A method and system for local dimming of a smartphone backlight

Abstract

This invention discloses a method and system for local dimming backlighting in smartphones, belonging to the field of smartphone dimming. The method includes the following steps: S1, screen partitioning; S2, content analysis and parameter acquisition; S3, target brightness mapping and color depth conversion prediction; S4, brightness command generation and block control; S5, color depth conversion execution and visual compensation; S6, dynamic feedback adjustment. Compared with traditional technologies, this invention solves the halo problem around small, bright objects in fixed physical partitions by dynamically reconstructing the backlight block boundaries through an embedded programmable partitioning engine; it solves the color banding defect caused by independent pixel driving when brightness suddenly drops by coupling and predicting the target brightness and color depth mode through a collaborative decision-making module; and it solves the visual fatigue caused by strong light response lag and touch flicker by fusing real-time data streams of ambient light and touch behavior through a multi-source sensing module.

Description

Technical Field

[0001] This invention belongs to the field of dimming backlighting, and more specifically, relates to a method for local dimming backlighting in smartphones. This invention also relates to a local dimming backlighting system for smartphones. Background Technology

[0002] Current high-end smartphone displays commonly employ local dimming technology, which enhances contrast by independently controlling backlight zones. Existing solutions mainly fall into two categories: hardware-driven fixed-zone dimming and software-driven content-adaptive dimming. The core implementation involves analyzing the brightness of each zone based on image content and then adjusting the LED array using PWM or current-driven adjustment.

[0003] While existing technologies can improve power consumption and image quality, they still suffer from fundamental flaws: insufficient flexibility in local dimming, with fixed hardware partitioning causing halos around small, bright objects; lack of collaborative optimization, with brightness adjustment and pixel driving operating independently, resulting in color banding in pixels that haven't reduced color depth when local backlight is significantly reduced; poor environmental adaptability, with delayed local dimming response in strong light causing a washed-out image, and dimming jitter causing eye fatigue during frequent touches; and one-sided power consumption control, as relying solely on backlight adjustment is insufficient to handle the pixel-level power consumption pressure of high-resolution screens. These issues have become key bottlenecks restricting the balance between HDR performance and battery life in smartphones. Therefore, we propose a local dimming backlighting method and system for smartphones. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a method and system for local dimming backlighting in smartphones. This invention overcomes the three major limitations of current local dimming technologies: rigid partitioning, lack of coordination, and environmental lag. It achieves hardware-reconfigurable partitioning control, pixel compensation linked to brightness and color depth, and closed-loop feedback adapted to human eyes. While eliminating halos and color banding, it achieves power consumption-aware full-scene image quality optimization, providing a high-precision, low-latency backlighting solution for smartphone HDR displays.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for local dimming backlight in a smartphone includes the following steps:

[0007] S1. Screen partitioning: The backlight area of ​​the smartphone's display panel is divided into multiple independently controllable backlight dimming blocks. The size and shape of the partitions are dynamically adjusted or selected from a set of preset rules based on the characteristics of the current display content, system power consumption strategy, or ambient light conditions.

[0008] S2. Content analysis and parameter acquisition: Real-time analysis of the brightness and color distribution characteristics of the frame image to be displayed during the frame buffer rendering stage or after the compositing layer processing.

[0009] S3. Target brightness mapping and color depth conversion prediction: Based on the local brightness characteristics and environmental parameters obtained from step S2, an independent target backlight brightness value is predicted and mapped for each backlight dimming block through a pre-stored lookup table or dynamic calculation model; at the same time, an appropriate color depth conversion mode target is predicted for the corresponding display panel area.

[0010] S4. Brightness command generation and block control: The target backlight brightness value of each backlight dimming block obtained in step S3 and its corresponding color depth conversion mode target are combined with the current load and power consumption budget constraints of the smartphone processor to generate an optimized set of local backlight brightness control commands.

[0011] S5. Color depth conversion execution and visual compensation: Based on the color depth conversion mode target corresponding to each block predicted in step S3, perform real-time color depth conversion processing on the original frame image data.

[0012] S6. Dynamic feedback adjustment: After steps S4 and S5 are executed, monitor the actual brightness output of each backlight dimming block, the actual display effect of the screen panel, and changes in user interaction status; use at least one ambient light sensor built into the smartphone to sample the reflected light of the actual display output brightness near the screen area, and combine it with the frequency of user touch operation or significant changes in content areas to establish closed-loop feedback.

[0013] Preferably, in step S1, when dynamically adjusting the size and shape of the partitions, adjacent blocks are automatically merged when the displayed content contains a large area of ​​dark field; when a high-contrast edge area is detected, the partition density is increased along the edge direction; and the minimum partition size is forced when the ambient light illuminance is >10,000 lux.

[0014] Preferably, in step S3, the dynamic calculation model is a lightweight convolutional neural network (CNN). The input layer receives the brightness histogram of the partition block, the mean value of the HSV color space, and the ambient light color temperature. The output layer synchronously generates the target backlight brightness and 8-bit / 10-bit color depth mode flag.

[0015] Preferably, in step S5, the visual compensation includes applying spatial filtering based on human visual characteristics to the color depth converted block to enhance high-frequency details; and superimposing a saturation gain coefficient that is positively correlated with the target backlight brightness of the block.

[0016] Preferably, in step S6, the closed-loop feedback adopts a dual-threshold mechanism: when the actual brightness deviates from the target value by more than 15% and lasts for more than 3 frames, emergency brightness calibration is triggered; when the user touch operation frequency is more than 5 times / second, the partition adjustment is frozen.

[0017] Preferably, the target backlight brightness value Calculated using the perception optimization model:

[0018]

[0019] in, The average brightness of pixels within block k. The maximum / minimum brightness values ​​within block k. The ambient light color temperature is α, β, and γ, which are weighting coefficients α+β+γ=1. This optimization model is a nonlinear sensing model that integrates the three factors of space, environment, and power consumption.

[0020] Preferably, saturation gain in visual compensation The compression intensity is adaptively adjusted according to color depth, specifically as follows:

[0021]

[0022] in, This is the amount by which the color depth is reduced; for example, if 10 bits is reduced to 8 bits, then Δbit = 2. This represents the target backlight brightness value for this block. This is the maximum backlight brightness of the system. The human eye sensitivity factor, η∈[0.1,0.3], increases in low-light environments.

[0023] A local dimming backlight system for a smartphone, the system being used to implement the above-described method, comprising:

[0024] The dynamic partition control module is used to divide the backlight area of ​​the display panel into multiple independently adjustable backlight dimming blocks; the environment fusion processor synchronously processes the multi-dimensional parameters output by the front light sensor, attitude sensor, and battery management IC.

[0025] The collaborative decision-making computing module includes: a dual-output prediction processor, which outputs independent target backlight brightness values ​​and adapted color depth conversion mode flags for each block in parallel based on brightness distribution characteristics and environmental parameters; and a power consumption constraint arbiter, which receives the SoC load status and the energy consumption budget threshold from the temperature sensor in real time and performs dynamic scaling on the target brightness values.

[0026] The partition drive execution module includes: a backlight instruction compilation unit that encodes the target brightness value and color depth mode flag into a partition block control instruction set; and a multi-channel driver IC array that is physically connected to the backlight LED matrix and outputs differentiated backlight current in response to the instruction set.

[0027] The color depth conversion and compensation engine, integrated into the display driver chip, includes: a real-time bit depth converter that performs bit depth compression on the original pixel data based on the color depth mode flag; and a visual perception compensation unit that injects adaptive contrast gain and color saturation correction signals into the compressed data.

[0028] The closed-loop feedback control module includes: a reflected light monitoring unit, an ambient light sensor array located on the screen bezel, which captures the actual output light signal of the panel; and a cross-domain adaptive controller, which dynamically updates the partitioning rules and compensation parameters based on brightness output deviation, user touch events, and content change signals.

[0029] Preferably, the dynamic partition control module includes a hardware-level partition block metadata storage, which stores a topology map dynamically generated by the programmable partition engine. This map switches to high resolution mode as the HDR content is activated. The multi-source perception processing module has a built-in high color gamut marker unit. When an area with DCI-P3 coverage > 90% is detected, it outputs a color depth protection latch signal to the collaborative decision-making module.

[0030] The technical effects and advantages of this invention are as follows: This invention provides a method and system for local dimming backlight in smartphones. Compared with traditional technologies, it solves the halo problem around small, bright objects in fixed physical partitions by dynamically reconstructing the backlight block boundaries through an embedded programmable partitioning engine; and solves the color banding defect caused by independent pixel driving when brightness drops suddenly by coupling and predicting the target brightness and color depth mode through a collaborative decision module.

[0031] By fusing real-time data streams of ambient light and touch behavior through a multi-source sensing module, visual fatigue caused by lag in response to strong light and touch flicker is resolved; by implementing block-adaptive color depth reduction calculations through a bit depth conversion engine, the battery life bottleneck caused by uncontrolled pixel-level power consumption on high-resolution screens is resolved; and by constructing an embedded optical feedback channel in the screen through a reflected light monitoring unit, the brightness inaccuracy caused by backlight aging in traditional open-loop control is resolved. Attached Figure Description

[0032] Figure 1 This is a flowchart of the smartphone local dimming backlight method of the present invention. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without inventive effort are within the scope of protection of this invention.

[0034] This invention provides, for example Figure 1A method and system for local dimming backlighting in smartphones achieves breakthroughs in image quality by coordinating brightness and color depth, eradicating the persistent halo / color banding issues in HDR content; it features environmental intelligence, achieving zero-perception dimming through dual sensing of touch status and ambient light; it significantly improves energy efficiency by using color depth reduction calculations to address the energy efficiency pain points of high-resolution screens; and it enhances reliability by using embedded optical feedback in the screen to ensure long-term consistency.

[0035] Specifically, the method for local dimming backlight in smartphones includes the following steps:

[0036] S1. Screen partitioning: The backlight area of ​​the smartphone's display panel is divided into multiple independently controllable backlight dimming blocks. The size and shape of the partitions are dynamically adjusted or selected from preset rules based on the characteristics of the current display content, system power consumption strategy, or ambient light conditions. The partitions do not depend on the physical LED arrangement and the number is greater than the physical number of backlight LEDs to achieve a resolution that exceeds physical limitations.

[0037] When dynamically adjusting the size and shape of partitions: when a large area of ​​dark content is detected in the displayed content, adjacent blocks are automatically merged; when a high-contrast edge area is detected, the partition density is increased along the edge direction; and the minimum partition size is forced when the ambient light level is >10,000 lux.

[0038] Dynamic adjustment of partition size satisfies:

[0039]

[0040] in, The actual area of ​​the current block (unit: mm²); k is a normalization coefficient related to the screen size (0.1-0.3); The percentage of edge pixels within the block (0-1); The system power consumption strategy factor is 0.7-1.2, and >1 when the battery is low. ϵ represents the average brightness of the block (0-1000 nits); ϵ is the zero-resistance constant (fixed at 0.1). This is the luminance sensitivity index (default value 1.5).

[0041] S2. Content analysis and parameter acquisition: During the frame buffer rendering stage or after the compositing layer processing, analyze the brightness distribution and color distribution characteristics of the frame image to be displayed in real time; and simultaneously acquire at least one environmental parameter, which includes at least one of the following: ambient light intensity or color temperature sensed by the smartphone's front light sensor, phone posture sensor data, battery power status, and currently running application type (foreground application or specified background application).

[0042] Brightness distribution analysis includes: identifying regions of salient importance (ROI) in the image, calculating the brightness variance between ROI and non-ROI regions, and initiating color depth conversion prediction when the variance is >50%; color distribution analysis includes detecting regions with P3 color gamut coverage of over 90%.

[0043] Add motion state analysis: If the maximum motion vector detected in the image is greater than 5 pixels / frame, then color depth conversion is disabled for high-speed motion areas, and the original color depth output is retained;

[0044] S3. Target brightness mapping and color depth conversion prediction: Based on the local brightness characteristics and environmental parameters obtained from step S2, an independent target backlight brightness value is predicted and mapped for each backlight dimming block through a pre-stored lookup table or dynamic calculation model; at the same time, an appropriate color depth conversion mode target is predicted for the corresponding display panel area.

[0045] The dynamic calculation model is a lightweight convolutional neural network (CNN). The input layer receives the brightness histogram of the partition, the mean value of the HSV color space, and the ambient light color temperature. The output layer simultaneously generates the target backlight brightness and 8-bit / 10-bit color depth mode flags.

[0046] The lookup table (LUT) is updated through online learning: it records the error rate between the actual power consumption of each block and the target value, and triggers LUT parameter optimization based on gradient descent when the cumulative error is >10%.

[0047] It should be noted that the target backlight brightness value Calculated using the perception optimization model:

[0048]

[0049] in, The average brightness of pixels within block k. The maximum / minimum brightness values ​​within block k. The ambient light color temperature is α, β, and γ, which are weighting coefficients α+β+γ=1. This optimization model is a nonlinear sensing model that integrates the three factors of space, environment, and power consumption.

[0050] Furthermore, the target color depth mode decision threshold for:

[0051]

[0052] in, This represents the maximum brightness difference within the block. Ambient illuminance (unit: lux) This is the current screen refresh rate. It is a hyperbolic tangent (compressing the effects of extreme values). , For calibration coefficients;

[0053] S4. Brightness command generation and block control: The target backlight brightness value and its corresponding color depth conversion mode target of each backlight dimming block obtained in step S3 are combined with the current load and power consumption budget constraints of the smartphone processor to generate an optimized set of local backlight brightness control instructions. This set of instructions acts on the display driver chip and dedicated backlight control IC of the smartphone to drive multiple backlight dimming blocks after partitioning to achieve differentiated brightness output.

[0054] It should be noted that the objective of instruction optimization under energy budget constraints is to minimize a function, expressed as:

[0055]

[0056] in, The ideal brightness predicted by CNN; This is the backlight energy consumption function; Ω represents the absolute value of the color depth change in block k; Ω is the screen resolution factor (FHD=1, QHD=1.3). , , For dynamic weights ∝1 / Remaining battery power;

[0057] In addition, the energy consumption budget constraint algorithm is as follows:

[0058]

[0059] in, This represents the total power consumption of the backlight system. Target brightness of block z; The backlight efficiency coefficient is (0.8-1.1). This refers to the upper limit of the processor temperature. The current temperature; Processor load rate; Heat dissipation model coefficient;

[0060] S5. Color Depth Conversion Execution and Visual Compensation: Based on the color depth conversion mode target corresponding to each block predicted in step S3, the original frame image data is processed for color depth conversion in real time. The color depth conversion mode target aims to reduce the actual driving intensity of screen pixels to save power consumption. The conversion method includes, but is not limited to, adaptively mapping a portion of high bit color depth, such as 10-bit image data, to low bit color depth, such as 8-bit data. At the same time, a visual perception compensation signal based on the actual target backlight brightness of the block is superimposed after the color depth conversion to maintain or enhance the contrast perception and color saturation consistency of the converted image.

[0061] Visual compensation includes: applying spatial filtering based on human visual characteristics (HVS) to the color depth converted blocks to enhance high-frequency details; and superimposing a saturation gain coefficient that is positively correlated with the target backlight brightness of the block;

[0062] Saturation gain in visual compensation The compression intensity is adaptively adjusted according to color depth, specifically as follows:

[0063]

[0064] in, This is the amount by which the color depth is reduced; for example, if 10 bits is reduced to 8 bits, then Δbit = 2. This represents the target backlight brightness value for this block. This is the maximum backlight brightness of the system. Human eye sensitivity factor, ∈[0.1,0.3], increases in low-light conditions;

[0065] Optional, pixel compensation value after color depth conversion Derived from the color psychology model, the expression is:

[0066]

[0067] in, These are the original pixel values. The weight of hue (warm colors > cool colors). Hue sensitivity function =1.2, =0.8, BF is the block boundary strength. For boundary protection coefficient, ∈[0.5,2];

[0068] The contrast gain of visual compensation is:

[0069]

[0070] in, This is the contrast gain factor (1.0-1.8). Weighted by human eye sensitivity (0.5 for standard mode, 1.2 for HDR mode); The target backlight value; The original image brightness; The number of bits for the output color depth (either 8 or 10).

[0071] S6. Dynamic feedback adjustment: After steps S4 and S5 are executed, monitor the actual brightness output of each backlight dimming block, the actual display effect of the screen panel, and changes in user interaction status; use at least one ambient light sensor built into the smartphone to sample the reflected light of the actual display output brightness near the screen area, and combine it with the frequency of user touch operation or significant changes in content areas to establish closed-loop feedback; based on the deviation and trend of the feedback signal from the expected target value, dynamically adjust the brightness mapping strategy in step S3, the brightness command intensity in step S4, and the color depth conversion and compensation parameters in step S5 in the next frame processing cycle to achieve a more accurate, lower power consumption, and flicker-free local dimming effect;

[0072] The closed-loop feedback adopts a dual-threshold mechanism: when the actual brightness deviates from the target value by more than 15% and lasts for more than 3 frames, emergency brightness calibration is triggered; when the user touch operation frequency is more than 5 times / second, the partition adjustment is frozen.

[0073] Specifically, dynamic feedback adjustment amount Generated by dual-channel error fusion:

[0074]

[0075] in, The target brightness value for block k (unit: nits); The measured brightness of the sensor (unit: nits); The rate of change of spatial frequency response over the past 3 frames; θ represents the touch intensity indicator (touch = 1, no touch = 0); θ and μ are adaptive weights (θ² + μ² = 1).

[0076] Additionally, a power-sensitive mode is set: when the battery level is less than 20%, all blocks are forced to use the 8-bit color depth conversion mode, and the peak backlight brightness is limited to ≤300 nits;

[0077] As an optional feature of this embodiment, the color depth conversion trigger condition is:

[0078]

[0079] in, The block brightness variation coefficient (threshold β=0.25); This is the CIELAB chromaticity plane offset (threshold γ=0.05); The ambient light color temperature;

[0080] The above method features dynamic and flexible partitioning, emphasizing that the partitioning is software-defined and dynamically adjusted, exceeding the limitations of physical LED layout, such as having a higher matrix number than the actual number of LEDs, thus improving the freedom and accuracy of dimming. This differs from traditional fixed hardware partitioning.

[0081] Perception-driven intelligent decision-making closely integrates environmental parameters (light, posture, power, application status) with content analysis to jointly determine the target brightness and target color depth mode for each zone; target color depth mode conversion, such as bit reduction in some areas, is an explicitly introduced core power-saving method; collaborative optimization and execution, the brightness command and the corresponding color depth conversion mode command are collaboratively optimized, and processor and power consumption constraints are taken into account before being delivered to the display system for execution;

[0082] Color depth conversion and perceptual compensation are key differentiators from most local dimming solutions that only adjust backlight brightness. By reducing the original color depth (bit reduction) of the target display area under specific backlight brightness, the power consumption of the panel pixel drive is directly reduced. At the same time, visual compensation based on the final brightness of that area is superimposed, such as improving contrast / saturation, to ensure the user's visual experience, especially brightness and color perception, are not affected by bit reduction and may even be enhanced. This achieves dual optimization of backlight power consumption and panel pixel power consumption. Closed-loop feedback and adaptation are introduced, utilizing the smartphone's own sensors, the ambient light sensor, and real-time feedback from user interaction to adjust the brightness and color depth conversion strategy in a closed loop. This ensures stable performance and adapts to complex environments and usage changes, such as from a dark room to sunlight, avoiding flickering and sudden brightness changes.

[0083] In addition, this embodiment also proposes a local dimming backlight system for smartphones, which is used to implement the above-mentioned method, including:

[0084] A dynamic zone control module is used to divide the backlight area of ​​the display panel into multiple independently adjustable backlight dimming blocks, which include:

[0085] The embedded programmable partitioning engine dynamically adjusts the partition area and geometry based on real-time received display content feature data, system power consumption strategy instructions, and ambient light sensor signals; the hardware-accelerated boundary reconstruction unit generates irregular curve partition boundaries in response to high-contrast edge detection results.

[0086] A multi-source sensing processing module, directly connected to the image rendering pipeline and sensor bus, includes:

[0087] The frame buffer analyzer extracts the brightness / color distribution features of the frame image after the GPU frame synthesis stage; the environment fusion processor synchronously processes the multi-dimensional parameters output by the front light sensor, attitude sensor, and battery management IC; the multi-source perception processing module has a built-in high color gamut marker unit, which outputs a color depth protection latch signal to the collaborative decision-making module when an area with DCI-P3 coverage > 90% is detected.

[0088] The collaborative decision-making computation module includes:

[0089] Dual-output predictive processor: Based on brightness distribution characteristics and environmental parameters, it outputs independent target backlight brightness values ​​and adaptive color depth conversion mode flags for each block in parallel; Power consumption constraint arbiter: It receives the SoC load status and the energy consumption budget threshold from the temperature sensor in real time, and performs dynamic scaling on the target brightness value; Collaborative decision-making module integrates a lightweight neural network accelerator, whose input layer receives the block brightness variance, chromaticity statistics and discrete battery power values, and whose output layer synchronously generates brightness and bit depth control vectors;

[0090] The partition driver execution module includes:

[0091] Backlight instruction compilation unit: Encodes the target brightness value and color depth mode flag into a block control instruction set. The backlight instruction compilation unit forcibly activates the block merging instruction in low power mode and outputs a global bit depth compression enable signal to the compensation engine; Multi-channel driver IC array: Physically connected to the backlight LED matrix, responding to the instruction set to output differentiated backlight current; The partition drive execution module includes an edge mixing circuit, which automatically inserts a gradient transition waveform when the brightness difference between adjacent blocks is >100 nits. This circuit is triggered synchronously with the backlight scanning timing.

[0092] The color depth conversion and compensation engine, integrated into the display driver chip, includes:

[0093] Real-time bit depth converter: performs bit depth compression on the original pixel data according to the color depth mode flag bit; Visual perception compensation unit: injects adaptive contrast gain and color saturation correction signals into the compressed data; The color depth conversion and compensation engine includes a text area protection channel. When the pixel analyzer recognizes that the proportion of text is >30%, it bypasses the bit depth converter and directly passes the original data.

[0094] A closed-loop feedback control module, comprising:

[0095] Reflected light monitoring unit: An ambient light sensor array located on the screen bezel captures the actual output light signal of the panel; Cross-domain adaptive controller: Dynamically updates partitioning rules and compensation parameters based on brightness output deviation, user touch events, and content change signals; Closed-loop feedback control module deploys a dual-threshold safety mechanism: When the reflected light intensity deviation is continuously >15% for 4 frame cycles, the emergency calibration protocol of the backlight driver IC is activated; When the touch sampling rate is >8 times / second, partition updates are frozen.

[0096] In addition, the sensors in the reflected light monitoring unit are arranged in an array with a spacing of ≤2mm, and the lens is coated with a bandpass filter (wavelength 450-650nm) to suppress ambient light noise.

[0097] This system, through hardware collaboration, integrates dynamic partitioning and bit depth conversion algorithms into dedicated hardware modules, such as partitioning engines, bit depth converters, and reflected light monitoring units, to solve the problem of lack of closed-loop feedback in traditional local dimming.

[0098] Cross-level optimization, frame buffer analyzer is directly connected to GPU rendering pipeline to achieve zero-latency content analysis power constraint arbiter directly reads and writes PMIC registers to achieve nanosecond-level power consumption response;

[0099] Enhanced security mechanisms include a dual-threshold security module, brightness deviation and touch freeze protection to completely eliminate visible flickering text areas, and a channel protection feature to prevent critical information from becoming blurred due to bit depth compression.

[0100] Production feasibility: the multi-channel driver IC array is compatible with existing AMOLED / LCD production lines, and the filter-coated sensor is a mature optical device.

[0101] In summary, compared with traditional technologies, this invention solves the halo problem around fragmented bright objects with fixed physical partitions by dynamically reconstructing the backlight block boundaries through an embedded programmable partitioning engine; and solves the color banding defect caused by independent pixel driving when brightness drops suddenly by coupling and predicting the target brightness and color depth mode through a collaborative decision module.

[0102] By fusing real-time data streams of ambient light and touch behavior through a multi-source sensing module, visual fatigue caused by lag in response to strong light and touch flicker is resolved; by implementing block-adaptive color depth reduction calculations through a bit depth conversion engine, the battery life bottleneck caused by uncontrolled pixel-level power consumption on high-resolution screens is resolved; and by constructing an embedded optical feedback channel in the screen through a reflected light monitoring unit, the brightness inaccuracy caused by backlight aging in traditional open-loop control is resolved.

[0103] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for local dimming backlight in a smartphone, characterized in that, Includes the following steps: S1. Screen partitioning: The backlight area of ​​the smartphone's display panel is divided into multiple independently controllable backlight dimming blocks. The size and shape of the partitions are dynamically adjusted or selected from a set of preset rules based on the characteristics of the current display content, system power consumption strategy, or ambient light conditions. S2. Content analysis and parameter acquisition: Real-time analysis of the brightness and color distribution characteristics of the frame image to be displayed during the frame buffer rendering stage or after the compositing layer processing. S3. Target brightness mapping and color depth conversion prediction: Based on the local brightness characteristics and environmental parameters obtained from step S2, an independent target backlight brightness value is predicted and mapped for each backlight dimming block through a pre-stored lookup table or dynamic calculation model; at the same time, an appropriate color depth conversion mode target is predicted for the corresponding display panel area. S4. Brightness command generation and block control: The target backlight brightness value of each backlight dimming block obtained in step S3 and its corresponding color depth conversion mode target are combined with the current load and power consumption budget constraints of the smartphone processor to generate an optimized set of local backlight brightness control commands. S5. Color depth conversion execution and visual compensation: Based on the color depth conversion mode target corresponding to each block predicted in step S3, perform real-time color depth conversion processing on the original frame image data. S6. Dynamic feedback adjustment: After steps S4 and S5 are executed, monitor the actual brightness output of each backlight dimming block, the actual display effect of the screen panel, and changes in user interaction status; use at least one ambient light sensor built into the smartphone to sample the reflected light of the actual display output brightness near the screen area, and combine it with the frequency of user touch operation or significant changes in content areas to establish closed-loop feedback.

2. The method for local dimming backlight in a smartphone according to claim 1, characterized in that, In step S1, when dynamically adjusting the size and shape of the partitions, adjacent blocks are automatically merged when the displayed content contains a large area of ​​dark field; when a high-contrast edge area is detected, the partition density is increased along the edge direction; and the minimum partition size is forced when the ambient light illuminance is >10,000 lux.

3. The method for local dimming backlight in a smartphone according to claim 1, characterized in that, In step S3, the dynamic calculation model is a lightweight convolutional neural network (CNN). The input layer receives the brightness histogram of the partition block, the mean value of the HSV color space, and the ambient light color temperature. The output layer synchronously generates the target backlight brightness and the 8-bit / 10-bit color depth mode flag.

4. A method for local dimming backlight in a smartphone according to claim 1, characterized in that, In step S5, the visual compensation includes applying spatial filtering based on human visual characteristics to the color depth converted block to enhance high-frequency details; and superimposing a saturation gain coefficient that is positively correlated with the target backlight brightness of the block.

5. A method for local dimming backlight in a smartphone according to claim 1, characterized in that, In step S6, the closed-loop feedback adopts a dual-threshold mechanism: when the actual brightness deviates from the target value by more than 15% and lasts for more than 3 frames, emergency brightness calibration is triggered; when the user touch operation frequency is more than 5 times / second, the partition adjustment is frozen.

6. A method for local dimming backlight in a smartphone according to claim 3, characterized in that, Target backlight brightness value Calculated using the perception optimization model: in, The average brightness of pixels within block k. The maximum / minimum brightness values ​​within block k. The ambient light color temperature is α, β, and γ, which are weighting coefficients α+β+γ=1. This optimization model is a nonlinear sensing model that integrates the three factors of space, environment, and power consumption.

7. A method for local dimming backlight in a smartphone according to claim 4, characterized in that, Saturation gain in visual compensation The compression intensity is adaptively adjusted according to color depth, specifically as follows: in, This is the amount by which the color depth is reduced; for example, if 10 bits is reduced to 8 bits, then Δbit = 2. This represents the target backlight brightness value for this block. This is the maximum backlight brightness of the system. The human eye sensitivity factor, η∈[0.1,0.3], increases in low-light environments.

8. A local dimming backlight system for a smartphone according to claim 1, wherein the system is used to implement the method according to any one of claims 1-7, characterized in that, include: The dynamic zone control module is used to divide the backlight area of ​​the display panel into multiple independently adjustable backlight dimming blocks. The environmental fusion processor simultaneously processes multi-dimensional parameters output from the front-facing light sensor, attitude sensor, and battery management IC. The collaborative decision-making computing module includes: a dual-output prediction processor, which outputs independent target backlight brightness values ​​and adapted color depth conversion mode flags for each block in parallel based on brightness distribution characteristics and environmental parameters; and a power consumption constraint arbiter, which receives the SoC load status and the energy consumption budget threshold from the temperature sensor in real time and performs dynamic scaling on the target brightness values. The partition drive execution module includes: a backlight instruction compilation unit that encodes the target brightness value and color depth mode flag into a partition block control instruction set; and a multi-channel driver IC array that is physically connected to the backlight LED matrix and outputs differentiated backlight current in response to the instruction set. The color depth conversion and compensation engine, integrated into the display driver chip, includes: a real-time bit depth converter that performs bit depth compression on the original pixel data based on the color depth mode flag; and a visual perception compensation unit that injects adaptive contrast gain and color saturation correction signals into the compressed data. The closed-loop feedback control module includes: a reflected light monitoring unit, an ambient light sensor array located on the screen bezel, which captures the actual output light signal of the panel; and a cross-domain adaptive controller, which dynamically updates the partitioning rules and compensation parameters based on brightness output deviation, user touch events, and content change signals.

9. A local dimming backlight system for a smartphone according to claim 8, characterized in that, The dynamic partition control module includes a hardware-level partition block metadata storage, which stores a topology map dynamically generated by the programmable partition engine. This map switches to high resolution mode as the HDR content is activated. The multi-source perception processing module has a built-in high color gamut marker unit. When an area with DCI-P3 coverage > 90% is detected, it outputs a color depth protection latch signal to the collaborative decision-making module.

Images