Liquid crystal display touch screen touch strategy optimization method and system based on use scenario recognition

Abstract

The application discloses a liquid crystal display touch control strategy optimization method and system based on a use scene recognition, relates to the technical field of display touch control, and comprises the following steps: acquiring data of a wearable device, and outputting a scene state feature sequence; performing steady-state migration analysis, recognizing and extracting a continuous interaction state, and outputting a touch control strategy optimization time window; performing dynamic balance of touch control interaction complexity and information priority, and generating an initial touch control strategy parameter set; extracting a motion intensity coefficient and an ambient light change coefficient, compensating the initial touch control strategy parameter set, and obtaining a modified touch control strategy parameter set; and optimizing and controlling a touch response mechanism of the liquid crystal display according to the touch control strategy optimization time window and the modified touch control strategy parameter set. The technical problems that the existing liquid crystal display scene recognition precision is insufficient, the touch control strategy lacks dynamic self-adaptive capability, and safety risks exist in the driving process are solved.

Description

Technical Field

[0001] This application relates to the field of display screen touch technology, specifically to a method and system for optimizing LCD display screen touch strategies based on usage scenario recognition. Background Technology

[0002] With the popularization of smart wearable devices and the increasing demand for in-vehicle information interaction, the intelligent optimization of the touch response strategy of LCD screens, as the core human-computer interaction interface of wearable devices, has become a key technology direction for improving driving safety.

[0003] However, existing technical solutions mainly focus on adjusting touch sensitivity and simplifying interface layout, relying heavily on single sensor data or user-triggered mechanisms for mode switching. This makes them prone to misjudging scenarios due to signal delays or connection interruptions. Furthermore, existing technologies often use preset, fixed parameter configurations, lacking the ability to adaptively compensate for real-time scene changes, making it difficult to balance ease of operation with the need to prevent accidental touches. Summary of the Invention

[0004] This application provides a method and system for optimizing LCD screen touch strategies based on usage scenario recognition, which solves the problems of insufficient scene recognition accuracy and lack of dynamic adaptive capability in existing LCD screens, leading to safety risks during driving.

[0005] The technical solution to the above-mentioned technical problems in this application is as follows: In a first aspect, this application provides a method for optimizing touch strategies for liquid crystal displays based on usage scenario recognition, the method comprising: Acquire inertial sensor data, ambient light sensor data, and current application status data from wearable devices, perform scene feature fusion, and output a scene status feature sequence; Perform steady-state transition analysis on the scene state feature sequence to identify and extract the continuous interaction state under the driving scene, and output the touch strategy optimization time window; Based on the scene state feature subsequence and user historical interaction response data within the time window of the touch strategy optimization, a dynamic balance between touch interaction complexity and information priority is achieved to generate an initial touch strategy parameter set that matches the driving scenario. Based on the scene state feature sequence, the motion intensity coefficient and ambient light change coefficient of the current scene are extracted, and the initial touch strategy parameter set is dynamically compensated to obtain the corrected touch strategy parameter set; The touch response mechanism of the LCD screen of the wearable device is optimized and controlled according to the optimized time window of the touch strategy and the modified touch strategy parameter set.

[0006] Secondly, this application provides a liquid crystal display touch strategy optimization system based on usage scenario recognition, including: The data acquisition module is used to acquire inertial sensor data, ambient light sensor data, and current application status data of the wearable device, and to perform scene feature fusion to output a scene status feature sequence. The feature analysis module is used to perform steady-state transition analysis of the scene state feature sequence, identify and extract the continuous interaction state under the driving scene, and output the touch strategy optimization time window. The parameter generation module is used to dynamically balance the complexity of touch interaction and information priority based on the scene state feature subsequence and user historical interaction response data within the time window of the touch strategy optimization, and generate an initial touch strategy parameter set that matches the driving scenario. The parameter compensation module is used to extract the motion intensity coefficient and ambient light change coefficient of the current scene based on the scene state feature sequence, and to dynamically compensate the initial touch strategy parameter set to obtain the corrected touch strategy parameter set. The control execution module is used to optimize and control the touch response mechanism of the LCD screen of the wearable device according to the optimized time window and the modified touch strategy parameter set.

[0007] This application provides one or more technical solutions, which have at least the following technical effects or advantages: This application provides a method and system for optimizing LCD touch strategies based on usage scenario recognition. First, by fusing scene features from multi-source sensor data, high-precision identification of driving scenarios is achieved, effectively overcoming misjudgments caused by delays or interruptions in single sensor signals. Second, a steady-state migration analysis mechanism is used to dynamically define the touch strategy optimization time window, ensuring that touch parameter adjustments only affect the continuous and stable range of the driving scenario. Third, a dynamic balance between touch interaction complexity and information priority is achieved based on historical user interaction response data, ensuring that the initial touch strategy parameter set matches the individual user's operating habits and response characteristics. Finally, through dual dynamic compensation of motion intensity coefficient and ambient light variation coefficient, the modified touch strategy parameter set possesses adaptive adjustment capabilities to real-time scene fluctuations, achieving a dynamic balance between operational convenience and anti-accidental touch requirements while ensuring driving safety.

[0008] Through the above technical solution, this application realizes the paradigm shift of LCD touch strategy from static preset to dynamic adaptive, improving the human-computer interaction safety and user experience of wearable devices in complex driving scenarios. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments 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 only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a flowchart illustrating the liquid crystal display touch strategy optimization method based on usage scenario recognition provided in an embodiment of this application; Figure 2 This is a schematic diagram of the structure of the LCD touch strategy optimization system based on usage scenario recognition provided in the embodiments of this application.

[0011] The components represented by each number in the attached diagram are explained below: Data acquisition module 11, feature analysis module 12, parameter generation module 13, parameter compensation module 14, execution control module 15. Detailed Implementation

[0012] This application provides a method and system for optimizing LCD touch strategies based on usage scenario recognition, which addresses the technical problem of insufficient scene recognition accuracy and lack of dynamic adaptive capability in existing LCD screens, leading to safety risks during driving.

[0013] Example 1, as Figure 1 As shown, this application provides a method for optimizing touch strategies for liquid crystal displays based on usage scenario recognition, including: S10: Acquire inertial sensor data, ambient light sensor data, and currently running application status data from the wearable device, perform scene feature fusion, and output a scene status feature sequence; In this embodiment of the application, the inertial sensor built into the wearable device collects the user's motion posture data at a preset sampling frequency of 50Hz, and obtains acceleration vector, angular velocity vector and azimuth angle data.

[0014] An ambient light sensor collects ambient light intensity values ​​at a low sampling frequency of 5Hz to identify changes in lighting conditions. Application status data is obtained in real time via an interface, including the currently running application package name, application type tag, and notification arrival records. The foreground application package name is a unique identifier for the application currently running on the screen and interacting with the user. The notification arrival records are a data set containing the source information and arrival time sequence of various notifications received by the wearable device at the current time.

[0015] Furthermore, the scene feature fusion process first performs sliding window filtering and zero-velocity correction on the inertial sensor data to eliminate sensor drift errors, and then extracts motion intensity features, including acceleration variance, angular velocity energy entropy, and attitude change rate; performs differential operations on the ambient light sensor data to extract illumination change trend features; and then aligns the motion intensity features, illumination change trend features, and application state data in time and concatenates them in dimensions to form a multi-dimensional scene state feature sequence, where each timestamp corresponds to a set of feature vectors containing motion, illumination, and application context information.

[0016] Specifically, step S10 in the method includes: The inertial sensor data includes at least three-axis acceleration data and three-axis angular velocity data; the ambient light sensor data includes at least time-series values ​​of light intensity; and the application status data includes at least the foreground application package name and notification arrival records. Extract the resultant acceleration amplitude from the triaxial acceleration data, calculate the variance of the resultant acceleration amplitude on the time axis, and obtain the acceleration fluctuation amount; The angular velocity change rate is extracted from the triaxial angular velocity data, and the number of times the angular velocity direction reverses per unit time is counted to obtain the angular velocity turning frequency. The acceleration fluctuation and angular velocity turning frequency are normalized respectively to obtain normalized values ​​of acceleration fluctuation and angular velocity turning frequency. The normalized values ​​of acceleration fluctuation and angular velocity turning frequency are weighted and fused to obtain the motion intensity coefficient. The difference in light intensity between adjacent time nodes is extracted from the time-series values ​​of light intensity to obtain the magnitude of light intensity variation. The frequency of light intensity change is obtained by counting the number of times the direction of light intensity change reverses per unit time. The amplitude of the light intensity change and the frequency of the light intensity change are normalized to obtain normalized values ​​of the amplitude of the light intensity change and the frequency of the light intensity change. The normalized values ​​of the light intensity variation amplitude and the normalized values ​​of the light intensity variation frequency are weighted and fused to obtain the ambient light variation coefficient. The motion intensity coefficient, ambient light change coefficient, foreground application package name, and notification arrival record are time-aligned and fused to generate a scene state feature sequence.

[0017] In this embodiment, firstly, the resultant acceleration amplitude is extracted from the triaxial acceleration data. Specifically, the sum of squares and square root of the triaxial acceleration components at each sampling moment are performed to obtain the instantaneous resultant acceleration amplitude, thereby constructing a time series of resultant acceleration amplitude. For this time series, the resultant acceleration amplitude sequence within a preset time window is taken, and the sum of squares of the differences between each data point and the sequence mean is calculated. This sum is then divided by the number of data points to obtain the acceleration fluctuation, which reflects the intensity of the user's body swaying during driving.

[0018] Secondly, the rate of change of angular velocity is extracted from the triaxial angular velocity data. First, the angular velocity components around each axis are extracted from the triaxial angular velocity data, and the difference in angular velocity values ​​at adjacent time points is calculated to obtain the rate of change of angular velocity. Further, the number of times the angular velocity direction reverses per unit time, i.e., the frequency of opposite signs between two adjacent sampling points, is counted to obtain the angular velocity turning frequency. This indicator reflects the frequent turning movements of the user's wrist or arm, which typically occur in scenarios with complex driving operations.

[0019] Specifically, the sequence of angular velocity change rates per unit time is traversed. When the product of two adjacent angular velocity change rates is negative, it is determined that an angular velocity direction reversal has occurred. The total number of direction reversals per unit time is the angular velocity turning frequency.

[0020] Furthermore, the acceleration fluctuation and angular velocity turning frequency are normalized respectively. The normalized value of acceleration fluctuation is (current acceleration fluctuation - minimum value) / (maximum value - minimum value), and the normalized value of angular velocity turning frequency is (current angular velocity turning frequency - minimum value) / (maximum value - minimum value).

[0021] Next, the normalized values ​​of acceleration fluctuation and angular velocity steering frequency are weighted and fused to obtain the motion intensity coefficient, i.e., motion intensity coefficient = α × normalized value of acceleration fluctuation + β × normalized value of angular velocity steering frequency. This normalization method can eliminate the influence of the difference in the dimensions of different sensors on the fusion result. α and β are preset weight coefficients, which can be optimized and determined by grid search or gradient descent method according to actual driving scenario data. Usually, the value of α is in the range of 0.6 to 0.8, and the value of β is in the range of 0.2 to 0.4, so as to highlight the dominant role of acceleration fluctuation in driving scenario recognition.

[0022] Secondly, regarding the ambient light sensor data, the difference in light intensity between adjacent time points is first extracted from the time-series values ​​of light intensity, i.e., the light intensity value at the current moment minus the light intensity value at the previous moment, to obtain the sequence of light intensity change amplitudes. Simultaneously, the number of times the direction of light intensity change reverses per unit time is counted. Specifically, a reversal of the direction of light intensity change is determined when the product of two adjacent light intensity differences is negative. The total number of reversals per unit time is the light change frequency. This indicator reflects the rapid switching of lighting conditions in the driving environment, such as when a vehicle enters or exits a tunnel or traverses a tree-lined road.

[0023] Next, the amplitude of light intensity change and the frequency of light intensity change are normalized respectively. The normalized value of the amplitude of light intensity change is (current amplitude of light intensity change - minimum value) / (maximum value - minimum value), and the normalized value of the frequency of light intensity change is (current frequency of light intensity change - minimum value) / (maximum value - minimum value).

[0024] The two are then weighted and fused to obtain the ambient light variation coefficient, which is: Ambient Light Variation Coefficient = γ × Normalized Value of Light Intensity Variation Amplitude + δ × Normalized Value of Light Variation Frequency. Here, γ and δ are preset weighting coefficients. Typically, γ ranges from 0.5 to 0.7, and δ ranges from 0.3 to 0.5, in order to balance the impact of the absolute amount of light intensity variation and the frequency of variation on scene perception.

[0025] Finally, the motion intensity coefficient, ambient light variation coefficient, foreground application package name, and notification arrival record are aligned according to a unified time base. The foreground application package name is represented by one-hot encoding, and the notification arrival record is represented by binary flag bits. These are then fused to generate a scene state feature sequence containing multi-dimensional information.

[0026] The frequency of light intensity changes is obtained by counting the number of times the direction of light intensity change reverses per unit time, including: Extract the first, second, and third light intensity values ​​at three consecutive time points from the light intensity time series values; Calculate the first difference between the second light intensity value and the first light intensity value, and calculate the second difference between the third light intensity value and the second light intensity value; When the product of the first difference and the second difference is negative, it is determined that a reversal of the direction of light intensity change has occurred at the second light intensity value; By iterating through the time series values ​​of light intensity, and accumulating the total number of times the direction of light intensity change reverses per unit time, the frequency of light change is obtained.

[0027] In this embodiment of the application, firstly, data from three consecutive time nodes are extracted from the light intensity time series values ​​and denoted as L(t-1), L(t), and L(t+1), which correspond to the first light intensity value, the second light intensity value, and the third light intensity value, respectively. Then, the first difference ΔL1=L(t)-L(t-1) and the second difference ΔL2=L(t+1)-L(t) are calculated.

[0028] When the product of ΔL1 and ΔL2 is negative, it indicates a change in the trend of light intensity change, meaning that a reversal in the direction of light intensity change occurs at the second light intensity value L(t). For example, if ΔL1 is positive and ΔL2 is negative, it means that the light intensity first increases and then decreases; if ΔL1 is negative and ΔL2 is positive, it means that the light intensity first decreases and then increases. Both of these cases are considered reversal events.

[0029] Furthermore, by iterating through the entire time series of light intensity values ​​and accumulating the count of all events that satisfy the direction reversal condition within a unit of time, the light change frequency can be obtained. Compared with the simple two-node difference method, this three-node determination mechanism can effectively filter out false direction changes caused by light sensor noise and improve the robustness of light change frequency calculation.

[0030] S20: Perform steady-state transition analysis on the scene state feature sequence to identify and extract the continuous interaction state under the driving scene, and output the touch strategy optimization time window; In this embodiment, steady-state migration analysis identifies a relatively stable continuous range of driving scenarios from a sequence of dynamically changing scene state features, thus avoiding interaction confusion that may be caused by adjusting touch strategies during drastic scene transitions.

[0031] Specifically, when the scene state change metric for multiple consecutive timestamps is below the steady-state determination threshold, the system is determined to enter the steady-state interval; when the scene state change metric exceeds the threshold and persists for a certain duration, a steady-state transition is determined. The start and end timestamps of the steady-state interval are extracted, and the final output is a touch strategy optimization time window that meets the minimum duration requirement, ensuring that touch parameter adjustments only take effect during stable driving scenarios and avoiding parameter jitter during scene switching.

[0032] Specifically, step S20 in the method includes: Extract the motion intensity coefficient and the foreground application package name from the scene state feature sequence; When the intensity coefficient of the motion is greater than the calibrated motion threshold and the foreground application package name belongs to the navigation application category, it is determined to be the entry point of the driving scene; Starting from the entry point of the driving scene, the scene state feature subsequence of subsequent time nodes is continuously scanned. The scene state feature subsequence is a continuous segment extracted from the scene state feature sequence. The starting time of the extraction is the entry point of the driving scene, and the ending time of the extraction is dynamically extended as the scanning process progresses. When the duration of time during which the intensity of motion is greater than the calibrated motion threshold reaches the dynamic duration threshold, the interval covered by the duration is set as the stable interval of the driving scenario. Based on the end point of the stable interval of the driving scenario, the motion intensity coefficient of subsequent time nodes is continuously monitored. When the motion intensity coefficient is less than or equal to the calibrated motion threshold and the duration reaches the exit confirmation time, it is determined as the exit point of the driving scenario. The time interval from the start time of the stable driving scenario to the exit point of the driving scenario is set as the touch strategy optimization time window.

[0033] In this embodiment, firstly, the intensity coefficient of motion and the name of the foreground application package are extracted from the scene state feature sequence as the basis for determining the driving scene. The intensity coefficient of motion reflects the intensity of the user's physical activity, while the name of the foreground application package indicates the type of application that the user is currently using. The combination of the two can effectively distinguish the driving scene from other daily activity scenes.

[0034] Specifically, the calibration motion threshold is obtained by collecting a large number of motion intensity coefficient samples in driving scenarios, statistically analyzing their distribution characteristics, and then calibrating the boundary value that distinguishes between driving and non-driving states. For example, the calibration motion threshold is set to 0.5, which can effectively distinguish between body swaying during driving and slight shaking in a stationary state.

[0035] At the same time, a whitelist of navigation application categories is established, which includes the package name identifiers of mainstream navigation applications. When the intensity coefficient of motion is greater than 0.5 and the foreground application package name belongs to the navigation application category, the entry point of the driving scenario is determined, indicating that the user is very likely to be driving and using the navigation function at the same time.

[0036] Secondly, starting from the entry point of the driving scene, a continuous scanning mechanism is initiated to dynamically capture subsequences of scene state features. The start time of this capture is fixed at the entry point of the driving scene, and the end time gradually extends forward as the scanning process progresses, forming a sliding time window. During the scanning process, the changes in the motion intensity coefficient are monitored in real time. When the coefficient remains greater than the calibrated motion threshold for a duration exceeding the dynamic duration threshold, the corresponding time interval is set as the stable interval of the driving scene. The dynamic duration threshold is set using an adaptive mechanism, with an initial value of 30 seconds. If an ambient light change coefficient exceeding 0.3 is detected within this interval, indicating a significant change in the driving environment, the dynamic duration threshold is extended to 60 seconds to ensure the reliability of the driving scene determination.

[0037] Secondly, the end of the stable driving scenario range marks the point where the user's driving state tends to be stable. Thereafter, the intensity coefficient of motion at subsequent time points is continuously monitored. When this coefficient is less than or equal to the calibrated motion threshold and the duration reaches the exit confirmation time, it is determined as the exit point of the driving scenario. The exit confirmation time is set to 20 seconds, which can effectively avoid misjudgment of the scenario due to brief periods of stillness such as temporary stops or waiting at red lights.

[0038] Finally, the time interval from the start time of the stable driving scenario to the exit point of the driving scenario is set as the touch strategy optimization time window. Within this window, targeted touch strategy optimization is performed to improve the interaction safety during driving.

[0039] The calculation steps for the dynamic duration threshold include: Obtain the motion intensity coefficient at the current time point; Calculate the ratio of the calibrated motion threshold to the current motion intensity coefficient, and multiply the ratio by the baseline duration threshold to obtain the initial value of the dynamic duration; The reference duration threshold is the standard duration used to confirm a stable state in a driving scenario when the motion intensity coefficient is equal to the calibrated motion threshold. The smaller value between the initial dynamic duration value and the maximum allowed duration threshold value is set as the upper limit of the dynamic duration value; The larger of the dynamic duration upper limit and the minimum allowed duration threshold is set as the dynamic duration threshold; The maximum allowed duration threshold is a first multiple of the baseline duration threshold, and the minimum allowed duration threshold is a second multiple of the baseline duration threshold, wherein the first multiple is greater than the second multiple.

[0040] In this embodiment, firstly, the motion intensity coefficient at the current time point is obtained, and the ratio of the calibrated motion threshold to the current motion intensity coefficient is calculated. When the ratio is less than 1, it indicates that the current motion state is more intense than the calibrated threshold, and a longer observation time is needed to confirm the stability of the scene; conversely, when the ratio r is close to 1, the required observation time is relatively short.

[0041] Secondly, the ratio is multiplied by the baseline duration threshold to obtain the initial value of the dynamic duration, that is, the initial value of the dynamic duration = baseline duration threshold × (calibrated motion threshold / current motion intensity coefficient). The baseline duration threshold is the standard duration used to confirm the stable state of the driving scenario when the motion intensity coefficient is exactly equal to the calibrated motion threshold. It is usually set to 30 seconds. This value is derived from statistical analysis of a large amount of driving scenario data, which can ensure the accuracy of the judgment while taking into account the real-time response.

[0042] Furthermore, to prevent the initial dynamic duration from being infinitely extended due to extremely intense motion, a maximum allowable duration threshold is set. The initial dynamic duration is compared with the maximum allowable duration threshold, and the smaller value is taken as the upper limit of the dynamic duration. The maximum allowable duration threshold is set as the first multiple of the baseline duration threshold, usually 4. When the motion intensity coefficient is much lower than the calibrated motion threshold, the initial dynamic duration will increase significantly. However, if the duration is confirmed to be too long, it will cause the optimization window startup delay to exceed 6 seconds, affecting the timely response of the touch strategy. Therefore, 4 times is used as the upper limit to ensure the timely response of the system.

[0043] Meanwhile, to prevent the initial dynamic duration from shortening excessively due to the motion state approaching the threshold, a minimum allowable duration threshold is set. The upper limit of the dynamic duration is compared with the minimum allowable duration threshold, and the larger value is taken as the final dynamic duration threshold. The minimum allowable duration threshold is set as the second multiple of the baseline duration threshold, usually 1.2, meaning the minimum allowable duration threshold is 15 seconds. When the motion intensity coefficient is much greater than the calibrated motion threshold, the initial dynamic duration will shrink significantly. However, if the duration is confirmed to be too short, requiring less than 1.8 seconds to determine stability, it may cause false triggering due to instantaneous violent vibrations. Therefore, 1.2 times is used as the lower limit to ensure the stability of the determination.

[0044] S30: Based on the scene state feature subsequence and user historical interaction response data within the time window of the touch strategy optimization, dynamically balance the touch interaction complexity and information priority, and generate an initial touch strategy parameter set that matches the driving scenario; In this embodiment, the dynamic balance between touch interaction complexity and information priority can achieve an optimal trade-off between the safety and efficiency of touch operation under the constraints of limited cognitive resources and interaction time in driving scenarios. Specifically, touch interaction complexity represents the cognitive load and operational precision required to complete a specific touch task, while information priority reflects the importance and urgency of different interface elements in the current driving situation. The dynamic balance between the two is achieved by establishing a multi-objective optimization model.

[0045] Based on the scene state feature subsequence and user historical interaction response data within the touch strategy optimization time window, an initial touch strategy parameter set matching the driving scenario is generated, which includes key parameters such as touch sensitivity level, touch area division density, interface element response latency threshold, and multi-level menu expansion depth limit.

[0046] Specifically, step S30 in the method includes: Extract the motion intensity coefficient, foreground application package name, and notification arrival record from the scene state feature subsequence within the time window of the touch strategy optimization; Based on the interaction records within the historical time interval, the user's historical interaction response data is obtained. The user's historical interaction response data includes the response time consumed by the user when viewing each notification source at a historical time node, the records of accidental touch operations that occurred during the historical interaction process, and the corresponding intensity coefficient of movement at that time. Based on the foreground application package name, retrieve from the user's historical interaction response data the viewing time and operation error rate of different notification sources under the application running state corresponding to the foreground application package name; Based on the motion intensity coefficient, retrieve the probability of accidental touches by the user for different touch gestures within the motion intensity range corresponding to the motion intensity coefficient from the user's historical interaction response data; The viewing time, operation error rate, and accidental touch probability are normalized and then weighted and fused to obtain the comprehensive interaction cost of each notification source; Based on the comprehensive interaction cost, the notification sources are prioritized to obtain the priority ranking result; An initial touch strategy parameter set is generated based on the priority sorting result and the accidental touch probability.

[0047] In this embodiment, firstly, the motion intensity coefficient, the foreground application package name, and the notification arrival record are extracted from the scene state feature subsequence within the touch strategy optimization time window as the scene basis for generating touch strategy parameters. The motion intensity coefficient affects the accuracy requirements of touch operation, the foreground application package name determines the type characteristics of the current interaction task, and the notification arrival record reflects the real-time status of external information interference.

[0048] Secondly, historical user interaction response data is obtained based on interaction records within historical time intervals. This data includes: the response time consumed by users when viewing various notification sources at historical time points, which quantifies the cognitive processing difficulty of different information sources; records of accidental touch operations during historical interactions, reflecting user operating habits and interface adaptation; and the corresponding motion intensity coefficient at that time, establishing a correlation mapping between motion state and interaction quality. The selection of historical time intervals employs a sliding window mechanism, typically covering the interaction data of the most recent 30 days to ensure the timeliness and representativeness of historical data.

[0049] Furthermore, based on the foreground application package name, conditional retrieval is performed from the user's historical interaction response data to obtain the viewing time and operation error rate of the user for different notification sources under the specific application running state. Here, viewing time refers to the average response time consumed by the user to view each notification source, and operation error rate is the ratio of the number of accidental touches when the user operates on each notification source to the total number of operations.

[0050] For example, when the backend application package name is a navigation application, the search retrieves the average viewing time of users receiving notifications from different sources such as instant messaging notifications, system prompts, and news pushes during navigation, as well as the accidental touch rate caused by distracted operation of the navigation interface. This search process adopts a strategy that combines exact matching and fuzzy matching. For application package names that do not appear in historical data, similarity inference is made based on the application category.

[0051] Simultaneously, based on the motion intensity coefficient, the probability of accidental touches within the corresponding motion intensity range is retrieved from the user's historical interaction response data. The motion intensity coefficient is divided into several discrete intervals, such as 0 to 0.3 for low intensity, 0.3 to 0.7 for medium intensity, and 0.7 to 1.0 for high intensity. The ratio of the number of accidental touches for each touch gesture type to the total number of operations is calculated within each interval to obtain the probability of accidental touches for each touch gesture within the motion intensity range.

[0052] Subsequently, the viewing time, operation error rate, and accidental touch probability were normalized to eliminate the difference in units. The Min-Max normalization method was used to map each indicator to the range of 0 to 1. The longer the viewing time, the larger the normalized value. The higher the operation error rate and accidental touch probability, the larger the normalized value.

[0053] For example, weighted fusion is performed by setting weight coefficients. The weight for viewing time is usually set to 0.4, the weight for operation error rate is set to 0.35, and the weight for accidental touch probability is set to 0.25. This weight configuration is determined based on the principle of prioritizing the scarcity of cognitive resources and operational safety in driving scenarios. The comprehensive interaction cost of each notification source is obtained through fusion calculation.

[0054] The notification sources are prioritized based on the overall interaction cost. The lower the overall interaction cost, the less processing burden and the higher the safety of the notification source in the driving scenario. The ranking results are divided into three levels: high priority, medium priority and low priority. High priority notifications are allowed to be presented in the form of sound, vibration and simplified interface in the driving scenario. Medium priority notifications are only prompted by vibration. Low priority notifications are completely suppressed until the driving scenario ends.

[0055] Finally, an initial set of touch strategy parameters is generated based on the priority ranking results and the probability of accidental touches. For motion intensity ranges with high accidental touch probabilities, the touch sensitivity level is reduced, the touch area division density is increased, and the response delay threshold of interface elements is extended to filter out erroneous operations. For ranges with low accidental touch probabilities, the sensitivity is moderately increased to ensure interaction efficiency. At the same time, a multi-level menu expansion depth limit is set according to the priority ranking results. High-priority tasks are allowed to expand to the third level, medium-priority tasks are limited to the second level, and low-priority tasks only retain the first-level entry, thereby achieving a dynamic balance between touch interaction complexity and information priority in driving scenarios.

[0056] The initial touch strategy parameter set is generated based on the priority sorting result and the accidental touch probability, including: Based on the priority ranking result, the top N notification sources with the lowest overall interaction cost are selected from all notification sources and set as the high-priority notification set. Here, N is determined according to the total number of notification sources in the notification arrival record at the current time node and is inversely proportional to the probability of accidental touch. Information display time is generated based on the number of notification sources in the high-priority notification set, wherein the information display time is the duration of a single display of each notification source in the high-priority notification set on the lock screen interface, and the information display time is inversely proportional to the number of notification sources in the high-priority notification set; Based on the accidental touch probability, interaction mode parameters are generated, wherein the interaction mode parameters include the sensitivity adjustment coefficient of sliding touch and the response area scaling coefficient of click touch. The sensitivity adjustment coefficient is obtained by multiplying the sensitivity reference value by a first adjustment factor, which is proportional to the difference between the normalized value of the false touch probability. The response region scaling factor is obtained by multiplying the response region baseline value by a second adjustment factor, which is proportional to the sum of the normalized values ​​of the false touch probability. The information display time and the interaction mode parameters are integrated into an initial touch strategy parameter set.

[0057] In this embodiment, firstly, based on the priority ranking result, the top N notification sources with the lowest overall interaction cost are selected from all notification sources and set as the high-priority notification set. The determination of N needs to take into account both notification processing requirements and driving safety constraints. The specific calculation formula is N=max{1, total number of notification sources × (1-normalized value of accidental touch probability)} and rounded down. The larger the normalized value of accidental touch probability, the higher the risk of touch operation in the current motion state, and the number of high-priority notifications needs to be reduced to reduce cognitive load. Conversely, the notification processing capacity can be appropriately increased.

[0058] For example, when the total number of notification sources is 10 and the normalized value of the accidental touch probability is 0.6, N=max{1, 10×(1-0.6)}=4, that is, the 4 notification sources with the lowest overall interaction cost are selected and included in the high-priority notification set.

[0059] Secondly, the information display time is generated based on the number of high-priority notification sources. Information display time refers to the duration of each high-priority notification displayed on the lock screen in a single instance. Information display time = Baseline display time / Number of high-priority notification sources, where the baseline display time is a fixed duration, such as 2 seconds. When the number of high-priority notification sources is 1, the information display time is the baseline display time; as the number increases, the display time decreases inversely, ensuring that the total duration of the lock screen carousel display is controlled within a reasonable range, avoiding excessive notifications that could distract the user from driving due to prolonged screen focus.

[0060] Next, interaction mode parameters are generated to adapt to the touch operation characteristics under the current motion state. The interaction mode parameters include the sensitivity adjustment coefficient of sliding touch and the response area scaling coefficient of click touch. The two optimize the touch experience from two dimensions: operation accuracy compensation and operation fault tolerance enhancement.

[0061] The sensitivity adjustment coefficient for swipe touch is calculated based on the dynamic correlation between the normalized value of the false touch probability and the sensitivity benchmark value. Sensitivity adjustment coefficient = sensitivity benchmark value × (1 - normalized value of false touch probability), where the sensitivity benchmark value is the default sensitivity setting. When the normalized value of the false touch probability is high, it indicates that the user has difficulty accurately controlling the swipe trajectory under the current motion state, and the sensitivity adjustment coefficient is correspondingly reduced, requiring a larger displacement to trigger a response, thus filtering out unintentional swipes caused by vehicle bumps. When the normalized value of the false touch probability is low, the sensitivity adjustment coefficient is moderately increased to ensure a smooth response for swipe operations.

[0062] The scaling factor for the response area of ​​a click-based touch event is calculated based on the positive correlation between the normalized value of the accidental touch probability and the baseline value of the response area. The scaling factor is calculated as: Response Area Baseline Value × (1 + Accidental Touch Probability Normalized Value), where the baseline value is the standard touch area size of the interface element. When the accidental touch probability normalized value is high, the scaling factor increases accordingly, expanding the actual response range of the clickable area outwards, making it easier for users to hit the target even when their gaze is off-center from the screen. When the accidental touch probability normalized value is low, the scaling factor approaches 1, maintaining the standard response area to avoid accidental triggering between adjacent elements.

[0063] Finally, the information display time and interaction mode parameters are integrated into an initial touch strategy parameter set. This parameter set serves as the basic configuration for subsequent adaptive optimization and will continue to be effective within the touch strategy optimization time window and dynamically corrected based on real-time feedback.

[0064] S40: Based on the scene state feature sequence, extract the motion intensity coefficient and ambient light change coefficient of the current scene, and dynamically compensate the initial touch strategy parameter set to obtain the corrected touch strategy parameter set; In this embodiment, the initial touch strategy parameter set is dynamically compensated to eliminate parameter adaptation deviations caused by instantaneous fluctuations in scene state, thereby further improving the environmental robustness of the touch strategy.

[0065] The motion intensity coefficient characterizes the stability of the vehicle's current driving state, while the ambient light variation coefficient reflects the impact of external light conditions on screen visibility and touch recognition accuracy. The two work together on the initial touch strategy parameter set, and the parameters are corrected in real time by establishing a compensation mapping relationship.

[0066] Specifically, step S40 in the method includes: Extract the motion intensity coefficient and ambient light change coefficient at the current time point from the scene state feature sequence; Based on the motion intensity coefficient, the information display time in the initial touch strategy parameter set is first compensated to obtain the compensated information display time. Based on the ambient light variation coefficient, a second compensation is performed on the compensated information display time to obtain the corrected information display time; Based on the motion intensity coefficient, a third compensation is performed on the interaction mode parameters in the initial touch strategy parameter set to obtain the corrected interaction mode parameters. The correction information display time and the correction interaction mode parameters are integrated into a correction touch strategy parameter set.

[0067] In this embodiment, firstly, the motion intensity coefficient and ambient light change coefficient at the current time point are extracted from the scene state feature sequence as two-dimensional inputs for dynamic compensation. The motion intensity coefficient is obtained through accelerometer data fusion calculation, while the ambient light change coefficient is determined based on the deviation rate between the ambient light sensor's collected values ​​and historical benchmark values.

[0068] Secondly, the information display time is compensated based on the motion intensity coefficient. This is used to shorten the information display time when the motion intensity coefficient increases. The compensated information display time = information display time × first compensation coefficient. The first compensation coefficient is usually between 0.3 and 0.5. This range has been verified by real vehicle testing and can control the duration of a single gaze to not exceed the safe threshold while ensuring information readability.

[0069] A second compensation is applied to the information display time after initial compensation based on the ambient light variation coefficient. This compensation extends the corrected information display time when the ambient light variation coefficient increases. The direction of compensation depends on the direction and magnitude of the light change. The corrected information display time = compensated information display time × second compensation coefficient, where the second compensation coefficient = 1 + ambient light variation coefficient. When the ambient light variation coefficient is positive, it indicates enhanced light, and the second compensation coefficient is greater than 1. The corrected information display time is extended to meet the visibility requirements under strong light. When the ambient light variation coefficient is negative, it indicates weakened light, and the second compensation coefficient is less than 1. The corrected information display time is shortened to prevent users from excessively staring at the screen due to pupil adjustment delay in low-light environments.

[0070] Thirdly, a third compensation is applied to the interaction mode parameters based on the motion intensity coefficient. This involves coordinated adjustments across two dimensions: swipe touch sensitivity and click response area. This reduces the sensitivity adjustment coefficient and increases the response area scaling coefficient when the motion intensity coefficient increases. The corrected information display time and corrected interaction mode parameters, after dual-dimensional compensation, are integrated into a corrected touch strategy parameter set. This parameter set is fully adapted to the current driving scenario characteristics and can be directly applied to real-time rendering and interaction response processing of the touch interface.

[0071] Specifically, based on the motion intensity coefficient, a third compensation is performed on the interaction mode parameters in the initial touch strategy parameter set, including: Obtain the current sensitivity adjustment coefficient and response area scaling coefficient from the initial touch strategy parameter set; Based on the ratio of the exercise intensity coefficient to the calibrated exercise threshold, a sensitivity compensation factor and a response region compensation factor are calculated respectively, wherein the sensitivity compensation factor is inversely proportional to the ratio, and the response region compensation factor is directly proportional to the ratio. Multiply the sensitivity adjustment coefficient by the sensitivity compensation factor to obtain the corrected sensitivity adjustment coefficient, and the corrected sensitivity adjustment coefficient is not less than the lower sensitivity threshold. Multiply the response region scaling factor by the response region compensation factor to obtain the corrected response region scaling factor, and the corrected response region scaling factor is not greater than the upper limit threshold of the scaling factor. The modified sensitivity adjustment coefficient and the modified response region scaling coefficient are integrated into a modified interaction mode parameter.

[0072] In this embodiment, firstly, the current sensitivity adjustment coefficient and response area scaling coefficient are extracted from the initial touch strategy parameter set as compensation benchmarks. The sensitivity adjustment coefficient reflects the response sensitivity of sliding touch in the current motion state, while the response area scaling coefficient characterizes the extent to which the effective touch range of the click operation is expanded.

[0073] Secondly, the ratio of the motion intensity coefficient to the calibrated motion threshold is calculated. This ratio quantifies the deviation of the current motion state from standard driving conditions. The calibrated motion threshold is typically set to 0.5, corresponding to a moderate-intensity normal driving state. The sensitivity compensation factor is calculated using the formula: Sensitivity Compensation Factor = Calibrated Motion Threshold / Motion Intensity Coefficient. When the motion intensity coefficient is higher than the calibrated motion threshold, the sensitivity compensation factor is less than 1, indicating that sensitivity needs to be reduced to suppress accidental touches. When the motion intensity coefficient is lower than the calibrated motion threshold, the sensitivity compensation factor is greater than 1, and sensitivity can be appropriately increased to ensure smooth operation.

[0074] The response area compensation factor is calculated using the formula Response Area Compensation Factor = Motion Intensity Coefficient / Calibrated Motion Threshold. It is proportional to the above ratio. When the motion intensity coefficient increases, the response area compensation factor increases accordingly, thereby compensating for the decrease in positioning accuracy caused by vehicle bumps by expanding the effective touch area.

[0075] Next, the sensitivity adjustment coefficient is multiplied by the sensitivity compensation factor to obtain the corrected sensitivity adjustment coefficient. Simultaneously, a lower sensitivity threshold is set to prevent excessive reduction that could lead to operational lag. This lower threshold is calibrated based on ergonomic experiments and is typically set to 0.3 times the baseline sensitivity. The response area scaling factor is multiplied by the response area compensation factor to obtain the corrected response area scaling factor. Simultaneously, an upper scaling factor threshold is set to prevent excessive expansion that could lead to overlapping interface elements and accidental touches. This upper threshold is determined based on the physical constraints of the touch interface layout and is typically set to 2.0 times the standard response area.

[0076] Finally, the corrected sensitivity adjustment coefficient and response area scaling coefficient are integrated into the corrected interaction mode parameters to form a touch response configuration that dynamically adapts to the current intensity of motion.

[0077] S50: Optimize and control the touch response mechanism of the LCD screen of the wearable device according to the optimized time window of the touch strategy and the modified touch strategy parameter set.

[0078] In this embodiment, the touch response mechanism of the LCD screen of the wearable device is optimized and controlled according to the touch strategy optimization time window and the modified touch strategy parameter set, so as to realize a complete closed loop from parameter generation to actual application of the touch strategy.

[0079] The touch strategy optimization time window defines the effective period of the current modified touch strategy parameter set. The duration of this window is usually set to 30 to 120 seconds, and the specific value is dynamically adjusted according to the fluctuation frequency of the scene state feature sequence. When the motion state tends to be stable, the window is appropriately extended to reduce the inconsistency of interaction caused by parameter switching. When the environmental conditions change drastically, the window is shortened to ensure the real-time adaptability of the strategy.

[0080] At the specific optimization and control level of the touch response mechanism, the notification carousel logic of the lock screen interface is first configured based on the corrected information display time. The single display duration of each notification source in the high-priority notification set is set as the corrected information display time, and the total carousel cycle is calculated to ensure that it does not exceed the safe gaze duration threshold. At the same time, based on the corrected sensitivity adjustment coefficient in the corrected interaction mode parameters, the sliding recognition algorithm parameters of the touch driver layer are recalibrated, including the minimum sliding displacement threshold, the sliding speed sampling window length, and the direction determination tolerance angle, so that the triggering conditions of the sliding operation match the user's control ability in the current motion state.

[0081] In summary, compared with existing technologies, this application realizes a paradigm shift from single static configuration to multi-dimensional adaptive optimization by constructing a dynamic mapping mechanism between scene state feature sequences and touch strategy parameters.

[0082] In summary, the embodiments of this application have at least the following technical effects: This application provides a method for optimizing LCD touch strategies based on usage scenario recognition. First, by fusing scene features from multi-source sensor data, high-precision identification of driving scenarios is achieved, effectively overcoming misjudgments caused by delays or interruptions in single sensor signals. Second, a steady-state migration analysis mechanism is used to dynamically define the touch strategy optimization time window, ensuring that touch parameter adjustments only affect the continuous and stable range of the driving scenario. Third, a dynamic balance between touch interaction complexity and information priority is achieved based on historical user interaction response data, ensuring that the initial touch strategy parameter set matches the individual user's operating habits and response characteristics. Finally, through dual dynamic compensation of motion intensity coefficient and ambient light variation coefficient, the modified touch strategy parameter set possesses adaptive adjustment capabilities to real-time scene fluctuations, achieving a dynamic balance between operational convenience and anti-accidental touch requirements while ensuring driving safety.

[0083] Through the above technical solution, this application realizes the paradigm shift of LCD touch strategy from static preset to dynamic adaptive, improving the human-computer interaction safety and user experience of wearable devices in complex driving scenarios.

[0084] Example 2, as Figure 2 As shown, based on the same inventive concept as the liquid crystal display touch strategy optimization method based on usage scenario recognition provided in Embodiment 1, this application also provides a liquid crystal display touch strategy optimization system based on usage scenario recognition, including: The data acquisition module 11 is used to acquire inertial sensor data, ambient light sensor data and currently running application status data of the wearable device, and perform scene feature fusion to output a scene status feature sequence. The feature analysis module 12 is used to perform steady-state transition analysis of the scene state feature sequence, identify and extract the continuous interaction state under the driving scene, and output the touch strategy optimization time window. The parameter generation module 13 is used to dynamically balance the complexity of touch interaction and information priority based on the scene state feature subsequence and user historical interaction response data within the time window of the touch strategy optimization, and generate an initial touch strategy parameter set that matches the driving scenario. The parameter compensation module 14 is used to extract the motion intensity coefficient and ambient light change coefficient of the current scene according to the scene state feature sequence, and to dynamically compensate the initial touch strategy parameter set to obtain the corrected touch strategy parameter set. The control execution module 15 is used to optimize and control the touch response mechanism of the liquid crystal display screen of the wearable device according to the optimized time window of the touch strategy and the modified touch strategy parameter set.

[0085] In one embodiment, the data acquisition module 11 is specifically used for: The inertial sensor data includes at least three-axis acceleration data and three-axis angular velocity data; the ambient light sensor data includes at least time-series values ​​of light intensity; and the application status data includes at least the foreground application package name and notification arrival records. Extract the resultant acceleration amplitude from the triaxial acceleration data, calculate the variance of the resultant acceleration amplitude on the time axis, and obtain the acceleration fluctuation amount; The angular velocity change rate is extracted from the triaxial angular velocity data, and the number of times the angular velocity direction reverses per unit time is counted to obtain the angular velocity turning frequency. The acceleration fluctuation and angular velocity turning frequency are normalized respectively to obtain normalized values ​​of acceleration fluctuation and angular velocity turning frequency. The normalized values ​​of acceleration fluctuation and angular velocity turning frequency are weighted and fused to obtain the motion intensity coefficient. The difference in light intensity between adjacent time nodes is extracted from the time-series values ​​of light intensity to obtain the magnitude of light intensity variation. The frequency of light intensity change is obtained by counting the number of times the direction of light intensity change reverses per unit time. The amplitude of the light intensity change and the frequency of the light intensity change are normalized to obtain normalized values ​​of the amplitude of the light intensity change and the frequency of the light intensity change. The normalized values ​​of the light intensity variation amplitude and the normalized values ​​of the light intensity variation frequency are weighted and fused to obtain the ambient light variation coefficient. The motion intensity coefficient, ambient light change coefficient, foreground application package name, and notification arrival record are time-aligned and fused to generate a scene state feature sequence.

[0086] Furthermore, in one embodiment of the application, the frequency of light intensity change is obtained by counting the number of times the direction of light intensity change reverses per unit time, including: Extract the first, second, and third light intensity values ​​at three consecutive time points from the light intensity time series values; Calculate the first difference between the second light intensity value and the first light intensity value, and calculate the second difference between the third light intensity value and the second light intensity value; When the product of the first difference and the second difference is negative, it is determined that a reversal of the direction of light intensity change has occurred at the second light intensity value; By iterating through the time series values ​​of light intensity, and accumulating the total number of times the direction of light intensity change reverses per unit time, the frequency of light change is obtained.

[0087] In one embodiment, the feature analysis module 12 is specifically used for: Extract the motion intensity coefficient and the foreground application package name from the scene state feature sequence; When the intensity coefficient of the motion is greater than the calibrated motion threshold and the foreground application package name belongs to the navigation application category, it is determined to be the entry point of the driving scene; Starting from the entry point of the driving scene, the scene state feature subsequence of subsequent time nodes is continuously scanned. The scene state feature subsequence is a continuous segment extracted from the scene state feature sequence. The starting time of the extraction is the entry point of the driving scene, and the ending time of the extraction is dynamically extended as the scanning process progresses. When the duration of time during which the intensity of motion is greater than the calibrated motion threshold reaches the dynamic duration threshold, the interval covered by the duration is set as the stable interval of the driving scenario. Based on the end point of the stable interval of the driving scenario, the motion intensity coefficient of subsequent time nodes is continuously monitored. When the motion intensity coefficient is less than or equal to the calibrated motion threshold and the duration reaches the exit confirmation time, it is determined as the exit point of the driving scenario. The time interval from the start time of the stable driving scenario to the exit point of the driving scenario is set as the touch strategy optimization time window.

[0088] Furthermore, in one embodiment of the application, the calculation step of the dynamic duration threshold includes: Obtain the motion intensity coefficient at the current time point; Calculate the ratio of the calibrated motion threshold to the current motion intensity coefficient, and multiply the ratio by the baseline duration threshold to obtain the initial value of the dynamic duration; The reference duration threshold is the standard duration used to confirm a stable state in a driving scenario when the motion intensity coefficient is equal to the calibrated motion threshold. The smaller value between the initial dynamic duration value and the maximum allowed duration threshold value is set as the upper limit of the dynamic duration value; The larger of the dynamic duration upper limit and the minimum allowed duration threshold is set as the dynamic duration threshold; The maximum allowed duration threshold is a first multiple of the baseline duration threshold, and the minimum allowed duration threshold is a second multiple of the baseline duration threshold, wherein the first multiple is greater than the second multiple.

[0089] Furthermore, the parameter generation module 13 is specifically used for: Extract the motion intensity coefficient, foreground application package name, and notification arrival record from the scene state feature subsequence within the time window of the touch strategy optimization; Based on the interaction records within the historical time interval, the user's historical interaction response data is obtained. The user's historical interaction response data includes the response time consumed by the user when viewing each notification source at a historical time node, the records of accidental touch operations that occurred during the historical interaction process, and the corresponding intensity coefficient of movement at that time. Based on the foreground application package name, retrieve from the user's historical interaction response data the viewing time and operation error rate of different notification sources under the application running state corresponding to the foreground application package name; Based on the motion intensity coefficient, retrieve the probability of accidental touches by the user for different touch gestures within the motion intensity range corresponding to the motion intensity coefficient from the user's historical interaction response data; The viewing time, operation error rate, and accidental touch probability are normalized and then weighted and fused to obtain the comprehensive interaction cost of each notification source; Based on the comprehensive interaction cost, the notification sources are prioritized to obtain the priority ranking result; An initial touch strategy parameter set is generated based on the priority sorting result and the accidental touch probability.

[0090] Further, based on the priority ranking result and the accidental touch probability, an initial touch strategy parameter set is generated, including: Based on the priority ranking result, the top N notification sources with the lowest overall interaction cost are selected from all notification sources and set as the high-priority notification set. Here, N is determined according to the total number of notification sources in the notification arrival record at the current time node and is inversely proportional to the probability of accidental touch. Information display time is generated based on the number of notification sources in the high-priority notification set, wherein the information display time is the duration of a single display of each notification source in the high-priority notification set on the lock screen interface, and the information display time is inversely proportional to the number of notification sources in the high-priority notification set; Based on the accidental touch probability, interaction mode parameters are generated, wherein the interaction mode parameters include the sensitivity adjustment coefficient of sliding touch and the response area scaling coefficient of click touch. The sensitivity adjustment coefficient is obtained by multiplying the sensitivity reference value by a first adjustment factor, which is proportional to the difference between the normalized value of the false touch probability. The response region scaling factor is obtained by multiplying the response region baseline value by a second adjustment factor, which is proportional to the sum of the normalized values ​​of the false touch probability. The information display time and the interaction mode parameters are integrated into an initial touch strategy parameter set.

[0091] In one embodiment, the parameter compensation module 14 is specifically used for: Extract the motion intensity coefficient and ambient light change coefficient at the current time point from the scene state feature sequence; Based on the motion intensity coefficient, the information display time in the initial touch strategy parameter set is first compensated to obtain the compensated information display time. Based on the ambient light variation coefficient, a second compensation is performed on the compensated information display time to obtain the corrected information display time; Based on the motion intensity coefficient, a third compensation is performed on the interaction mode parameters in the initial touch strategy parameter set to obtain the corrected interaction mode parameters. The correction information display time and the correction interaction mode parameters are integrated into a correction touch strategy parameter set.

[0092] Furthermore, based on the motion intensity coefficient, a third compensation is performed on the interaction mode parameters in the initial touch strategy parameter set, including: Obtain the current sensitivity adjustment coefficient and response area scaling coefficient from the initial touch strategy parameter set; Based on the ratio of the exercise intensity coefficient to the calibrated exercise threshold, a sensitivity compensation factor and a response region compensation factor are calculated respectively, wherein the sensitivity compensation factor is inversely proportional to the ratio, and the response region compensation factor is directly proportional to the ratio. Multiply the sensitivity adjustment coefficient by the sensitivity compensation factor to obtain the corrected sensitivity adjustment coefficient, and the corrected sensitivity adjustment coefficient is not less than the lower sensitivity threshold. Multiply the response region scaling factor by the response region compensation factor to obtain the corrected response region scaling factor, and the corrected response region scaling factor is not greater than the upper limit threshold of the scaling factor. The modified sensitivity adjustment coefficient and the modified response region scaling coefficient are integrated into a modified interaction mode parameter.

Claims

1. A method for optimizing LCD touch strategies based on usage scenario recognition, characterized in that, The method includes: Acquire inertial sensor data, ambient light sensor data, and current application status data from wearable devices, perform scene feature fusion, and output a scene status feature sequence; Perform steady-state transition analysis on the scene state feature sequence to identify and extract continuous interaction states in the driving scenario, and output a touch strategy optimization time window, including: Extract the motion intensity coefficient and the foreground application package name from the scene state feature sequence; When the intensity coefficient of the motion is greater than the calibrated motion threshold and the foreground application package name belongs to the navigation application category, it is determined to be the entry point of the driving scene; Starting from the entry point of the driving scene, the scene state feature subsequence of subsequent time nodes is continuously scanned. The scene state feature subsequence is a continuous segment extracted from the scene state feature sequence. The starting time of the extraction is the entry point of the driving scene, and the ending time of the extraction is dynamically extended as the scanning process progresses. When the duration of time during which the intensity of motion is greater than the calibrated motion threshold reaches the dynamic duration threshold, the interval covered by the duration is set as the stable interval of the driving scenario. Based on the end point of the stable interval of the driving scenario, the motion intensity coefficient of subsequent time nodes is continuously monitored. When the motion intensity coefficient is less than or equal to the calibrated motion threshold and the duration reaches the exit confirmation time, it is determined as the exit point of the driving scenario. The time interval from the start time of the stable driving scenario to the exit point of the driving scenario is set as the touch strategy optimization time window; Based on the scene state feature subsequence and user historical interaction response data within the optimized touch strategy time window, a dynamic balance is achieved between touch interaction complexity and information priority, generating an initial touch strategy parameter set matching the driving scenario, including: Extract the motion intensity coefficient, foreground application package name, and notification arrival record from the scene state feature subsequence within the time window of the touch strategy optimization; Based on the interaction records within the historical time interval, the user's historical interaction response data is obtained. The user's historical interaction response data includes the response time consumed by the user when viewing each notification source at a historical time node, the records of accidental touch operations that occurred during the historical interaction process, and the corresponding intensity coefficient of movement at that time. Based on the foreground application package name, retrieve from the user's historical interaction response data the viewing time and operation error rate of different notification sources under the application running state corresponding to the foreground application package name; Based on the motion intensity coefficient, retrieve the probability of accidental touches by the user for different touch gestures within the motion intensity range corresponding to the motion intensity coefficient from the user's historical interaction response data; The viewing time, operation error rate, and accidental touch probability are normalized and then weighted and fused to obtain the comprehensive interaction cost of each notification source; Based on the comprehensive interaction cost, the notification sources are prioritized to obtain the priority ranking result; Based on the priority sorting results and the accidental touch probability, an initial touch strategy parameter set is generated; Based on the scene state feature sequence, the motion intensity coefficient and ambient light change coefficient of the current scene are extracted, and the initial touch strategy parameter set is dynamically compensated to obtain a corrected touch strategy parameter set, including: Extract the motion intensity coefficient and ambient light change coefficient at the current time point from the scene state feature sequence; Based on the motion intensity coefficient, the information display time in the initial touch strategy parameter set is first compensated to obtain the compensated information display time. Based on the ambient light variation coefficient, a second compensation is performed on the compensated information display time to obtain the corrected information display time; Based on the motion intensity coefficient, a third compensation is performed on the interaction mode parameters in the initial touch strategy parameter set to obtain the corrected interaction mode parameters. The correction information display time and the correction interaction mode parameters are integrated into a correction touch strategy parameter set; The touch response mechanism of the LCD screen of the wearable device is optimized and controlled according to the optimized time window of the touch strategy and the modified touch strategy parameter set.

2. The LCD touch strategy optimization method based on usage scenario recognition according to claim 1, characterized in that, Acquire inertial sensor data, ambient light sensor data, and currently running application status data from the wearable device, perform scene feature fusion, and output a scene status feature sequence, including: The inertial sensor data includes at least three-axis acceleration data and three-axis angular velocity data; the ambient light sensor data includes at least time-series values ​​of light intensity; and the application status data includes at least the foreground application package name and notification arrival records. Extract the resultant acceleration amplitude from the triaxial acceleration data, calculate the variance of the resultant acceleration amplitude on the time axis, and obtain the acceleration fluctuation amount; The angular velocity change rate is extracted from the triaxial angular velocity data, and the number of times the angular velocity direction reverses per unit time is counted to obtain the angular velocity turning frequency. The acceleration fluctuation and angular velocity turning frequency are normalized respectively to obtain normalized values ​​of acceleration fluctuation and angular velocity turning frequency. The normalized values ​​of acceleration fluctuation and angular velocity turning frequency are weighted and fused to obtain the motion intensity coefficient. The difference in light intensity between adjacent time nodes is extracted from the time-series values ​​of light intensity to obtain the magnitude of light intensity variation. The frequency of light intensity change is obtained by counting the number of times the direction of light intensity change reverses per unit time. The amplitude of the light intensity change and the frequency of the light intensity change are normalized to obtain normalized values ​​of the amplitude of the light intensity change and the frequency of the light intensity change. The normalized values ​​of the light intensity variation amplitude and the normalized values ​​of the light intensity variation frequency are weighted and fused to obtain the ambient light variation coefficient. The motion intensity coefficient, ambient light change coefficient, foreground application package name, and notification arrival record are time-aligned and fused to generate a scene state feature sequence.

3. The LCD touch strategy optimization method based on usage scenario recognition according to claim 2, characterized in that, The frequency of light intensity changes is obtained by counting the number of times the direction of change of light intensity reverses per unit time, including: Extract the first, second, and third light intensity values ​​at three consecutive time points from the light intensity time series values; Calculate the first difference between the second light intensity value and the first light intensity value, and calculate the second difference between the third light intensity value and the second light intensity value; When the product of the first difference and the second difference is negative, it is determined that a reversal of the direction of light intensity change has occurred at the second light intensity value; By iterating through the time series values ​​of light intensity, and accumulating the total number of times the direction of light intensity change reverses per unit time, the frequency of light change is obtained.

4. The LCD touch strategy optimization method based on usage scenario recognition according to claim 1, characterized in that, The calculation steps for the dynamic duration threshold include: Obtain the motion intensity coefficient at the current time point; Calculate the ratio of the calibrated motion threshold to the current motion intensity coefficient, and multiply the ratio by the baseline duration threshold to obtain the initial value of the dynamic duration; The reference duration threshold is the standard duration used to confirm a stable state in a driving scenario when the motion intensity coefficient is equal to the calibrated motion threshold. The smaller value between the initial dynamic duration value and the maximum allowed duration threshold value is set as the upper limit of the dynamic duration value; The larger of the dynamic duration upper limit and the minimum allowed duration threshold is set as the dynamic duration threshold; The maximum allowed duration threshold is a first multiple of the baseline duration threshold, and the minimum allowed duration threshold is a second multiple of the baseline duration threshold, wherein the first multiple is greater than the second multiple.

5. The LCD touch strategy optimization method based on usage scenario recognition according to claim 1, characterized in that, Based on the priority ranking result and the accidental touch probability, an initial touch strategy parameter set is generated, including: Based on the priority ranking result, the top N notification sources with the lowest overall interaction cost are selected from all notification sources and set as the high-priority notification set. Here, N is determined according to the total number of notification sources in the notification arrival record at the current time node and is inversely proportional to the probability of accidental touch. Information display time is generated based on the number of notification sources in the high-priority notification set, wherein the information display time is the duration of a single display of each notification source in the high-priority notification set on the lock screen interface, and the information display time is inversely proportional to the number of notification sources in the high-priority notification set; Based on the accidental touch probability, interaction mode parameters are generated, wherein the interaction mode parameters include the sensitivity adjustment coefficient of sliding touch and the response area scaling coefficient of click touch. The sensitivity adjustment coefficient is obtained by multiplying the sensitivity reference value by a first adjustment factor, which is proportional to the difference between the normalized value of the false touch probability. The response region scaling factor is obtained by multiplying the response region baseline value by a second adjustment factor, which is proportional to the sum of the normalized values ​​of the false touch probability. The information display time and the interaction mode parameters are integrated into an initial touch strategy parameter set.

6. The method for optimizing LCD touch strategies based on usage scenario recognition according to claim 1, characterized in that, Based on the motion intensity coefficient, a third compensation is performed on the interaction mode parameters in the initial touch strategy parameter set, including: Obtain the current sensitivity adjustment coefficient and response area scaling coefficient from the initial touch strategy parameter set; Based on the ratio of the exercise intensity coefficient to the calibrated exercise threshold, a sensitivity compensation factor and a response region compensation factor are calculated respectively, wherein the sensitivity compensation factor is inversely proportional to the ratio, and the response region compensation factor is directly proportional to the ratio. Multiply the sensitivity adjustment coefficient by the sensitivity compensation factor to obtain the corrected sensitivity adjustment coefficient, and the corrected sensitivity adjustment coefficient is not less than the lower sensitivity threshold. Multiply the response region scaling factor by the response region compensation factor to obtain the corrected response region scaling factor, and the corrected response region scaling factor is not greater than the upper limit threshold of the scaling factor. The modified sensitivity adjustment coefficient and the modified response region scaling coefficient are integrated into a modified interaction mode parameter.

7. A liquid crystal display touch strategy optimization system based on usage scenario recognition, characterized in that, The method for optimizing the touch strategy of a liquid crystal display screen based on usage scenario recognition as described in any one of claims 1-6 includes: The data acquisition module is used to acquire inertial sensor data, ambient light sensor data, and current application status data of the wearable device, and to perform scene feature fusion to output a scene status feature sequence. The feature analysis module is used to perform steady-state transition analysis on the scene state feature sequence of the driving scenario, identify and extract the continuous interaction state under the driving scenario, and output the touch strategy optimization time window, including: Extract the motion intensity coefficient and the foreground application package name from the scene state feature sequence; When the intensity coefficient of the motion is greater than the calibrated motion threshold and the foreground application package name belongs to the navigation application category, it is determined to be the entry point of the driving scene; Starting from the entry point of the driving scene, the scene state feature subsequence of subsequent time nodes is continuously scanned. The scene state feature subsequence is a continuous segment extracted from the scene state feature sequence. The starting time of the extraction is the entry point of the driving scene, and the ending time of the extraction is dynamically extended as the scanning process progresses. When the duration of time during which the intensity of motion is greater than the calibrated motion threshold reaches the dynamic duration threshold, the interval covered by the duration is set as the stable interval of the driving scenario. Based on the end point of the stable interval of the driving scenario, the motion intensity coefficient of subsequent time nodes is continuously monitored. When the motion intensity coefficient is less than or equal to the calibrated motion threshold and the duration reaches the exit confirmation time, it is determined as the exit point of the driving scenario. The time interval from the start time of the stable driving scenario to the exit point of the driving scenario is set as the touch strategy optimization time window; The parameter generation module is used to dynamically balance touch interaction complexity and information priority based on the scene state feature subsequence and user historical interaction response data within the time window of the touch strategy optimization, and to generate an initial touch strategy parameter set that matches the driving scenario, including: Extract the motion intensity coefficient, foreground application package name, and notification arrival record from the scene state feature subsequence within the time window of the touch strategy optimization; Based on the interaction records within the historical time interval, the user's historical interaction response data is obtained. The user's historical interaction response data includes the response time consumed by the user when viewing each notification source at a historical time node, the records of accidental touch operations that occurred during the historical interaction process, and the corresponding intensity coefficient of movement at that time. Based on the foreground application package name, retrieve from the user's historical interaction response data the viewing time and operation error rate of different notification sources under the application running state corresponding to the foreground application package name; Based on the motion intensity coefficient, retrieve the probability of accidental touches by the user for different touch gestures within the motion intensity range corresponding to the motion intensity coefficient from the user's historical interaction response data; The viewing time, operation error rate, and accidental touch probability are normalized and then weighted and fused to obtain the comprehensive interaction cost of each notification source; Based on the comprehensive interaction cost, the notification sources are prioritized to obtain the priority ranking result; Based on the priority sorting results and the accidental touch probability, an initial touch strategy parameter set is generated; The parameter compensation module is used to extract the motion intensity coefficient and ambient light change coefficient of the current scene based on the scene state feature sequence, and to dynamically compensate the initial touch strategy parameter set to obtain a corrected touch strategy parameter set, including: Extract the motion intensity coefficient and ambient light change coefficient at the current time point from the scene state feature sequence; Based on the motion intensity coefficient, the information display time in the initial touch strategy parameter set is first compensated to obtain the compensated information display time. Based on the ambient light variation coefficient, a second compensation is performed on the compensated information display time to obtain the corrected information display time; Based on the motion intensity coefficient, a third compensation is performed on the interaction mode parameters in the initial touch strategy parameter set to obtain the corrected interaction mode parameters. The correction information display time and the correction interaction mode parameters are integrated into a correction touch strategy parameter set; The control execution module is used to optimize and control the touch response mechanism of the LCD screen of the wearable device according to the optimized time window and the modified touch strategy parameter set.

Images