A data intelligent adjustment method of an adaptive screen of a foldable mobile phone

Abstract

The application relates to the technical field of foldable mobile phone screen adjustment, and discloses a data intelligent adjustment method for the adaptive screen of a foldable mobile phone. The method obtains monitoring data of multiple sensors of the foldable mobile phone within a preset time period, including deformation data of a screen bending area, ambient light intensity data and user holding pressure distribution data. After comparative analysis, real-time deformation deviation of each area of the screen is calculated, each partition light adaptation coefficient is determined, pressure distribution characteristics are generated, and a health degree evaluation value is obtained by comprehensively considering the two. Based on this, the monitoring data sample state is classified, and the screen display parameters are dynamically adjusted. The method can accurately match the screen bending state, ambient light and user holding habit, effectively improve the screen display effect, meet the diversified needs of users, prolong the service life of the screen, and bring a better experience to users of the foldable mobile phone.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of foldable mobile phone screen adjustment, in particular to a data intelligent adjustment method for the adaptive screen of a foldable mobile phone. BACKGROUND

[0002] With the rapid development of technology, foldable mobile phones, as a new type of mobile terminal device, are gradually entering people's lives. Its unique foldable design brings users a more diversified use experience. It can be carried conveniently as a normal mobile phone in the folded state, and it can provide more screen space when unfolded to meet the needs of more scenarios such as watching videos and processing documents.

[0003] In the actual use process of foldable mobile phones, the adaptive adjustment of the screen has always been a key factor restricting the improvement of user experience. At present, the screen adjustment function of most foldable mobile phones is relatively simple. For the deformation of the screen bending area, there is often a lack of continuous and accurate monitoring and corresponding adjustment mechanism. During the screen bending process, the display parameters of each area of the screen cannot be dynamically and reasonably adjusted according to the real-time deformation, which may cause abnormal screen display effects such as picture distortion and color deviation, greatly affecting the user's visual experience.

[0004] In terms of ambient light intensity, although existing foldable mobile phones have certain brightness adjustment functions, they usually only make simple linear adjustments based on the single data of the ambient light sensor. When the ambient light intensity fluctuates frequently in a short period of time, the screen brightness cannot be adjusted in time and accurately, resulting in over-brightness or over-darkness, which causes trouble to the user. In the outdoor environment with strong sunlight, the insufficient screen brightness makes it difficult to see the content; while in the indoor environment with dim light, the over-bright screen is dazzling, affecting the use comfort.

[0005] As for the user's holding pressure distribution, the current mobile phones basically do not take it into account in screen adjustment. When using foldable mobile phones, users have various holding methods, and different holding pressure distributions may affect the display requirements of the screen. When holding with one hand, the user may pay more attention to the display clarity of the local area of the screen; when holding with both hands, the requirements for the display effect of the whole screen are different. However, the existing screen adjustment method does not distinguish and optimize it accordingly, and cannot meet the diversified use requirements of users.

[0006] Due to these deficiencies in the screen adjustment of existing foldable mobile phones, a more intelligent and comprehensive data intelligent adjustment method for the adaptive screen is urgently needed to improve the screen display effect and user experience of foldable mobile phones. SUMMARY

[0007] The present application aims to provide a data intelligent adjustment method for adaptive screen of foldable mobile phone to solve the problems in the background art.

[0008] To achieve the above-mentioned purpose, the present application provides a data intelligent adjustment method for adaptive screen of foldable mobile phone, which comprises:

[0009] Obtaining monitoring data of multiple sensors of the foldable mobile phone within a preset time period, the monitoring data including deformation data of the screen bending area, ambient light intensity data and user holding pressure distribution data;

[0010] By comparing the change trend of the deformation data of the screen bending area in each preset time period with the historical deformation data, the real-time deformation deviation of each area of the screen is calculated;

[0011] Analyzing the fluctuation range of the current ambient light intensity data within the preset time period and the real-time deformation deviation to determine the light adaptation coefficient of each partition of the screen;

[0012] By comparing the distribution difference of the user holding pressure distribution data in the symmetric area of the screen and the pressure gradient change between adjacent partitions, the pressure distribution characteristics of each partition of the screen are generated;

[0013] Comprehensive analysis of the light adaptation coefficient and the pressure distribution characteristics, the health degree evaluation value of the screen under the current use state is calculated;

[0014] Based on the health degree evaluation value, all monitoring data samples are classified, and the screen display parameters are dynamically adjusted through the classification results.

[0015] Preferably, the process of calculating the real-time deformation deviation of each area of the screen comprises:

[0016] Arranging the deformation data of the screen bending area in time sequence to generate a deformation curve;

[0017] Calculating the curvature accumulation of the deformation curve in the complete time interval;

[0018] Obtaining the average value of the curvature accumulation of the historical deformation curve;

[0019] The real-time deformation deviation is represented by the difference between the curvature accumulation of the current deformation curve and the average value of the curvature accumulation.

[0020] Preferably, the process of determining the light adaptation coefficient of each partition of the screen comprises:

[0021] Calculating the range value of the ambient light intensity data within the preset time period;

[0022] The light adaptation coefficient is positively correlated with the range value and the real-time deformation deviation, respectively;

[0023] When the fluctuation range of ambient light intensity data exceeds a preset threshold, the weighting factor of the light adaptation coefficient is increased.

[0024] Preferably, the process of generating the pressure distribution characteristics of each partition of the screen includes:

[0025] Calculate the similarity of pressure distribution in symmetrical regions of the screen;

[0026] Obtain the rate of change of pressure gradient between adjacent partitions;

[0027] The pressure distribution similarity and pressure gradient change rate are normalized and fused.

[0028] When the rate of change of the pressure gradient exceeds a critical value, a mechanism for reconstructing the pressure distribution characteristics is triggered.

[0029] Preferably, the process of calculating the health assessment value of the screen in its current usage state includes:

[0030] The light adaptation coefficient and pressure distribution characteristics are weighted and fused together;

[0031] Calculate the cumulative value of the fusion result for all partitions of the screen;

[0032] When the screen is fully unfolded, reduce the weight of the curved area partition;

[0033] When the screen is bent, increase the weight of the bent area partition.

[0034] Preferably, the process of classifying the status of all monitoring data samples based on the health assessment value includes:

[0035] Establish dynamic classification thresholds based on health assessment values;

[0036] Density clustering algorithm is used to classify the state of monitoring data samples;

[0037] The screen status is divided into three categories: high-risk status, warning status, and safe status.

[0038] When a sample is classified as high-risk, the screen saver mechanism is automatically triggered.

[0039] Preferably, the dynamic adjustment of screen display parameters based on classification results includes:

[0040] Obtain user's historical operation habits data;

[0041] Establish a linkage mechanism between the screen brightness adjustment model and the color temperature compensation model;

[0042] Input health assessment values, ambient light intensity data, and user historical operation habit data into the linkage model;

[0043] Output the brightness correction parameters and color temperature compensation parameters for each zone of the screen.

[0044] Preferably, the method further includes a phased adjustment process:

[0045] When the screen changes from a bent state to an unfolded state, the adjustment process is divided into an initial transition phase and a steady-state adaptation phase.

[0046] In the initial transition phase, linear interpolation is used to quickly approximate the target display parameters;

[0047] During the steady-state adaptation phase, the exponential decay method is used for fine-tuning compensation.

[0048] The timing of switching between the two stages is dynamically adjusted based on the rate of change of ambient light intensity.

[0049] Preferably, the method further includes a user habit learning mechanism:

[0050] Record the final adjustment parameters under different environmental scenarios;

[0051] Analyze the frequency and magnitude of user-manually overriding automatic adjustment parameters;

[0052] When the frequency of manual overwriting exceeds the preset number of times, update the user's historical operation habit data;

[0053] The updated user history operation data is fed back into the screen brightness adjustment model.

[0054] Preferably, the method further includes an exception handling process:

[0055] Continuously monitor the deviation between the actual display parameters and the target parameters of each screen partition;

[0056] When the deviation value continues to exceed the allowable range, the sensor data verification mechanism is activated;

[0057] If the sensor data verification passes, the health assessment value is recalculated;

[0058] If the sensor data verification fails, switch to the backup adjustment strategy and generate a fault log.

[0059] Compared with the prior art, the beneficial effects of the present invention are:

[0060] In handling screen bending deformation, the system continuously acquires deformation data from multiple sensors over a preset time period and compares this data with historical deformation trends to calculate the real-time deformation deviation of each area. This process accurately captures subtle changes during screen bending. Based on the real-time deformation deviation, the system can finely adjust the display parameters of each area of ​​the screen. When a large deformation is detected in a certain area of ​​the screen, the system specifically optimizes parameters such as pixel arrangement and color display in that area to avoid image distortion and color falsification caused by deformation, ensuring a clear and accurate image for the user and greatly improving display quality under bending conditions.

[0061] To address changes in ambient light intensity, this method analyzes the fluctuation range of current ambient light intensity data within a preset time period and, combined with real-time screen deformation deviation, determines the light adaptation coefficient for each screen zone. In bright outdoor environments, the system automatically increases screen brightness based on the light adaptation coefficient while optimizing color contrast to ensure clear and readable content. In low-light indoor environments, it reduces screen brightness and adjusts color saturation to avoid eye strain from an overly bright screen, allowing users to comfortably use their phones under different lighting conditions and improving the screen's adaptability to ambient light.

[0062] Considering the distribution of user grip pressure, the system generates pressure distribution characteristics for each screen partition by comparing the differences in user grip pressure distribution data across symmetrical areas of the screen and the changes in pressure gradient between adjacent partitions. Based on these characteristics, the system can understand the user's grip habits and focus areas. When it detects that the user is holding the screen with one hand and the pressure is concentrated on one side, it automatically enhances the display clarity and color vibrancy of that side of the screen to highlight the user's focus. When holding the screen with both hands, it optimizes the overall display effect, providing a balanced and comfortable visual experience to meet the screen display needs of users in diverse grip scenarios.

[0063] By combining the light adaptation coefficient and pressure distribution characteristics, the system calculates the screen's health assessment value under its current usage state. Based on this assessment, all monitoring data samples are categorized, and the screen display parameters are dynamically adjusted. This comprehensive and intelligent adjustment method can match user scenarios and needs in real time and accurately, effectively improving screen display performance and extending screen lifespan. It reduces screen aging and damage caused by improper adjustments, lowers user costs, and provides users with a more convenient, comfortable, and personalized foldable phone experience, thus promoting the further development and widespread adoption of foldable phone technology. Attached Figure Description

[0064] Figure 1 This is a schematic diagram illustrating the working principle of the adaptive screen intelligent adjustment method for foldable mobile phones according to the present invention.

[0065] Figure 2 A flowchart for calculating real-time deformation deviations in different areas of the screen;

[0066] Figure 3 A flowchart generated to show the pressure distribution characteristics of each partition on the screen;

[0067] Figure 4 A flowchart for classifying the status of monitoring data samples based on health assessment values. Detailed Implementation

[0068] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0069] Please see Figure 1 This invention provides a data intelligent adjustment method for the adaptive screen of a foldable mobile phone, the method comprising:

[0070] The system collects monitoring data over a preset time period using an array of sensors integrated at multiple locations on the device. This data includes: deformation data of the screen's bending area acquired by a micro-strain sensor with a sampling frequency of 100Hz, capable of capturing microscopic deformation characteristics during the folding process; ambient light intensity data acquired by an ambient light sensor supporting a dynamic range of 10,000-100,000 lux; and user grip pressure distribution data acquired by a piezoelectric pressure sensor array containing 64 sensing units, capturing pressure distribution changes at a sampling rate of 120 times per second.

[0071] The deformation analysis module arranges the collected deformation data in a time series and calculates the real-time deformation deviation value of each area of ​​the screen by comparing the deformation characteristics of the current time period with those of the same historical period. The ambient light processing module analyzes the fluctuation characteristics of light intensity data within a time period and, combined with the real-time deformation deviation data, uses a weighted statistical algorithm to determine the light adaptation coefficient of each partition of the screen. The stress analysis module identifies the grip pattern through machine learning algorithms, compares the pressure distribution differences in symmetrical areas of the screen, and calculates the pressure gradient changes between adjacent partitions to generate corresponding pressure distribution feature vectors. The light adaptation coefficient and stress distribution features are input into the evaluation model, and a dynamic weight allocation algorithm is used to calculate the health assessment value of the screen under its current usage state. This assessment value comprehensively reflects the screen's physical state, environmental adaptability, and usage load.

[0072] The system categorizes monitored data samples based on health assessment results, employing an unsupervised clustering algorithm to classify screen states into different risk levels. Based on the classification results, the system dynamically adjusts display parameters for each screen zone, including brightness, color temperature, and refresh rate, and continuously optimizes the adjustment effect through a closed-loop feedback mechanism. The entire process runs in real-time within the embedded system, with the complete cycle from data acquisition to parameter adjustment controlled within 200 milliseconds, ensuring a smooth user experience.

[0073] Example 1: See Figure 2 This paper describes the processing and analysis of deformation data in the bending area of ​​a foldable phone screen. A high-precision strain sensor array is embedded in the bending area inside the phone. These sensors continuously collect microscopic deformation data of the screen at a fixed sampling frequency. When the user folds or unfolds the phone, the sensors capture the stress changes experienced by the screen material in real time, forming a raw deformation data stream. This data stream undergoes noise reduction and filtering by a preprocessing module to eliminate interference signals caused by temperature fluctuations or instantaneous impacts, retaining valid data points reflecting the true deformation state. The processed data is arranged in timestamp order, forming a continuous time-series dataset. The system uses a five-minute time unit to divide the continuous data stream into discrete time segments, each containing three thousand sampling points. Based on the data points within these time segments, the system constructs a smooth deformation curve using a cubic spline interpolation algorithm. This curve accurately reflects the deformation trajectory of the screen's bending area within the corresponding time period. Each deformation curve contains curvature change characteristics. The system calculates the curvature value using a numerical differentiation method and integrates the absolute value of the curvature along the time axis to obtain a cumulative curvature index. This metric quantifies the total bending intensity that a screen experiences over a specific period of time.

[0074] Historical deformation data is stored using a circular buffer structure, retaining complete deformation records for the most recent thirty days. The system automatically calculates a moving average of the historical cumulative curvature daily using a weighted average algorithm, assigning higher weight to recent data. For example, during weekday commuting hours, the system focuses on recording user behavior data involving frequent screen folding in subway cars; while in indoor usage scenarios on weekends, it primarily collects deformation characteristics when the screen is extended for extended periods. When calculating the real-time deformation deviation for the current time period, the system first retrieves the historical average for the same time period (e.g., 8:00-8:05 AM on weekdays) from the historical database. This average is calculated based on data from the past twenty similar time periods. The real-time deviation is defined as the algebraic difference between the current cumulative curvature and the historical average. A positive value indicates that the current bending intensity is higher than the historical norm, while a negative value reflects that the bending intensity is lower than expected. In specific application scenarios, when users continuously perform rapid folding operations, the system detects a sharp increase in the cumulative curvature, resulting in a significant positive deviation; conversely, when the screen remains in a semi-extended state for an extended period, a continuous negative deviation may occur. This deviation value is input as a key parameter into the subsequent light adaptation coefficient calculation module, providing a basis for the deformation state of the screen partition adjustment.

[0075] In a typical application scenario, users frequently fold their phones to check information during their morning commute. The system records that the deformation curve exhibits dense fluctuations during this period, with the cumulative curvature reaching its peak. During lunch breaks, users unfold their phones to watch videos, and the deformation curve flattens out. On the way home in the evening, frequent folding operations occur again, but the cumulative curvature is about 15% lower than in the morning. Based on this, the system determines the screen's fatigue state and generates a negative deviation value. At night, the system automatically updates historical data, incorporating the day's data into the historical database and recalculating the moving average, achieving dynamic evolution of the historical benchmark. The entire process forms a closed-loop feedback, ensuring that the real-time deformation deviation always reflects the latest screen usage status. The system has a special abnormal fluctuation monitoring mechanism. When the deviation value for three consecutive time segments exceeds three times the standard deviation, a hardware self-check program is automatically triggered to check for sensor malfunctions or mechanical structural abnormalities. This implementation method ensures that deformation monitoring has both historical continuity and timely capture of changes in usage habits, establishing an accurate physical state benchmark for adaptive adjustment.

[0076] Example 2: See Figure 3The system employs a collaborative processing mechanism involving ambient light and grip pressure. A light sensor array is distributed along the screen edge and the back of the device, capturing ambient light data at a sampling frequency of twenty times per second. Within a preset five-minute timeframe, the system continuously records light intensity values, automatically eliminating abnormal fluctuations caused by instantaneous reflections or shadows. The calculation of the light intensity range is based on the effective data set, taking the arithmetic difference between the maximum and minimum light intensity readings. This range reflects the fluctuation range of ambient brightness; for example, when walking outdoors under tree shade, the light intensity reading will periodically change within the range of 20,000 to 80,000 lux; while in an indoor office environment, the light intensity fluctuation range narrows to 300 to 500 lux. The system also receives real-time deformation deviation data from Example 1, which is derived from the cumulative curvature difference. The light adaptation coefficient is generated using a bivariate coupling algorithm, and its value increases with both the light intensity range and the deformation deviation. The specific correlation ratio was determined through calibration experiments. In a standard test environment, when the illumination difference reaches 30,000 lux and the deformation deviation exceeds 15%, the baseline value of the light adaptation coefficient is set to 0.7. The system monitors illumination fluctuation characteristics in real time. When it detects that the illumination intensity changes by more than 10,000 lux within ten seconds, it automatically increases the weighting factor of the light adaptation coefficient, giving this coefficient a higher proportion in subsequent calculations. This design makes the screen adjustment mechanism sensitive to sudden changes in illumination, such as moving from a dark cinema to a sunny street, allowing the system to quickly respond to strong light environments.

[0077] The generation of pressure distribution features relies on a matrix of 64 pressure-sensitive sensors embedded in the screen bezel. The sensors capture the user's grip pressure at a sampling rate of 100 times per second, forming a pressure distribution heatmap. The system divides the screen into eight symmetrical regions and calculates the similarity of pressure distribution in the corresponding regions. Similarity evaluation uses a vector space model, converting sensor readings for each region into feature vectors and quantifying the similarity using the cosine of the angle between the two regions. In scenarios where the user holds the screen with one hand, the pressure vectors in the left and right regions show significant differences, and the cosine similarity may drop below 0.3; however, when the user holds the screen with both hands, the similarity between the symmetrical regions can reach above 0.8. The rate of change of pressure gradient between adjacent regions is calculated using spatial difference, with the system comparing pressure value differences point by point along the screen's central axis. When a pressure difference between adjacent sensor nodes exceeds 50 kPa, it is marked as a point of significant gradient change. During user adjustments to their grip posture, finger movements cause rapid changes in the pressure gradient; for example, when the thumb slides from the center of the screen to the edge, the local rate of change of pressure gradient may exceed 80 kPa / mm. The system normalizes and fuses similarity indices and gradient change rates, mapping the two types of parameters to the 0-1 range and synthesizing a comprehensive feature value according to a preset weight ratio. When the pressure gradient change rate exceeds the critical value of 75 kPa / mm for three consecutive seconds, a feature reconstruction mechanism is triggered. The reconstruction process includes re-delineating the boundaries of the pressure-sensitive region and activating a backup feature extraction algorithm. For example, in special scenarios where the user is wearing gloves, the original pressure distribution pattern becomes ineffective, and the system automatically switches to a feature recognition mode based on pressure change trends to ensure the continuity of grip state analysis.

[0078] In a typical subway commuting scenario, when a user stands and holds their phone, the left thumb and right index finger create an asymmetrical pressure distribution. The system detects a similarity of only 0.25 between the left and right sides, with a pressure gradient of 65 kPa / mm between the thumb pressing area and adjacent areas. As the train enters a tunnel, the ambient light drops sharply from 30,000 lux to 50 lux, resulting in a light intensity difference of 29,950 lux. Simultaneously, the system receives deformation deviation data, showing that the current screen bending intensity is 12% higher than the historical average. Based on these parameters, the light adaptation coefficient is calculated to be 0.68, and the pressure distribution feature value is labeled as 0.42. When the user switches hands, the pressure distribution pattern changes drastically within a short period, and the rate of change of the pressure gradient exceeds a critical value, triggering a reconstruction mechanism. The system automatically adjusts the pressure zones from eight to twelve and activates a time-weighted feature extraction algorithm to effectively capture the transition state during the grip mode change. The reconstructed feature analysis shows an abnormally high pressure point in the right little finger area under the new grip posture, and the system transmits this information to the health assessment module. The entire process is completed within milliseconds, ensuring that the screen adjustment system responds synchronously to changes in user behavior and sudden environmental changes. The system's built-in anomaly detection module continuously monitors the consistency of data logic. When the light intensity is displayed as nighttime levels but the pressure distribution exhibits outdoor grip characteristics, the sensor cross-validation process is automatically initiated to rule out the possibility of data acquisition anomalies.

[0079] Example 3: See Figure 4 This involves a comprehensive processing flow of screen health assessment and status classification. The system receives the light adaptability coefficient from Example 2. and pressure distribution characteristic values These two parameters quantify the impact of ambient lighting conditions and user interaction patterns on screen status, respectively. The health assessment value is calculated using a weighted fusion model, the core expression of which is:

[0080]

[0081] in: This indicates the overall health assessment value of the screen. This represents the total number of screen partitions. and Representing the first The light adaptation coefficient and pressure distribution characteristic value of each partition. and It is the corresponding dynamic weighting coefficient, the value of which depends on the physical state of the screen. This refers to the importance weight of the partitions, which is dynamically adjusted based on the partition's location and function. When the screen is fully unfolded, the system uses a Hall sensor to detect that the unfolding angle exceeds 170 degrees; at this time, the corresponding bending area... The value is reduced to the range of 0.3-0.5, while the non-bent area retains the standard weight of 1.0. When the screen is bent, the gyroscope and strain sensor work together to detect the bending angle. When the bending angle exceeds 45 degrees, the system will reduce the weight of the bent area. The value was increased to the range of 1.2-1.5, while the weighting of inactive areas was appropriately reduced.

[0082] The dynamic classification threshold is established based on the historical distribution characteristics of health assessment values. The system maintains a sliding window containing the most recent 500 health assessment values, and the statistical characteristics of this window are updated synchronously each time a new health assessment value is calculated. Classification Threshold and Determined in the following ways: Take the 85th percentile of the health assessment values ​​within the window. The 15th percentile is used. This dynamic threshold mechanism allows the classification criteria to adapt to the usage habits of different users. For example, for users who frequently fold their phones, their baseline health assessment value is higher, and the corresponding classification threshold will also be raised.

[0083] The density clustering algorithm uses the DBSCAN method to project monitoring data samples into a three-dimensional feature space, with the three dimensions corresponding to health assessment values. Deformation deviation and pressure gradient The algorithm automatically identifies cluster centers by finding densely populated regions of samples, setting a minimum sample size of 5 and a neighborhood radius of 0.5 standard units. During clustering, the system first standardizes the input features to eliminate dimensional differences. Then, it calculates the Euclidean distance between samples and constructs a density connectivity graph. Finally, it outputs three state categories: a high-risk state corresponds to a health assessment value higher than [a certain value]. Furthermore, the abnormal areas have low sample density; the health assessment value corresponding to the warning status is between and The transition zone between these; the safety status corresponds to a health assessment value lower than [a certain value]. And the core region with a high sample density.

[0084] When a sample is classified as high-risk, the system automatically triggers a multi-level screen protection mechanism. The first level of protection includes limiting screen brightness to 50% of maximum brightness and reducing the refresh rate to 60Hz. If the health assessment value continues to deteriorate, the second level of protection activates, automatically disabling high-power display effects and lowering the screen color temperature. In extreme cases, the system will forcibly switch to a plain text display mode until the health status improves. The entire state classification process is executed every 30 seconds to ensure timely response to changes in screen state. The system also records each classification result, forming a state transition graph for analyzing screen state evolution patterns.

[0085] In a typical application scenario, users use their phones in a beach environment where strong sunlight reduces the light adaptation coefficient. The deformation generally increased, while sand particles entering the hinge mechanism caused deformation deviation. An abnormal increase. The health assessment value calculated by the system. The value reached 0.85, exceeding the current threshold. The device was identified as high-risk by the density clustering algorithm. The system immediately triggered a protection mechanism, reducing screen brightness and displaying a message advising the user to avoid frequently folding the phone in dusty environments. Upon returning indoors, the health assessment value... The value gradually dropped to 0.62, and the system automatically lifted the protection restrictions, restoring normal display parameters.

[0086] Example 4: Dynamic adjustment mechanism involving screen display parameters. The system extracts user's historical operation habit data from local storage. This data records the user's manual adjustments to screen brightness, color temperature, and other parameters over the past thirty days in a timestamp sequence. In the coffee shop scenario, the system detects that the user typically adjusts the screen color temperature to 6500K around 10 AM; this preference data is labeled as a weekday mode feature. The screen brightness adjustment model and the color temperature compensation model are linked through a three-layer neural network. The network input layer contains four nodes: health assessment value H (from Example 3), ambient light intensity Lux, user habit parameter U_h, and current screen shape identifier S_f. The hidden layer uses the ReLU activation function, and the output layer generates two control parameters: brightness correction coefficient ΔB and color temperature compensation value ΔC. The neural network weights are obtained through supervised learning training. The training dataset contains 100,000 sets of sensor data and corresponding records of the user's final adjustment results.

[0087] When a user unfolds their folded phone inside a subway car, the system detects the screen's transition from a folded to an unfolded state. The adjustment process is automatically divided into two phases: an initial transition phase lasting 800 milliseconds, using a linear interpolation algorithm to quickly adjust display parameters. The system uses the current brightness value B_c as the starting point and the target brightness value B_t as the ending point, calculating the interpolation result every 20 milliseconds. The instant the train exits the tunnel, ambient light surges from 50 lux to 30,000 lux, and the system detects a light intensity change rate of 6,000 lux per second, immediately initiating the transition phase. The linear interpolation process completes the brightness jump from 150 nits to 450 nits within 500 milliseconds. The steady-state adaptation phase then begins, with its duration automatically adjusted based on environmental stability. This phase uses an exponential decay algorithm for fine-tuning, setting the initial decay step size to 5 nits / second, and detecting ambient light changes every 100 milliseconds. When the light intensity fluctuation is less than 50 lux per second, the decay step size automatically decreases to 1 nit / second. In an office setting, when a user unfolds their foldable phone and places it on a table, the system detects that the ambient light level has stabilized at 400 lux. After three minutes of steady-state operation, the adjustment process automatically ends. See Table 1 for example data on screen parameter linkage adjustments.

[0088] Table 1: Record of multi-scene screen parameter linkage adjustment.

[0089]

[0090] In the library scenario, the ambient light stabilizes at 300 lux when the user unfolds their phone. During the initial transition phase, the system linearly interpolates the brightness from 100 nits in standby mode to the target value of 280 nits within 600 milliseconds. In the steady-state adaptation phase, slight light fluctuations caused by page turning are detected, and the brightness is fine-tuned in increments of 2 nits per second, eventually stabilizing at 275 nits. Color temperature is synchronized from an initial 6500K to the user's preferred 6100K, with a channel-specific progressive adjustment: the blue channel decreases by 0.3% per frame, the red channel increases by 0.1%, and the green channel remains unchanged. When the user adjusts the phone to a semi-folding stand mode, the system detects the change in shape based on accelerometer data, immediately recalculates the target parameters, and initiates a new adjustment cycle. The entire adjustment process maintains a smooth color temperature transition across different screen zones, avoiding visible color banding. The system continuously records actual display parameters in the background, and when a persistent deviation between the automatic adjustment result and the user's manual settings is detected, the system automatically marks this scene feature for subsequent model optimization.

[0091] Example 5: The system runs a background scene recording module, which generates scene identification codes by integrating light intensity sensor, color temperature sensor, and GPS positioning data. Each identification code corresponds to a specific environmental combination; for example, the code "L5500K_CT6500_GPS121.5" represents 5500 lux illumination, 6500K color temperature, and a specific latitude and longitude location. After the screen completes automatic adjustment, the system captures the final stable brightness value, color temperature value, and partition parameters, establishes a mapping relationship with the current scene identification code, and stores it in a habit database. When the user manually overrides the automatic adjustment parameters, the system records the operation timestamp, modification magnitude, and parameter difference before and after the modification. The modification frequency statistics use a sliding time window algorithm, counting the number of overwrite operations in 72-hour cycles. When the count exceeds five times, a learning mechanism is triggered.

[0092] In a typical office scenario, a user manually adjusted the automatically set color temperature from 5000K to 4500K at 10 AM for three consecutive days. The system detected that the frequency of this adjustment had reached a threshold. The learning module activated a weighted average algorithm, merging historical habit data with the new operation record at a 3:7 ratio. The original database recorded the color temperature value for this scenario as 4800K, the new record as 4500K, and the updated stored value as 4590K. The updated user habit data was immediately fed back to the brightness adjustment model. When the model detected the same scenario identifier again, the initial output color temperature target value was adjusted to 4590K. The system also recorded the user's operation amplitude characteristics. When it detected that the user habitually increased the brightness by more than 20%, the preset brightness baseline value was increased by 5% in subsequent adjustments.

[0093] The anomaly handling process includes real-time monitoring and a multi-level response mechanism. The screen driver chip has a built-in parameter feedback loop that collects the actual brightness and color temperature data of each zone every 100 milliseconds and compares it with the target parameters output by the neural network model. Deviation calculation uses a zone difference accumulation algorithm. When the accumulated deviation exceeds the allowable range and persists for three seconds, the sensor data verification program is initiated. The verification process first performs sensor self-diagnosis: the light sensor verifies circuit integrity through dark current testing, the pressure sensor performs zero-point drift calibration, and the deformation sensor performs a stepped signal response test. Data consistency checks verify physical logic correlation. For example, when the screen is stationary, the accelerometer reading should be a 1g vertical component. If a drastic change in pressure distribution is detected without updating the acceleration data, it is marked as a data conflict.

[0094] In a coastal highway driving scenario, the user mounts their phone on a car mount. The system suddenly detects a persistently excessive brightness deviation in the right-side partition, triggering a verification process. Self-diagnosis reveals a response delay in the three channels of the light sensor, and data consistency checks show that the pressure reading in this area is zero, but the brightness demand is abnormally high. The system determines the sensor is faulty and switches to a backup adjustment strategy: mirroring the left-side partition's light sensor data to compensate for the right-side data, with the pressure distribution referencing the historical driving mode average. Simultaneously, an encrypted fault log is generated, recording the time of the anomaly, environmental parameters, and handling measures. The log is uploaded to the cloud diagnostic platform via a secure channel. After the user ends their trip, the system automatically prompts for hardware testing.

[0095] In a beach recreational setting, a mobile phone is placed under a parasol in a semi-folded position. The system detects abnormal fluctuations in the color temperature deviation value. The verification program confirms that the sensor is working properly, but the data logic is abnormal: the light intensity is displayed as 20,000 lux, while the color temperature reading is only 3,000K, contradicting the high color temperature characteristics of the beach environment. The system recalculates the health assessment value and finds that strong reflected light is causing the color temperature sensor to misjudge. Based on the new assessment value, the color temperature compensation model is adjusted, the output value is corrected to 5500K, and this environmental characteristic is marked. In subsequent encounters with similar strong reflection scenarios, the system automatically loads the compensation parameters to reduce misjudgments. All anomaly handling is completed in the background, without affecting the smoothness of the user interface. The fault log uses a cyclic overwrite storage strategy, retaining the most recent fifty anomaly records for after-sales analysis.

[0096] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0097] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent data adjustment of an adaptive screen in a foldable mobile phone, characterized in that, include: Acquire monitoring data from multiple sensors of the foldable phone within a preset time period. The monitoring data includes deformation data of the screen bending area, ambient light intensity data, and user grip pressure distribution data. By comparing the deformation data of the screen bending area with the historical deformation data within each preset time period, the real-time deformation deviation of each area of ​​the screen is calculated. Analyze the fluctuation range of the current ambient light intensity data within a preset time period and the real-time deformation deviation to determine the light adaptation coefficient of each screen partition; By comparing the differences in the distribution of user grip pressure data in symmetrical areas of the screen, as well as the changes in pressure gradient between adjacent partitions, pressure distribution characteristics of each partition of the screen are generated. Based on the light adaptation coefficient and pressure distribution characteristics, calculate the health assessment value of the screen under the current usage state; Based on the health assessment value, all monitoring data samples are classified into different states, and the screen display parameters are dynamically adjusted according to the classification results.

2. The adaptive screen data intelligent adjustment method for foldable mobile phones according to claim 1, characterized in that, The real-time deformation deviation process for each region of the calculation screen includes: Arrange the deformation data of the bent area of ​​the screen according to the time series to generate a deformation curve; Calculate the cumulative curvature of the deformation curve over the entire time interval; Obtain the average cumulative curvature of the historical deformation curve; The real-time deformation deviation is characterized by the difference between the cumulative curvature of the current deformation curve and the average value of the cumulative curvature.

3. The adaptive screen data intelligent adjustment method for foldable mobile phones according to claim 2, characterized in that, The process of determining the light adaptation coefficient of each partition of the screen includes: Calculate the range of ambient light intensity data within a preset time period; The optical adaptation coefficient is positively correlated with the range value and the real-time deformation deviation, respectively. When the fluctuation range of ambient light intensity data exceeds a preset threshold, the weighting factor of the light adaptation coefficient is increased.

4. The adaptive screen data intelligent adjustment method for foldable mobile phones according to claim 3, characterized in that, The process of generating the pressure distribution characteristics of each partition of the screen includes: Calculate the similarity of pressure distribution in symmetrical regions of the screen; Obtain the rate of change of pressure gradient between adjacent partitions; The pressure distribution similarity and pressure gradient change rate are normalized and fused. When the rate of change of the pressure gradient exceeds a critical value, a mechanism for reconstructing the pressure distribution characteristics is triggered.

5. The adaptive screen data intelligent adjustment method for foldable mobile phones according to claim 1, characterized in that, The process of calculating the health assessment value of the screen under its current usage status includes: The light adaptation coefficient and pressure distribution characteristics are weighted and fused together; Calculate the cumulative value of the fusion result for all partitions of the screen; When the screen is fully unfolded, reduce the weight of the curved area partition; When the screen is bent, increase the weight of the bent area partition.

6. The adaptive screen data intelligent adjustment method for foldable mobile phones according to claim 5, characterized in that, The process of classifying the status of all monitoring data samples based on the health assessment value includes: Establish dynamic classification thresholds based on health assessment values; Density clustering algorithm is used to classify the state of monitoring data samples; The screen status is divided into three categories: high-risk status, warning status, and safe status. When a sample is classified as high-risk, the screen saver mechanism is automatically triggered.

7. The adaptive screen data intelligent adjustment method for foldable mobile phones according to claim 6, characterized in that, The dynamic adjustment of screen display parameters based on classification results includes: Obtain user's historical operation habits data; Establish a linkage mechanism between the screen brightness adjustment model and the color temperature compensation model; Input health assessment values, ambient light intensity data, and user historical operation habit data into the linkage model; Output the brightness correction parameters and color temperature compensation parameters for each zone of the screen.

8. The adaptive screen data intelligent adjustment method for foldable mobile phones according to claim 7, characterized in that, It also includes a phased adjustment process: When the screen changes from a bent state to an unfolded state, the adjustment process is divided into an initial transition phase and a steady-state adaptation phase. In the initial transition phase, linear interpolation is used to quickly approximate the target display parameters; During the steady-state adaptation phase, the exponential decay method is used for fine-tuning compensation. The timing of switching between the two stages is dynamically adjusted based on the rate of change of ambient light intensity.

9. The adaptive screen data intelligent adjustment method for foldable mobile phones according to claim 8, characterized in that, It also includes a user habit learning mechanism: Record the final adjustment parameters under different environmental scenarios; Analyze the frequency and magnitude of user-manually overriding automatic adjustment parameters; When the frequency of manual overwriting exceeds the preset number of times, update the user's historical operation habit data; The updated user history operation data is fed back into the screen brightness adjustment model.

10. The adaptive screen data intelligent adjustment method for foldable mobile phones according to claim 9, characterized in that, It also includes exception handling procedures: Continuously monitor the deviation between the actual display parameters and the target parameters of each screen partition; When the deviation value continues to exceed the allowable range, the sensor data verification mechanism is activated; If the sensor data verification passes, the health assessment value is recalculated; If the sensor data verification fails, switch to the backup adjustment strategy and generate a fault log.

Images