content change is large to improve the response speed, thereby achieving an adaptive balance between flicker suppression and response sensitivity.

[0032] After obtaining the output luminance control values ​​for each partition, a power budget constraint is further introduced to achieve the overall energy-saving goal. The controller estimates the power consumption of the current frame based on the output partition luminance control values ​​and compares it with the power budget. When the estimated power consumption exceeds the power budget, the controller constructs partition priorities and gradually lowers the output partition luminance control values ​​from low to high priority, while cyclically estimating power consumption during the lowering process until the power budget is met. The partition priority is determined by the larger of the partition brightness peak pull and the partition shadow texture conservation value, giving higher priority to partitions with stronger requirements for brightness fidelity or shadow level maintenance. This ensures that partitions containing bright small targets or shadow level sensitive partitions are protected first during power back-off. The principle is that when the total power consumption is limited, proportional reduction can easily cause significant degradation in the visual appearance of key areas. By using a priority strategy dominated by brightness peak pull and shadow texture conservation values, the impact of power back-off can be concentrated on partitions with lower visual sensitivity, so that the overall power consumption meets the budget while maintaining the consistency and stability of brightness details and shadow levels as much as possible.

[0033] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An adaptive energy-saving control algorithm for zoned brightness of electronic screens, characterized in that: include: The input video frames are processed in a uniform brightness domain and divided into multiple partitions according to the screen. For each partition, a partition logarithmic brightness histogram and its cumulative distribution are constructed. The high quantiles and full-screen reference quantiles of the partitions are extracted from the cumulative distribution, and the quantile difference between the two in the logarithmic domain is used to generate the traction of the bright peaks of the partitions through saturation gating mapping. Within a preset dark area interval, the discrete second-order difference of the cumulative distribution is calculated to determine the dark area inflection point, and the partition dark fringe conservation quantity is generated based on the density difference above and below the inflection point through saturation gating mapping. Based on the global brightness reference, the regional brightness peak traction amount and the regional dark fringe conservation amount, the regional candidate luminescence control amount is generated, and the regional target luminescence control amount is obtained by the neighborhood difference constraint. When the target luminance control amount in the partition is reduced, pixel compensation gain is generated and a high brightness capping is set; The temporal hysteresis bandwidth is constructed based on the inter-frame variation of the partitioned bright peak traction and partitioned dark fringe conservation. The output partitioned luminous control quantity is obtained by performing hysteresis judgment and rate limiting update on the partitioned target luminous control quantity through the temporal hysteresis bandwidth. When the power consumption exceeds the budget, the output partitioned luminous control quantity is rolled back.

2. The electronic screen partition brightness adaptive energy-saving control algorithm according to claim 1, characterized in that, The high quantiles of the partition and the reference quantiles of the full screen are preset cumulative distribution quantile positions. The quantile positions are obtained by querying the threshold of the cumulative distribution. The traction of the bright peak in the partition changes monotonically with the logarithmic domain quantile difference and saturates at the preset upper limit.

3. The electronic screen partition brightness adaptive energy-saving control algorithm according to claim 1, characterized in that, The partitioned logarithmic brightness histogram is obtained by downsampling and statistically analyzing the pixels in the partitions, and the cumulative distribution is obtained by summing the histograms bin by bin and normalizing them.

4. The electronic screen partition brightness adaptive energy-saving control algorithm according to claim 1, characterized in that, The dark region is a set of continuous bins corresponding to the low brightness range in the logarithmic brightness histogram. The dark inflection point is the bin position in the dark region where the absolute value of the cumulative distribution discrete second difference reaches an extreme value.

5. The electronic screen partition brightness adaptive energy-saving control algorithm according to claim 1, characterized in that, The calculation logic of the partition dark fringe conservation quantity includes: determining the accumulated histogram count of the bins at and below the dark inflection point as the density below the inflection point, and determining the accumulated histogram count of the bins within the preset bandwidth range above the dark inflection point as the density above the inflection point; using the logarithmic difference between the density above the inflection point and the density below the inflection point as input, and obtaining the partition dark fringe conservation quantity through saturation gating mapping.

6. The electronic screen partition brightness adaptive energy-saving control algorithm according to claim 1, characterized in that, The generation of candidate emission control quantities for each region includes: taking complementary values ​​for the global brightness reference, the region bright peak pulling amount, and the region dark fringe conservation amount; multiplying the complementary values ​​to obtain a joint complementary value; and taking the complementarity of the joint complementary value to obtain a composite value; and clamping the composite value to limit it between the preset lower limit and upper limit of the emission control quantity to obtain the candidate emission control quantity for each region.

7. The electronic screen partition brightness adaptive energy-saving control algorithm according to claim 1, characterized in that, The neighborhood difference constraint determines the set of adjacent partitions based on the preset neighborhood structure, and determines the upper and lower boundaries of the neighborhood based on the candidate emission control quantities of the adjacent partitions. The upper boundary of the neighborhood is determined by the maximum value of the candidate emission control quantities of the adjacent partitions combined with the preset neighborhood difference threshold, and the lower boundary of the neighborhood is determined by the minimum value of the candidate emission control quantities of the adjacent partitions combined with the preset neighborhood difference threshold. Clamp the candidate emission control values ​​of any partition into the range defined by the upper and lower bounds of the neighborhood, and repeat the process until the neighborhood difference constraint is satisfied or the convergence condition is met.

8. The electronic screen partition brightness adaptive energy-saving control algorithm according to claim 1, characterized in that, The temporal hysteresis bandwidth is obtained by power-mapping the inter-frame variation amplitudes of the partitioned bright peak pull and the partitioned dark fringe conservation. The smaller the variation amplitude, the larger the hysteresis bandwidth. The inter-frame variation amplitude is the larger of the inter-frame difference amplitudes of the partitioned bright peak pull and the partitioned dark fringe conservation.

9. The electronic screen partition brightness adaptive energy-saving control algorithm according to claim 8, characterized in that, When the difference between the partition target luminous control quantity and the output partition luminous control quantity of the previous frame is less than the hysteresis bandwidth, the output of the previous frame is not updated. When the aforementioned condition is not met, the partition target luminous control quantity is updated at a rate-limited manner according to the upper limit of the update step size related to the hysteresis bandwidth, and the output partition luminous control quantity is obtained.

10. The electronic screen partition brightness adaptive energy-saving control algorithm according to claim 9, characterized in that, When the estimated power consumption exceeds the power consumption budget, the output partition light emission control quantity is gradually reduced according to the partition priority determined by the partition bright peak traction quantity and the partition dark fringe conservation quantity, and the estimation is repeated until the power consumption budget is met. The partition priority is determined by the larger of the partition bright peak traction quantity and the partition dark fringe conservation quantity, and is gradually reduced according to the partition priority from low to high.

Images