Mini LED intelligent brightness adjusting system and adjusting method thereof

Abstract

The application discloses a Mini LED intelligent brightness adjusting system and an adjusting method thereof, and belongs to the technical field of display screens. The system comprises a data acquisition module, a temperature rise pre-judgment module, a partition brightness adjusting decision module, an execution control module and a dynamic threshold generation module connected in sequence. The data acquisition module is used for acquiring partition data and environmental parameters of each Mini LED lamp bead. The temperature rise pre-judgment module predicts the peak junction temperature of each partition through an LSTM prediction model. The dynamic threshold generation module dynamically adjusts the temperature warning threshold in combination with the use condition of the lamp bead and scene parameters. The partition brightness adjusting decision module determines a partition brightness adjusting strategy according to the real-time junction temperature, the predicted peak junction temperature and the warning threshold, and the adjusting strategy is executed by the execution control module. The application can realize fine temperature rise management and control at the partition level, control backlight temperature rise, prolong the service life of the lamp bead, reduce the loss of picture peak brightness and contrast, maximize the retention of picture quality and viewing experience, and solve the defects of the existing fixed threshold and global brightness reduction scheme.

Description

Technical Field

[0001] This application belongs to the field of display screen technology, specifically relating to a Mini LED intelligent brightness adjustment system and its adjustment method. Background Technology

[0002] With the development of display technology, Mini LED backlighting technology has become the mainstream backlighting solution for mid-to-high-end smart TVs due to its advantages such as high contrast, high peak brightness, long lifespan, and low power consumption. Mini LED backlit TVs use hundreds or thousands of independent backlight zones, enabling precise local dimming and significantly improving the dynamic range and picture quality. However, at the same time, the high density and compact arrangement of Mini LED chips cause the backlight module to generate a lot of heat in high-brightness, high-duty-cycle operating scenarios (such as HDR high-brightness images and full-screen white field displays), leading to a rapid increase in the junction temperature of the LED chips. Prolonged operation under overheated conditions can result in rapid luminous efficacy decay, color coordinate drift, and a significantly shortened lifespan. In severe cases, it can even lead to LED failure and backlight module damage. Furthermore, the temperature rise of the backlight module can also be conducted to core components such as the TV motherboard and driver chip, affecting the overall stability of the TV.

[0003] Existing technologies for thermal management and brightness adjustment solutions for Mini LED TVs have the following main drawbacks: The management and brightness adjustment are disconnected. Most of them adopt the "one-size-fits-all" protection scheme for overheating. When the overall backlight temperature is detected to be too high, the backlight brightness is directly reduced globally. Without combining the local dimming characteristics of Mini LED, it is impossible to achieve fine temperature rise control at the local level. As a result, the image quality performance of a large number of low-temperature zones is wasted, and the peak brightness and contrast of the picture are severely lost, which greatly affects the viewing experience.

[0004] The temperature rise threshold is set to a fixed value without being dynamically adjusted based on the lifespan of the LED chips and the decay of their luminous efficacy. In the case of new TVs, the threshold is too conservative, wasting the performance potential of the LED chips. After long-term use, the luminous efficacy of the LED chips decays and the temperature rise is higher at the same brightness, and the fixed threshold cannot provide effective protection.

[0005] Therefore, there is an urgent need to develop an intelligent brightness adjustment solution for Mini LED smart TVs that incorporates temperature rise feedback. Summary of the Invention

[0006] This application discloses a Mini LED intelligent brightness adjustment system and method, which effectively controls backlight temperature rise and extends the life of LED beads while maximizing the preservation of image quality and user viewing experience.

[0007] To address the aforementioned technical problems, this application provides a Mini LED intelligent brightness adjustment system, which employs the following technical solution: A Mini LED intelligent brightness adjustment system includes a data acquisition module, a temperature rise prediction module, a zone brightness adjustment decision module, and an execution control module connected in sequence. It also includes a dynamic threshold generation module connected to the temperature rise prediction module. The data acquisition module is used to collect parameter data during the operation of the Mini LED; The temperature rise prediction module is used to predict the peak junction temperature of each Mini LED chip zone based on the parameter data during Mini LED operation and using the built-in temperature prediction model. The zone brightness adjustment decision module is used to determine the appropriate zone brightness adjustment strategy based on the real-time junction temperature, peak junction temperature and temperature warning threshold of each Mini LED bead zone. The execution control module is used to execute the zone brightness adjustment strategy.

[0008] The dynamic threshold generation module is used to dynamically adjust the temperature warning threshold of each Mini LED bead zone.

[0009] Furthermore, the data acquisition module includes a zone temperature acquisition unit, an ambient light acquisition unit, and an ambient temperature acquisition unit, wherein: The zone temperature acquisition unit is used to acquire the real-time substrate temperature of each Mini LED chip zone, and calculate the actual junction temperature data of each Mini LED chip zone based on the preset substrate temperature-chip junction temperature conversion formula. The ambient light acquisition unit is used to acquire the real-time ambient light intensity of the environment being viewed by the device. The ambient temperature acquisition unit is used to collect the real-time ambient temperature when the device is playing music.

[0010] Furthermore, the data acquisition module also includes a screen content acquisition unit, an LED parameter acquisition unit, and a user behavior acquisition unit, wherein: The image content acquisition unit is used to analyze the input video signal frame by frame to obtain the average brightness of the image and the grayscale value of each Mini LED bead zone. The LED chip parameter acquisition unit is used to acquire Mini LED chip parameters, including rated junction temperature, rated luminous efficacy parameters, junction temperature-lifetime model, factory-calibrated brightness-temperature rise calibration curve, and cumulative usage time of the chip. The user behavior collection unit is used to collect user behavior preference data.

[0011] Furthermore, the Mini LED intelligent brightness adjustment system also includes a scene adaptation module, which is connected to the temperature rise prediction module and the zone brightness adjustment decision module. The scene adaptation module has multiple scene modes and an adjustment strategy library corresponding to each scene mode. The multiple scene modes include at least movie mode, game mode, standard mode and eye protection mode.

[0012] Furthermore, the dynamic threshold generation module is connected to the data acquisition module and the scene adaptation module. Specifically, the dynamic threshold generation module is used to dynamically adjust the temperature warning threshold of each Mini LED bead zone based on the rated junction temperature of the Mini LED bead, combined with the cumulative usage time of the bead, the luminous efficacy attenuation coefficient, and the current scene mode.

[0013] Furthermore, the temperature warning threshold includes a temperature rise warning threshold and an overheat protection threshold, wherein the overheat protection threshold is greater than the temperature rise warning threshold. The zone brightness adjustment decision module specifically includes: The temperature comparison unit is used to compare the real-time junction temperature and peak junction temperature of the partition with the temperature warning threshold, respectively. The first comparison result unit is used to execute the conventional partitioned light control strategy when both the real-time junction temperature and the peak junction temperature of the partition are lower than the temperature rise warning threshold. The second comparison result unit is used to execute an active and gradual pre-adjustment strategy when the real-time junction temperature or peak junction temperature of the partition reaches the temperature rise warning threshold and is lower than the overheat protection threshold. The third comparison result unit is used to execute a graded limiting adjustment strategy when the real-time junction temperature or peak junction temperature of a zone reaches the overheat protection threshold.

[0014] Furthermore, the Mini LED intelligent brightness adjustment system also includes an image quality processing module, which comprises an image quality smoothing module unit and an image quality compensation module unit, wherein: The image quality smoothing module unit is used to partition the Mini LED beads whose brightness needs to be adjusted. It adopts the brightness gradient transition algorithm between adjacent partitions to control the brightness difference between the adjusted partition and the adjacent partition within the preset brightness difference range. The image quality compensation module unit is used to dynamically adjust the color gamut, contrast, and gamma value of the Mini LED lamp bead partitions based on a preset color gamut-brightness compensation matrix.

[0015] Furthermore, the Mini LED intelligent brightness adjustment system also includes a closed-loop feedback and self-calibration module, which comprises a closed-loop feedback unit and a self-calibration unit, wherein: The closed-loop feedback unit is used to collect the junction temperature change and actual brightness output data of the Mini LED bead partition after brightness adjustment in real time, compare the predicted temperature rise and brightness, calculate the error value, correct the weight parameters of the temperature prediction model through the gradient descent algorithm, and correct the compensation coefficient of the brightness-temperature rise calibration curve. The self-calibration unit is used to automatically perform full-zone brightness-temperature rise calibration tests when the cumulative usage time of the LED beads reaches the calibration cycle or when the TV is in standby idle state. It sequentially inputs fixed brightness drive values ​​to each Mini LED bead zone, collects the corresponding temperature rise data and actual luminous efficacy data, updates the LED bead luminous efficacy attenuation coefficient, junction temperature-lifetime model parameters and brightness adjustment compensation parameters, and completes the system self-calibration.

[0016] To address the aforementioned technical problems, this application also provides a Mini LED intelligent brightness adjustment method, which employs the following technical solution: A Mini LED intelligent brightness adjustment method, applied to the Mini LED intelligent brightness adjustment system as described above, the Mini LED intelligent brightness adjustment method includes: After the TV is turned on, the system loads the Mini LED lamp bead rated parameters, junction temperature-lifetime model, factory-calibrated brightness-temperature rise calibration curve, cumulative lamp bead usage time and user historical settings parameters to complete system initialization and self-test. According to the set sampling frequency, the junction temperature data, ambient light brightness data, input video image content and ROI area data, and user's currently selected viewing mode data of each Mini LED bead zone are collected in real time. Using the collected data as input, the temperature prediction model predicts the peak junction temperature and temperature rise curve of each Mini LED bead zone. At the same time, combined with the cumulative usage time of the LED beads and the current scene mode, the temperature warning threshold of each Mini LED bead zone is dynamically generated. The real-time junction temperature of each zone, the predicted peak junction temperature, and the temperature warning threshold are matched to divide the corresponding temperature rise control levels. Combined with the viewing priority of the screen area and the adaptation rules of the current scene, the target brightness value and adjustment step size of each Mini LED bead zone are generated. Based on the target brightness value of each Mini LED bead zone, the brightness of adjacent zones is gradient smoothed, and the color gamut and contrast compensation parameters of the corresponding bead zone are generated. The backlight driver IC executes the PWM dimming command of each Mini LED bead, collects the junction temperature and actual brightness data of each Mini LED bead after adjustment in real time, calculates the error value, and corrects the compensation coefficient of the temperature prediction model and calibration curve. When the cumulative usage time of the LED beads reaches the calibration cycle or the TV is in standby idle state, perform a full-zone brightness-temperature rise calibration test, update the model parameters and calibration curves, and complete the system self-calibration.

[0017] Furthermore, using the collected data as input, a temperature prediction model is employed to predict the peak junction temperature and temperature rise curve of each Mini LED bead zone, specifically including: The collected data is preprocessed to obtain a standard input sequence. The preprocessing includes outlier detection and removal, missing data compensation, data smoothing filtering, and normalization. The standard input sequence is input into the temperature prediction model, and the historical temperature rise trend, brightness change trend, APL change rate, and temporal correlation features of ambient temperature change on the thermal behavior of the zone are extracted through the temperature prediction model. Based on the temporal correlation characteristics, the temperature prediction model is used to output the junction temperature prediction sequence of each Mini LED chip partition within a preset time period in the future, and to generate the future temperature rise prediction curve corresponding to each Mini LED chip partition. Peak search and comparison are performed on multiple junction temperature prediction values ​​in the future temperature rise prediction curve, and the highest junction temperature value within the preset time period is extracted as the predicted peak junction temperature of each Mini LED bead zone. According to the preset rolling update cycle, the collected data is refreshed by sliding the window, the earliest time period data is removed and the latest collected data is added, and the temperature prediction model inference is re-executed to update the future temperature rise curve and peak junction temperature of each Mini LED bead zone in real time.

[0018] Compared with the prior art, the embodiments of this application have the following main advantages: First, this solution combines the local dimming characteristics of Mini LED to achieve precise temperature rise control at the local level. It only adjusts the brightness of local areas that exceed the temperature limit, while maintaining the original output of local areas with low temperature and high brightness. This greatly reduces the loss of peak brightness and contrast, and maximizes the preservation of the original image quality and viewing experience.

[0019] Secondly, this solution uses a dynamic threshold generation module to dynamically adjust the temperature threshold based on the cumulative usage time of the LED beads, the luminous efficacy decay, and the scene priority. In the new device stage, it can fully release the performance potential of the LED beads, and after long-term use, it can also match the thermal characteristics of the decayed LED beads, achieving effective overheat protection and balancing performance and LED bead lifespan management.

[0020] Third, this solution uses an LSTM model to predict future temperature rise trends, allowing for early and gradual adjustments to avoid sudden brightness changes caused by rapid temperature increases. It also incorporates brightness smoothing and image quality compensation algorithms to further reduce the impact of brightness adjustments on the viewing experience. Furthermore, it can adapt to the long-term decay of LEDs through a closed-loop self-calibration mechanism, maintaining control accuracy and image quality even after long-term use. Attached Figure Description

[0021] To more clearly illustrate the solution of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This paper shows a system structure block diagram of a Mini LED intelligent brightness adjustment system according to an embodiment of the present application; Figure 2 It shows Figure 1 The structural block diagram of the data acquisition module 101 in the middle; Figure 3 This paper shows a system structure block diagram of another Mini LED intelligent brightness adjustment system according to an embodiment of the present application; Figure 4 It shows Figure 1 Or the structural block diagram of the partition brightness adjustment decision module 103 in 3; Figure 5 It shows Figure 3 Structural block diagram of the image quality processing module 107 in the image; Figure 6 It shows Figure 3 The structural block diagram of the closed-loop feedback and self-calibration module 108 in the middle; Figure 7 A flowchart of a Mini LED intelligent brightness adjustment method according to an embodiment of this application is shown. Detailed Implementation

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0024] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0025] Existing technologies for thermal management and brightness adjustment solutions for Mini LED TVs have the following main drawbacks: 1. The management and brightness adjustment are disconnected. Most of them adopt the "one-size-fits-all" protection solution for overheating. When the overall backlight temperature is detected to be too high, the backlight brightness is directly reduced globally. Without combining the local dimming characteristics of Mini LED, it is impossible to achieve fine temperature rise control at the local level. As a result, the image quality performance of a large number of low-temperature zones is wasted, and the peak brightness and contrast of the picture are severely lost, which greatly affects the viewing experience. 2. The passive overheat protection only performs derating operation after the temperature exceeds the standard, which has obvious adjustment lag. This causes large fluctuations in backlight temperature, which not only fails to effectively avoid lifespan damage caused by short-term overheating of LED beads, but also easily causes sudden increases and decreases in screen brightness, resulting in a very poor user experience. 3. The temperature rise threshold is set to a fixed value without being dynamically adjusted based on the lifespan of the LED beads and the decay of their luminous efficacy. The threshold is too conservative for new TVs, wasting the performance potential of the LED beads. After long-term use, the luminous efficacy of the LED beads decays and the temperature rise is higher at the same brightness. The fixed threshold cannot provide effective protection. 4. It does not adapt and adjust according to the content of the screen and the viewing scenario, and cannot distinguish between the core area and non-core area of ​​the screen. It also cannot match the corresponding temperature rise control and brightness adjustment strategies for different user needs in different scenarios such as movies, games, and eye protection, and cannot achieve a balance between image quality, lifespan, and experience. 5. Lack of a closed-loop self-calibration mechanism. As the luminous efficacy of the LEDs decays with long-term use, the correspondence between brightness and temperature rise shifts, and the accuracy of the original adjustment model drops significantly, making it impossible to guarantee the adjustment effect and protection capability throughout the entire life cycle of the TV.

[0026] Therefore, there is an urgent need to develop an intelligent brightness adjustment solution for Mini LED smart TVs that incorporates temperature rise feedback, in order to solve the aforementioned pain points of existing technologies, effectively control backlight temperature rise, extend the lifespan of LED chips, and maximize the preservation of picture quality and user viewing experience.

[0027] The purpose of this application is to address the shortcomings of existing technologies by providing a MiniLED intelligent brightness adjustment system and method for smart TVs based on temperature rise feedback. Through zone-level temperature rise acquisition and prediction, dynamic threshold generation, priority-based adjustment of picture, scene adaptive adaptation and closed-loop self-calibration, it achieves refined and coordinated control of MiniLED backlight temperature rise and brightness, solving the problems of lagging adjustment, one-size-fits-all brightness reduction, and inability to balance picture quality and lifespan in existing technologies.

[0028] Example 1: To achieve the above-mentioned objectives, this application adopts the following technical solution, please refer to [link / reference]. Figure 1 This application discloses a Mini LED intelligent brightness adjustment system 100, including a data acquisition module 101, a temperature rise prediction module 102, a zone brightness adjustment decision module 103, and an execution control module 104 connected in sequence. It also includes a dynamic threshold generation module 105, which is connected to the temperature rise prediction module 102. The data acquisition module 101 is used to acquire parameter data during the operation of the Mini LED; The temperature rise prediction module 102 is used to predict the peak junction temperature of each Mini LED chip zone based on the parameter data during Mini LED operation using the built-in temperature prediction model. The partition brightness adjustment decision module 103 is used to determine the appropriate partition brightness adjustment strategy based on the real-time junction temperature, peak junction temperature and temperature warning threshold of each Mini LED bead partition. The execution control module 104 is used to execute the zone brightness adjustment strategy.

[0029] The dynamic threshold generation module 105 is used to dynamically adjust the temperature warning threshold of each Mini LED bead zone.

[0030] Among them, the data acquisition module 101 serves as the data foundation of the system, responsible for collecting data from all dimensions, including temperature rise, ambient light, screen content, LED parameters, and user settings; the temperature rise prediction module 102 and the dynamic threshold generation module 105 work together to upgrade from passive protection to active prediction, identifying temperature rise risks in advance and generating appropriate safety thresholds; the zone brightness adjustment decision module 103 is the core of the system, combining temperature rise status, screen priority, and scene rules to generate refined adjustment strategies; and the execution control module 104 is responsible for the implementation of dimming commands.

[0031] Specifically, the parameter data for Mini LED operation includes the real-time junction temperature of each Mini LED chip zone, real-time ambient light intensity, real-time ambient temperature, average screen brightness, screen grayscale value corresponding to each Mini LED chip zone, Mini LED chip parameters (including rated junction temperature, rated luminous efficacy parameters, junction temperature-lifetime model, factory-calibrated brightness-temperature rise calibration curve, and cumulative chip usage time), and user behavior preference data. The data acquisition module 101 is further subdivided into multiple functional units. For example, the zone temperature acquisition unit acquires the real-time substrate temperature of each zone and obtains the actual junction temperature according to a preset conversion relationship. The ambient light acquisition unit and the ambient temperature acquisition unit acquire ambient light intensity and ambient temperature, respectively. The screen content acquisition unit analyzes the input video frame by frame to extract the average screen brightness and the screen grayscale value corresponding to each zone. The chip parameter acquisition unit acquires basic parameters such as rated junction temperature and rated luminous efficacy, as well as the cumulative usage time. The user behavior acquisition unit acquires user behavior preference data. To adapt to users' viewing needs in different scenarios, the system also has a scene adaptation module. The scene adaptation module is connected to the temperature rise prediction module and the zone brightness adjustment decision module. It has multiple scene modes such as movie, game, standard, and eye protection, as well as the corresponding adjustment strategy library for each mode. It can match the corresponding adjustment rules according to the mode currently selected by the user.

[0032] In one specific embodiment of this application, the temperature prediction model is an LSTM (Long Short-Term Memory) prediction model. The LSTM prediction model can extract temporal correlation features from the input time-series temperature, brightness, and screen data, learn the temperature rise change law of Mini LED partitions under different input conditions, and output a more accurate peak junction temperature prediction result. Compared with the traditional static temperature rise calculation method, it is more suitable for complex temperature rise change scenarios during dynamic screen playback.

[0033] The temperature prediction model takes the parameter data of the Mini LED operation collected by the data acquisition module 101 as input, and combines it with the scene mode selected by the user to predict the peak junction temperature of each Mini LED chip zone. Then, the zone brightness adjustment decision module 103 determines the appropriate zone brightness adjustment strategy based on the real-time junction temperature, peak junction temperature and temperature warning threshold of each Mini LED chip zone.

[0034] The dynamic threshold generation module 105 dynamically adjusts the temperature rise warning threshold and overheat protection threshold of each zone based on the rated junction temperature of the Mini LED beads, combined with the cumulative usage time of the beads, the luminous efficacy attenuation coefficient, and the current scene mode. The overheat protection threshold is higher than the temperature rise warning threshold.

[0035] The partition brightness adjustment decision module 103 is equipped with a temperature comparison unit, which compares the real-time junction temperature and peak junction temperature of each partition with the temperature warning threshold, corresponding to three different adjustment strategies: when both temperatures are below the temperature rise warning threshold, the conventional partition light control strategy is executed; when either temperature reaches the temperature rise warning threshold but does not exceed the overheat protection threshold, the active gradual pre-adjustment strategy is executed; when either temperature reaches the overheat protection threshold, the graded limiting adjustment strategy is executed to achieve graded control under different temperature rise risks.

[0036] The execution control module 104 receives the PWM dimming commands for each zone output by the zone brightness adjustment decision module 103, and transmits them to the Mini LED backlight driver IC via the SPI bus. The driver IC controls the LED drive current and PWM duty cycle of the corresponding zone to complete the zone brightness adjustment, and at the same time feeds back the drive execution results in real time to form a complete control closed loop.

[0037] Further, please see Figure 2 The data acquisition module 101 includes a zone temperature acquisition unit 110, an ambient light acquisition unit 120, and an ambient temperature acquisition unit 130, wherein: The partition temperature acquisition unit 110 is used to acquire the real-time substrate temperature of each Mini LED chip partition and calculate the actual junction temperature data of each Mini LED chip partition based on the preset substrate temperature-chip junction temperature conversion formula. The ambient light acquisition unit 120 is used to acquire the real-time ambient light brightness of the environment being viewed by the device. The ambient temperature acquisition unit 130 is used to acquire the real-time ambient temperature when the device is playing music.

[0038] The zone temperature acquisition unit 110 uses a sampling frequency of 10Hz to acquire the real-time substrate temperature of each zone through an NTC thermistor. Based on a preset substrate temperature-LED junction temperature conversion formula, it calculates the actual junction temperature data of the LEDs in each zone in real time. The ambient light acquisition unit 120 uses a sampling frequency of 5Hz to acquire the real-time ambient light brightness of the viewing environment through a photosensitive sensor, in lux. The ambient temperature acquisition unit 130 uses a sampling frequency of 1Hz to acquire the real-time ambient temperature around the TV through an ambient temperature sensor, in °C, providing basic input data for the temperature rise prediction model based on the environmental thermal impact.

[0039] Further, see also Figure 2 The data acquisition module 101 also includes a screen content acquisition unit 140, an LED parameter acquisition unit 150, and a user behavior acquisition unit 160, wherein: The image content acquisition unit 140 is used to analyze the input video signal frame by frame to obtain the average brightness of the image and the grayscale value of the image corresponding to each Mini LED bead zone. The LED parameter acquisition unit 150 is used to acquire Mini LED parameters, including rated junction temperature, rated luminous efficacy, junction temperature-lifetime model, factory-calibrated brightness-temperature rise calibration curve, and cumulative usage time of the LED. User behavior collection unit 160 is used to collect user behavior preference data.

[0040] The image content acquisition unit 140 analyzes the input HDMI / TV video signal frame by frame to obtain the average brightness (APL), the grayscale value of each zone, and HDR10 / Dolby Vision format parameters. At the same time, it uses a built-in lightweight AI image recognition model to locate the ROI (region of interest) of the image (face, central subject, dynamic high-brightness subject) and divide the viewing priority of the image area. The LED parameter acquisition unit 150 has a built-in Flash storage chip to store the rated junction temperature (125℃), rated luminous efficacy parameters, Arrhenius junction temperature-lifetime model, factory-calibrated brightness-temperature rise curve, and cumulative usage time statistics of the Mini LED LEDs. The user behavior acquisition unit 160 collects data in real time on the user's set viewing mode (movie / game / standard / eye protection), manual brightness setting, eye protection mode on / off, color temperature preference, etc.

[0041] This application sets up a data acquisition module 101 to collect basic data from multiple dimensions across the entire scenario. This provides sufficient input features for the LSTM temperature rise prediction model, ensuring prediction accuracy, and also provides data support for dynamic threshold adjustment and scenario adaptation. It is the foundation for the entire system to achieve refined management and control.

[0042] Further, please see Figure 3 The Mini LED intelligent brightness adjustment system also includes a scene adaptation module 106, which is connected to the temperature rise prediction module 102 and the zone brightness adjustment decision module 103 respectively. The scene adaptation module 106 has multiple scene modes and an adjustment strategy library corresponding to each scene mode. The multiple scene modes include at least movie mode, game mode, standard mode and eye protection mode.

[0043] Among them, the movie mode prioritizes image quality, with an adjustment step size of 3% / s. When the temperature rise exceeds the limit, the third and second priority areas are adjusted first. Only when the temperature still exceeds the limit after adjusting all non-core areas will the first priority area be fine-tuned to minimize the loss of peak brightness of the main screen. The game mode prioritizes low latency, increases the temperature sampling frequency to 20Hz, shortens the adjustment response cycle to 10ms, disables unnecessary image quality compensation calculations, sets the adjustment step size to an upper limit of 2% / s to avoid screen fluctuations, and performs peak brightness exemption for short-term high-brightness game screens, with an exemption duration of up to 5s. The standard mode is a balanced strategy, which uses a baseline threshold and baseline adjustment parameters to balance temperature rise control, image quality and overall power consumption. The eye protection mode prioritizes health, lowering the overheat protection threshold by 5°C. It also limits the global maximum brightness to no more than 400 nits in conjunction with ambient light control, thereby reducing flicker and blue light output while managing temperature rise.

[0044] Furthermore, the dynamic threshold generation module is specifically used to dynamically adjust the temperature rise warning threshold and overheat protection threshold of each Mini LED bead zone based on the rated junction temperature of the Mini LED bead, combined with the cumulative usage time of the bead, the luminous efficacy attenuation coefficient, and the current scene mode.

[0045] The temperature warning threshold includes a temperature rise warning threshold and an overheat protection threshold. The overheat protection threshold is higher than the temperature rise warning threshold, and the warning threshold is lower than the overheat protection threshold. For example, the overheat protection threshold does not exceed 85% of the rated junction temperature of the LED. Based on the rated junction temperature of 125℃, the baseline warning threshold is set to 90℃, and the baseline overheat protection threshold is set to 100℃ (80% of the rated junction temperature). When the cumulative usage time of the LED exceeds 10,000 hours, the overheat protection threshold is lowered by 3℃; when it exceeds 20,000 hours, the overheat protection threshold is lowered by 5℃. At the same time, the threshold is dynamically adjusted according to the current scene mode: 2℃ is increased in movie mode, 4℃ in game mode, and 5℃ in eye protection mode.

[0046] Further, please see Figure 4 The zone brightness adjustment decision module 103 specifically includes: Temperature comparison unit 131 is used to compare the real-time junction temperature of the partition with the peak junction temperature of the partition and the temperature warning threshold, respectively. The first comparison result unit 132 is used to execute a conventional partitioned light control strategy when both the real-time junction temperature and the peak junction temperature of the partition are lower than the temperature rise warning threshold. The second comparison result unit 133 is used to execute an active gradual pre-adjustment strategy when the real-time junction temperature or peak junction temperature of the partition reaches the temperature rise warning threshold and is lower than the overheat protection threshold. The third comparison result unit 134 is used to execute a graded limiting adjustment strategy when the real-time junction temperature or peak junction temperature of the partition reaches the overheat protection threshold.

[0047] Specifically, this application adopts a built-in three-level temperature rise control strategy, which combines the priority of screen areas to perform differentiated adjustments: Normal temperature rise range: The real-time junction temperature and the predicted peak junction temperature of each zone are both lower than the warning threshold. Based on the ambient light brightness and the content of the screen, regular zone dimming is performed without any brightness restrictions related to temperature rise, so as to fully release the peak brightness performance of the LEDs. Among them, regular zone dimming refers to controlling the backlight brightness of each zone individually according to the target brightness set by the user based on the input screen content and ambient brightness, so as to preserve the high brightness details and contrast of the original HDR screen and give full play to the performance advantages of Mini LED backlight. Warning temperature rise range: If the real-time junction temperature or predicted peak junction temperature of a zone reaches the warning threshold but is below the overheat protection threshold, active gradual pre-adjustment is performed. The brightness of the corresponding zones is reduced in sequence according to the priority of Level 3 > Level 2 > Level 1, with an adjustment step of 3% / s, until the predicted peak junction temperature falls back below the warning threshold. For example, if the junction temperature of a zone in the black border area of ​​the screen is predicted to reach 93℃ in 15 seconds, the brightness of that zone is reduced from 200nit to 150nit in advance to suppress the rising temperature trend and not affect the main screen viewing experience. Overheating temperature rise range: When the real-time junction temperature or predicted peak junction temperature of a zone reaches the overheating protection threshold, a graded limiting adjustment is performed. First, the brightness of the zone in the third priority area is reduced by a maximum of 50%. If the temperature continues to rise, the second priority area is gradually reduced. Only when the temperature still exceeds the threshold after the above adjustment is the brightness limited in the first priority area, with an adjustment step of 5% / s. At the same time, a peak brightness exemption strategy is implemented. When the corresponding image of a zone is identified as a short-duration high-brightness dynamic image (such as HDR explosion effect, lightning image) with a duration of ≤3s, even if the junction temperature reaches the overheating protection threshold, the brightness is not reduced, and the peak image quality effect is preserved. When the duration of the high-brightness image is >3s, the exemption state is exited, and a gradual derating is performed.

[0048] Further, see also Figure 3 and Figure 5 The Mini LED intelligent brightness adjustment system also includes an image quality processing module 107, which includes an image quality smoothing module unit 171 and an image quality compensation module unit 172, wherein: The image quality smoothing module unit 171 is used to partition the Mini LED beads whose brightness needs to be adjusted, and adopts the adjacent partition brightness gradient transition algorithm to control the brightness difference between the adjusted partition and the adjacent partition within the preset brightness difference range. The image quality compensation module unit 172 is used to dynamically adjust the color gamut, contrast and gamma value of the Mini LED lamp bead partitions based on the preset color gamut-brightness compensation matrix.

[0049] Specifically, the image smoothing module unit 171 uses an adjacent partition brightness gradient transition algorithm to control the brightness difference between the adjusted partition and the adjacent partition within 20% for the partition whose brightness needs to be adjusted, so as to avoid image banding and light spot problems caused by the brightness jump of the partition; the image quality compensation module unit 172 dynamically adjusts the image color gamut, contrast and gamma value of the partition whose brightness is reduced based on the preset color gamut-brightness compensation matrix, so as to make up for the loss of image layering caused by the brightness reduction and ensure the overall image viewing consistency.

[0050] The adjacent partition brightness gradient transition algorithm is a gradual interpolation algorithm based on the brightness difference between adjacent partitions. The algorithm adjusts the brightness value from the adjusted partition to the edge area according to the brightness difference between the currently adjusted partition and its surrounding adjacent partitions, resulting in a smooth transition effect for brightness changes. It avoids abrupt boundaries between bright and dark areas, significantly improving the overall visual quality of the adjusted image. The preset color gamut-brightness compensation matrix is ​​obtained through factory calibration, recording the corresponding color gamut compression ratio, contrast gain parameter, and gamma correction coefficient for different brightness reduction levels. The corresponding compensation parameters can be obtained directly by looking up the table, resulting in low computational load and no additional system load. It can quickly complete image quality compensation processing, ensuring natural color levels in the adjusted image and avoiding issues such as graying and loss of detail caused by localized brightness reduction.

[0051] Further, see also Figure 3 and Figure 6 The Mini LED intelligent brightness adjustment system also includes a closed-loop feedback and self-calibration module 108, which includes a closed-loop feedback unit 181 and a self-calibration unit 182, wherein: The closed-loop feedback unit 181 is used to collect the junction temperature change and actual brightness output data of the Mini LED bead partition after brightness adjustment in real time, perform predicted temperature rise comparison and brightness comparison, calculate the error value, correct the weight parameters of the LSTM prediction model through the gradient descent algorithm, and correct the compensation coefficient of the brightness-temperature rise calibration curve. The self-calibration unit 182 is used to automatically perform full-zone brightness-temperature rise calibration test when the cumulative usage time of the LED beads reaches the calibration cycle or when the TV is in standby idle state. It sequentially inputs fixed brightness drive values ​​to each Mini LED bead zone, collects the corresponding temperature rise data and actual luminous efficacy data, updates the LED bead luminous efficacy attenuation coefficient, junction temperature-lifetime model parameters and brightness adjustment compensation parameters, and completes the system self-calibration.

[0052] Specifically, the closed-loop feedback unit 181 collects the zone junction temperature changes and actual brightness output data in real time after brightness adjustment, compares the predicted temperature rise with the target brightness value, calculates the error value, and corrects the weight parameters of the LSTM prediction model and the compensation coefficient of the brightness-temperature rise calibration curve through the gradient descent algorithm, continuously improving the prediction and adjustment accuracy. The self-calibration unit 182 counts the cumulative usage time of the LED beads. Every 1000 hours of cumulative usage, or when the TV is in standby idle state, it automatically performs a full-zone brightness-temperature rise calibration test, sequentially inputs fixed brightness drive values ​​to each zone, collects the corresponding temperature rise data and actual luminous efficacy data, updates the LED bead luminous efficacy attenuation coefficient, junction temperature-lifetime model parameters, and brightness adjustment compensation parameters, completes the system self-calibration, and adapts to the luminous efficacy attenuation changes of LED beads after long-term use.

[0053] The Mini LED intelligent brightness adjustment system of this application has at least the following technical effects: 1. Achieve zone-level fine-grained temperature rise control. Combining the characteristics of Mini LED zone light control, perform differentiated brightness adjustment for zones with different temperature rise states. Only control zones with excessive temperature rise, maximize the preservation of image quality performance of low-temperature zones, and significantly reduce the loss of image brightness and contrast while effectively controlling temperature rise. 2. Adopting a temperature rise proactive prediction mechanism, the system uses a predictive model to identify the risk of temperature rise in different zones in advance, upgrading from passive overheat protection to proactive gradual pre-adjustment. This avoids large fluctuations in backlight temperature, effectively preventing lifespan damage caused by short-term overheating of LED beads and avoiding visual discomfort caused by sudden changes in screen brightness. 3. Achieve dynamic generation of temperature rise threshold. Based on the junction temperature-lifetime model of the LED beads, the threshold is dynamically adjusted by combining the cumulative usage time of the LED beads, the light efficiency decay, and the scene mode. This fully releases the performance of the LED beads in the new TV stage and strengthens the protection capability after long-term use, thus balancing the performance potential and lifespan of the LED beads. 4. Achieve adaptive adaptation by combining screen content and scene mode. Prioritize the adjustment of non-core area brightness by dividing the screen ROI area to minimize the loss of image quality of the main screen. At the same time, match exclusive adjustment strategies for different scenarios such as movies, games, and eye protection to meet the core needs of users in different scenarios and achieve a multi-dimensional balance of thermal management, image quality, and experience. 5. It has full closed-loop feedback and self-calibration capabilities. It continuously corrects the model parameters through feedback of actual data after adjustment, and performs full-zone calibration regularly to adapt to the light efficiency decay changes of the lamp beads throughout their entire life cycle. This ensures the adjustment accuracy and protection effect during long-term use of the TV, effectively extending the service life of the backlight module and the whole machine.

[0054] Example 2: To achieve the aforementioned objectives, this application further discloses an adjustment method applied to the aforementioned Mini LED intelligent brightness adjustment system. Please refer to [link to relevant documentation]. Figure 7 This application discloses a Mini LED intelligent brightness adjustment method, applied to the Mini LED intelligent brightness adjustment system described above. The Mini LED intelligent brightness adjustment method includes: S1, after the TV is turned on, loads the Mini LED lamp bead rated parameters, junction temperature-lifetime model, factory-calibrated brightness-temperature rise calibration curve, cumulative lamp bead usage time and user historical settings parameters, and completes system initialization and self-test; Specifically, the LED parameter acquisition unit of the data acquisition module reads all the basic parameters and historical statistics of LEDs stored in the Flash storage chip, and the user behavior acquisition unit synchronously loads the user's stored scene mode preferences, brightness and color temperature settings, etc., to complete the initial parameter configuration of the temperature prediction model, detect the working status of each acquisition unit and adjustment module, and enter the normal working process after confirming that there are no abnormalities.

[0055] S2, according to the set sampling frequency, collects in real time the junction temperature data, ambient light brightness data, input video image content and ROI area data, and the user's currently selected viewing mode data of each Mini LED bead zone; Specifically, the zone temperature acquisition unit polls and collects the temperature signals output by the NTC thermistors of all zones at a fixed frequency of 10Hz, converting them into real-time junction temperature values ​​for each zone. Simultaneously, the ambient light acquisition unit collects the ambient light brightness in front of the screen using an ambient light sensor. The image content acquisition unit analyzes the input video signal frame by frame, obtaining APL, zone grayscale values, and HDR format parameters. Lightweight AI is used to identify and divide ROI regions and determine image priorities. The scene adaptation module synchronously reads user behavior data such as the currently selected viewing mode and manual brightness settings. The system integrates all input data for the current period and sends it to the temperature rise prediction module and the dynamic threshold generation module.

[0056] S3 uses the collected data as input and predicts the peak junction temperature and temperature rise curve of each Mini LED bead zone through the temperature prediction model. At the same time, it dynamically generates the temperature warning threshold of each Mini LED bead zone by combining the cumulative usage time of the bead and the current scene mode. Specifically, in the temperature rise prediction module, the system inputs the real-time junction temperature of each partition, the junction temperature change trend of the past 10 cycles, and the grayscale target brightness of the partition corresponding to the input video frame into the temperature prediction model to predict the peak junction temperature and overall temperature rise trend of each partition within the next 15 seconds. At the same time, in the dynamic threshold generation module, the system uses the rated junction temperature of the LED as a benchmark, and first makes a preliminary adjustment to the benchmark threshold based on the luminous efficacy attenuation coefficient corresponding to the cumulative usage time of the LED. Then, it corrects the threshold in combination with the current scene mode, and finally outputs the temperature rise warning threshold and overheat protection threshold for each partition.

[0057] S4 matches the real-time junction temperature of the partition, the predicted peak junction temperature and the temperature warning threshold, divides the corresponding temperature rise control level, and generates the target brightness value and adjustment step size of each Mini LED bead partition by combining the viewing priority of the screen area and the adaptation rules of the current scene. Specifically, the partition brightness adjustment decision module makes decisions based on three control levels: normal temperature rise, warning temperature rise, and overheating temperature rise. When the temperature rise is within the normal range, the target brightness is generated directly based on the ambient light and the content of the screen without any additional brightness restrictions. When the temperature rise is within the warning range, the brightness of the corresponding partitions is adjusted sequentially according to the priority order of Level 3, Level 2, and Level 1, with a set step size, until the predicted peak junction temperature falls below the warning threshold. When the temperature rise is within the overheating range, the maximum brightness reduction is first applied to the Level 3 priority partitions. If the temperature still exceeds the limit, the brightness of the Level 2 priority partitions is then reduced. Only if the temperature still exceeds the threshold after the first two adjustments are completed is the brightness limit applied to the Level 1 priority partitions. At the same time, peak brightness exemption is applied to short-term high-brightness dynamic images. After the exemption ends, the adjustment is resumed.

[0058] S5, based on the target brightness value of each Mini LED bead zone, performs gradient smoothing on the brightness of adjacent zones, and simultaneously generates color gamut and contrast compensation parameters for the corresponding bead zone. Specifically, the image smoothing module first calculates the brightness difference between each zone that needs brightness adjustment and its adjacent zones. If the brightness difference exceeds a preset 20% threshold, a linear gradient transition is performed between the adjusted zone and its adjacent zones to control the brightness difference between adjacent zones within 20%, thus avoiding image banding or visible light spots due to sudden brightness changes. Then, the image compensation module retrieves a preset color gamut-brightness compensation matrix for the zones with reduced brightness. Based on the current target brightness value of the zone, it matches the corresponding color gamut compression ratio, contrast enhancement coefficient, and gamma correction value to adjust the color parameters of the corresponding image of the zone. This compensates for the loss of image transparency and layering caused by the reduction in brightness, ensuring a unified and harmonious overall image.

[0059] S6 executes the PWM dimming command of each Mini LED bead through the backlight driver IC, collects the junction temperature and actual brightness data of each Mini LED bead after adjustment in real time, calculates the error value, and corrects the compensation coefficients of the LSTM prediction model and calibration curve. Specifically, after receiving the target brightness command for each zone, the execution control module (backlight driver IC) outputs the drive current according to the corresponding PWM duty cycle to complete the brightness adjustment of each zone. After the adjustment is completed, the real-time junction temperature and actual brightness output data of the adjusted zone are read synchronously in the next data acquisition cycle. The actual junction temperature change is compared with the temperature rise result of the temperature prediction model, and the actual output brightness is compared with the target brightness. The temperature rise prediction error value and the brightness adjustment error value are calculated respectively. The error value is substituted into the gradient descent algorithm for iterative update to correct the hidden layer weight parameters of the temperature prediction model. At the same time, the compensation coefficient of the brightness-temperature rise calibration curve is updated to complete a closed-loop feedback correction, providing a more accurate parameter basis for the prediction and adjustment of the next cycle.

[0060] S7: When the cumulative usage time of the LED beads reaches the calibration cycle or the TV is in standby idle state, perform full-zone brightness-temperature rise calibration test, update model parameters and calibration curves, and complete system self-calibration.

[0061] Specifically, the closed-loop feedback and self-calibration module calculates the cumulative usage time of each LED bead. When the cumulative usage time reaches the preset 1000-hour calibration cycle, or when the TV is triggered into standby mode and remains idle for more than the preset time, the full-zone calibration process is automatically initiated. The system sequentially lights up each Mini LED bead zone individually, inputs a PWM drive signal with a fixed duty cycle to the zone, and after maintaining stable drive output, collects the current stable junction temperature and actual output brightness of the zone. It compares the data with the initial calibration reference data, calculates the actual luminous efficacy attenuation coefficient of the current LED bead, updates the junction temperature-lifetime model parameters of the corresponding zone, refits and generates the brightness-temperature rise calibration curve of the zone, and updates the compensation parameters in the brightness adjustment process. After all zones are calibrated, all updated parameters are stored in the non-volatile Flash chip, the system self-calibrates, exits the calibration process, and returns to standby mode.

[0062] The Mini LED intelligent brightness adjustment method of this application enables precise active temperature control of Mini LED backlight zones under different usage stages and viewing scenarios. This avoids damage to the LED chips due to overheating, extends the lifespan of the backlight module, and maximizes the preservation of peak brightness, contrast, and layering of the image. It balances heat dissipation protection, image quality performance, and user viewing experience. Furthermore, through a self-calibration mechanism throughout the entire life cycle, it can maintain a stable and accurate adjustment effect even after long-term use. This solves the problem of image quality loss caused by the one-size-fits-all brightness reduction in traditional unified dimming overheat protection solutions, and also makes up for the shortcomings of fixed threshold protection that cannot adapt to changes in LED chip decay.

[0063] Furthermore, using the collected data as input, a temperature prediction model is employed to predict the peak junction temperature and temperature rise curve of each Mini LED bead zone, specifically including: The collected data is preprocessed to obtain a standard input sequence. The preprocessing includes outlier detection and removal, missing data compensation, data smoothing filtering, and normalization. The standard input sequence is input into the temperature prediction model, and the historical temperature rise trend, brightness change trend, APL change rate, and temporal correlation features of ambient temperature change on the thermal behavior of the zone are extracted through the temperature prediction model. Based on the temporal correlation characteristics, the temperature prediction model is used to output the junction temperature prediction sequence of each Mini LED chip partition within a preset time period in the future, and to generate the future temperature rise prediction curve corresponding to each Mini LED chip partition. Peak search and comparison are performed on multiple junction temperature prediction values ​​in the future temperature rise prediction curve, and the highest junction temperature value within the preset time period is extracted as the predicted peak junction temperature of each Mini LED bead zone. According to the preset rolling update cycle, the collected data is refreshed by sliding the window, the earliest time period data is removed and the latest collected data is added, and the temperature prediction model inference is re-executed to update the future temperature rise curve and peak junction temperature of each Mini LED bead zone in real time.

[0064] In one specific embodiment of this application, the temperature prediction model is an LSTM prediction model. The prediction scheme implemented by the LSTM prediction model is designed with ensuring the timeliness of the prediction as its core, and it perfectly matches the actual changing patterns of the Mini LED display scenario: when actually playing content, the brightness and APL of the screen are constantly changing dynamically. If only a 15-second prediction is made based on a single 30-second data set, as time goes by, the screen content has changed, and the original prediction will gradually deviate from the actual junction temperature change, resulting in increasingly larger deviations. The mechanism of scrolling and refreshing the input window every 5 seconds ensures that the model input always retains the latest actual operating data, without consuming too much computing power due to excessively frequent refreshes, thus achieving a good balance between prediction accuracy and operating performance.

[0065] From data acquisition to output results, targeted optimizations have been made based on the characteristics of Mini LED partitions: First, in the selection of input features, the core variables that have the most direct impact on junction temperature changes have been included, namely partition drive brightness, screen APL changes and ambient temperature. The junction temperature of a Mini LED is essentially the result of balancing the heat generated by the LED chip during operation with its heat dissipation. The heat generated by the LED chip is directly correlated with the driving brightness; the higher the driving brightness, the more heat is generated per unit time, and the faster the junction temperature rises. Changes in the average brightness (APL) of the screen directly affect the overall heat generation of all zones on the screen. If the APL suddenly jumps from low brightness to high brightness, the heat dissipation environment of the entire screen will be changed, and the temperature rise rate of individual zones will also accelerate. This is the underlying logic that enables LSTM networks to learn the pattern of "sudden APL jump → impending sharp increase in temperature." The influence of ambient temperature is also significant. When the ambient temperature is already high, the temperature difference between the zone and the environment decreases, and the heat dissipation efficiency drops. Under the same driving brightness, the junction temperature of a zone will rise faster than in a low-temperature environment, and the final peak junction temperature will be higher. This pattern is also incorporated into the model inference process as a core feature, effectively improving the prediction accuracy under different usage scenarios.

[0066] Secondly, in terms of model selection, a pre-trained lightweight LSTM network was adopted, which meets the edge-side operation requirements of display devices: Traditional statistical prediction methods such as ARIMA have poor fitting effects on nonlinear junction temperature changes and are difficult to handle temperature rise prediction under the combined influence of multiple variables; while large-scale deep learning models cannot run on edge chips of consumer-grade display products such as TVs and monitors, which would lead to excessive power consumption and latency. Lightweight LSTM is very suitable for handling this type of time-series prediction problem. It can remember the changing trends of variables such as junction temperature and brightness over the past 30 seconds through a gating mechanism, and grasp the key patterns affecting the temperature rise trend. At the same time, because the model is small, only forward inference is needed after pre-training, and no training is required on the edge. The computational load is very small, which can meet the performance requirement of inference refresh every 5 seconds and will not put too much load on the main chip of the display device.

[0067] During actual data collection, due to factors such as the stability of the acquisition module and signal transmission interference, occasional outliers such as temperature jumps and missing brightness data frames may occur. If these outliers are not removed in advance, they will seriously interfere with the model's judgment of trends, causing the prediction results to deviate completely from reality. Min-max normalization, which scales all features to the [0,1] interval, solves the problem of excessively large differences in the numerical ranges of different features. The numerical range of temperature is usually between 20℃ and 100℃, while the numerical range of APL is 0 to 1. If normalization is not performed, the model will naturally be more biased towards the temperature feature with larger numerical values, ignoring the influence of small numerical features such as APL, resulting in a decrease in prediction accuracy. Time alignment ensures that the feature data corresponds to the actual state at each time point, preventing time misalignment and avoiding incorrect feature association.

[0068] The final output shows the temperature rise curve and peak junction temperature of each zone in the next 15 seconds. Based on the predicted peak junction temperature, the system can adjust the maximum brightness of each zone in advance, avoiding the reduction of LED lifespan due to the junction temperature exceeding the threshold. It also avoids the lag of traditional overheat protection schemes that only activate when the junction temperature exceeds the threshold. This truly achieves proactive control of the junction temperature. At the same time, because the rolling refresh mechanism always keeps the predicted results and the current actual operating status in line, there will be no problem of excessive advance adjustment leading to unnecessary brightness reduction, thus balancing display performance and device reliability.

[0069] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application 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 specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. A Mini LED intelligent brightness adjustment system, characterized in that, The system includes a data acquisition module, a temperature rise prediction module, a zone brightness adjustment decision module, and an execution control module connected in sequence. It also includes a dynamic threshold generation module connected to the temperature rise prediction module. The data acquisition module is used to collect parameter data during the operation of the Mini LED; The temperature rise prediction module is used to predict the peak junction temperature of each Mini LED chip zone based on the parameter data of the Mini LED during operation and using a built-in temperature prediction model. The partition brightness adjustment decision module is used to determine an appropriate partition brightness adjustment strategy based on the real-time junction temperature, peak junction temperature and temperature warning threshold of each Mini LED bead partition. The execution control module is used to execute the partition brightness adjustment strategy; The dynamic threshold generation module is used to dynamically adjust the temperature warning threshold of each Mini LED bead zone.

2. The Mini LED intelligent brightness adjustment system as described in claim 1, characterized in that, The data acquisition module includes a zoned temperature acquisition unit, an ambient light acquisition unit, and an ambient temperature acquisition unit, wherein: The partition temperature acquisition unit is used to acquire the real-time substrate temperature of each Mini LED bead partition, and calculate the actual junction temperature data of each Mini LED bead partition based on the preset substrate temperature-bead junction temperature conversion formula. The ambient light acquisition unit is used to acquire the real-time ambient light brightness of the environment being viewed by the device; The ambient temperature acquisition unit is used to acquire the real-time ambient temperature when the device is playing music.

3. The Mini LED intelligent brightness adjustment system as described in claim 2, characterized in that, The data acquisition module further includes a screen content acquisition unit, an LED parameter acquisition unit, and a user behavior acquisition unit, wherein: The image content acquisition unit is used to analyze the input video signal frame by frame to obtain the average brightness of the image and the grayscale value of the image corresponding to each Mini LED bead partition. The LED bead parameter acquisition unit is used to acquire Mini LED bead parameters, including rated junction temperature, rated luminous efficacy parameters, junction temperature-lifetime model, factory-calibrated brightness-temperature rise calibration curve, and cumulative usage time of the LED bead. The user behavior collection unit is used to collect user behavior preference data.

4. The Mini LED intelligent brightness adjustment system as described in claim 3, characterized in that, The Mini LED intelligent brightness adjustment system also includes a scene adaptation module, which is connected to the temperature rise prediction module and the zone brightness adjustment decision module respectively. The scene adaptation module has multiple scene modes and an adjustment strategy library corresponding to each scene mode. The multiple scene modes include at least movie mode, game mode, standard mode and eye protection mode.

5. The Mini LED intelligent brightness adjustment system as described in claim 4, characterized in that, The dynamic threshold generation module is connected to the data acquisition module and the scene adaptation module. Specifically, the dynamic threshold generation module is used to dynamically adjust the temperature warning threshold of each Mini LED bead zone based on the rated junction temperature of the Mini LED bead, combined with the cumulative usage time of the bead, the luminous efficacy attenuation coefficient, and the current scene mode.

6. The Mini LED intelligent brightness adjustment system as described in claim 1, characterized in that, The temperature warning threshold includes a temperature rise warning threshold and an overheat protection threshold, wherein the overheat protection threshold is greater than the temperature rise warning threshold, and the zone brightness adjustment decision module specifically includes: A temperature comparison unit is used to compare the real-time junction temperature of the partition and the peak junction temperature of the partition with the temperature warning threshold, respectively. The first comparison result unit is used to execute a conventional partition light control strategy when the real-time junction temperature and peak junction temperature of the partition are both lower than the temperature rise warning threshold. The second comparison result unit is used to execute an active gradual pre-adjustment strategy when the real-time junction temperature or the peak junction temperature of the partition reaches the temperature rise warning threshold and is lower than the overheat protection threshold. The third comparison result unit is used to execute a graded limiting adjustment strategy when the real-time junction temperature or peak junction temperature of the partition reaches the overheat protection threshold.

7. The Mini LED intelligent brightness adjustment system as described in claim 1, characterized in that, The Mini LED intelligent brightness adjustment system also includes an image quality processing module, which comprises an image quality smoothing unit and an image quality compensation unit, wherein: The image quality smoothing module unit is used to partition the Mini LED beads whose brightness needs to be adjusted, and adopt the adjacent partition brightness gradient transition algorithm to control the brightness difference between the adjusted partition and the adjacent partition within a preset brightness difference range. The image quality compensation module unit is used to dynamically adjust the color gamut, contrast, and gamma value of the Mini LED bead partitions based on a preset color gamut-brightness compensation matrix for the partitions with reduced brightness.

8. The Mini LED intelligent brightness adjustment system as described in claim 7, characterized in that, The Mini LED intelligent brightness adjustment system also includes a closed-loop feedback and self-calibration module, which comprises a closed-loop feedback unit and a self-calibration unit, wherein: The closed-loop feedback unit is used to collect the junction temperature change and actual brightness output data of the Mini LED bead partition after brightness adjustment in real time, compare the predicted temperature rise and brightness, calculate the error value, correct the weight parameters of the temperature prediction model and the compensation coefficient of the brightness-temperature rise calibration curve through the gradient descent algorithm. The self-calibration unit is used to automatically perform full-zone brightness-temperature rise calibration test when the cumulative usage time of the LED beads reaches the calibration cycle or when the TV is in standby idle state. It sequentially inputs fixed brightness drive values ​​to each Mini LED bead zone, collects corresponding temperature rise data and actual luminous efficacy data, updates the LED bead luminous efficacy attenuation coefficient, junction temperature-lifetime model parameters and brightness adjustment compensation parameters, and completes system self-calibration.

9. A method for intelligent brightness adjustment of Mini LED, characterized in that, The Mini LED intelligent brightness adjustment method, applied to the Mini LED intelligent brightness adjustment system as described in any one of claims 1-8, comprises: After the TV is turned on, the system loads the Mini LED lamp bead rated parameters, junction temperature-lifetime model, factory-calibrated brightness-temperature rise calibration curve, cumulative lamp bead usage time and user historical settings parameters to complete system initialization and self-test. According to the set sampling frequency, the junction temperature data, ambient light brightness data, input video image content and ROI area data, and user's currently selected viewing mode data of each Mini LED bead zone are collected in real time. Using the collected data as input, the temperature prediction model predicts the peak junction temperature and temperature rise curve of each Mini LED bead zone. At the same time, combined with the cumulative usage time of the LED beads and the current scene mode, the temperature warning threshold of each Mini LED bead zone is dynamically generated. The real-time junction temperature and predicted peak junction temperature of each zone are matched with the temperature warning threshold to divide the corresponding temperature rise control level. Combined with the viewing priority of the screen area and the adaptation rules of the current scene, the target brightness value and adjustment step size of each Mini LED bead zone are generated. Based on the target brightness value of each Mini LED bead zone, the brightness of adjacent zones is gradient smoothed, and color gamut and contrast compensation parameters of the corresponding bead zone are generated. The backlight driver IC executes the PWM dimming command of each Mini LED bead, collects the junction temperature and actual brightness data of each Mini LED bead after adjustment in real time, calculates the error value, and corrects the compensation coefficient of the temperature prediction model and calibration curve. When the cumulative usage time of the LED beads reaches the calibration cycle or the TV is in standby idle state, perform a full-zone brightness-temperature rise calibration test, update the model parameters and calibration curves, and complete the system self-calibration.

10. The Mini LED intelligent brightness adjustment method as described in claim 9, characterized in that, The step of using the collected data as input to predict the peak junction temperature and temperature rise curve of each Mini LED bead zone through a temperature prediction model specifically includes: The collected data is preprocessed to obtain a standard input sequence. The preprocessing includes outlier detection and removal, missing data compensation, data smoothing filtering, and normalization. The standard input sequence is input into the temperature prediction model, and the model is used to extract the temporal correlation features of historical temperature rise trend, brightness change trend, APL change rate, and ambient temperature change on the thermal behavior of the zone. Based on the time-series correlation features, the temperature prediction model is used to output the junction temperature prediction sequence corresponding to each Mini LED bead partition within a preset time period in the future, and to generate the future temperature rise prediction curve corresponding to each Mini LED bead partition. Peak search and comparison are performed on multiple junction temperature prediction values ​​in the future temperature rise prediction curve, and the highest junction temperature value within a preset time period is extracted as the predicted peak junction temperature of each Mini LED bead partition. According to the preset rolling update cycle, the collected data is refreshed by sliding the window, the earliest time period data is removed and the latest collected data is added, and the temperature prediction model inference is re-executed to update the future temperature rise curve and peak junction temperature of each Mini LED bead zone in real time.

Images