Standby picture processing method of Micro LED display screen

Abstract

The invention provides a standby picture processing method for a Micro LED display screen, and the method comprises the steps: separating overlapped tracks through a clustering algorithm if a track overlapping analysis result shows that the tracks of a plurality of touch points are overlapped, calculating the track overlapping degree and the separation confidence coefficient, and obtaining a touch track separation result; combining the pressure mutation detection result with the coordinate change of the touch point, calculating a touch offset through weighted average to obtain a touch offset correction parameter, and adjusting a touch coordinate mapping relation according to the correction parameter; recognizing a user touch posture mode by adopting a posture recognition technology through a touch offset correction result in combination with the change of the finger contact angle and the contact area of the user, extracting finger posture feature parameters, and generating user touch posture data; and according to the touch offset correction result and the pressure mutation detection result, combining with the finger posture characteristic parameters, modeling personalized touch habits of the user through posture analysis, and obtaining touch sensitivity adjustment parameters.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a standby screen processing method of a Micro LED display screen. Background Art

[0002] In the field of modern display technology, micro-LED displays, due to their high brightness, high contrast, and low power consumption, have become a key development direction for intelligent interactive devices. They are particularly crucial in the interactive design of standby screens, where optimizing touch guidance directly impacts the user experience and the level of device intelligence. However, current touch interaction methods for standby screens still have significant shortcomings. Many systems struggle to handle complex touch behaviors, prone to slow responses or misjudgments, resulting in frustrating user operations and a diminished user experience. Specifically, touch interaction in standby mode presents unique technical challenges. The primary issue lies in the complexity of user touch intent. When users perform multiple touches rapidly in a short period of time, touch traces often overlap, making it difficult for the system to accurately discern the user's true intent. This confusion can further trigger changes in user behavior. For example, adjusting the touch rhythm by long-pressing can lead to sudden changes in touch pressure. This sudden change in pressure can easily cause the touch position to shift, leading to a series of false touches. These intertwined technical factors make touch parsing in the standby screen extremely complex, requiring rapid identification of user intent while avoiding operational confusion caused by misjudgments. Therefore, how to dynamically analyze touch trajectories, map pressure changes, and analyze user postures in the standby screen to generate adaptive guidance animations to improve touch accuracy and smoothness has become a key issue that needs to be solved urgently. Summary of the Invention

[0003] The present invention provides a standby screen processing method for a Micro LED display, which mainly includes: When a user's finger touches the surface of the Micro LED display, the system switches from the standby screen to the active screen and displays the touch response interface. It also acquires multi-touch data, analyzes the coordinate position and timestamp information of each touch point, identifies the overlapping area of ​​different finger trajectories, and obtains the trajectory overlap analysis results. If the track overlap analysis results show that several touch point tracks overlap, the overlapping tracks are separated by a clustering algorithm, and the track overlap degree and separation confidence are calculated to obtain the touch track separation result; If the track overlap analysis results show that there are multiple touch point tracks overlapping, the position density distribution of each touch point is analyzed, adjacent touch points are identified and classified as the same finger operation, and the overlap degree and separation reliability between each track are calculated to obtain the touch track separation result; Combine the pressure mutation detection results with the touch point coordinate changes, calculate the touch offset through weighted average, obtain the touch offset correction parameters, and adjust the touch coordinate mapping relationship according to the correction parameters; By using the touch offset correction results and combining the changes in the user's finger contact angle and contact area, gesture recognition technology is used to identify the user's touch gesture pattern, extract the finger gesture feature parameters, and generate user touch gesture data; Based on the touch offset correction results and pressure mutation detection results, combined with the finger posture characteristic parameters, the user's personalized touch habits are modeled through posture analysis to obtain the touch sensitivity adjustment parameters; Generate adaptive guidance animation data based on the current interactive scenario requirements, combine touch sensitivity adjustment parameters, and generate interactive response data by optimizing the animation playback rhythm and the priority order of user operation response. When the user completes the touch operation, the system maintains the current screen display status.

[0004] The technical solution provided by the embodiment of the present invention may have the following beneficial effects: The present invention discloses a standby screen processing method for a Micro LED display screen, which realizes the accurate recognition of complex touch operations through multi-touch data analysis and trajectory overlap parsing. When the user's finger touches the display screen, the present invention can accurately separate the overlapping trajectories, and combine pressure mutation detection and coordinate change to calculate the touch offset for correction. At the same time, the present invention can also identify the user's touch posture, extract finger posture feature parameters, and model personalized touch habits, thereby optimizing touch sensitivity. Finally, the present invention generates an adaptive guidance animation according to the needs of the interactive scene, and optimizes the interactive response in combination with touch parameters to provide a smooth and accurate touch experience. This intelligent touch technology significantly improves the recognition accuracy and operation sensitivity of multi-touch, providing strong support for complex touch interaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] Figure 1 Flowchart of a standby screen processing method of a Micro LED display screen according to the present invention.

[0006] Figure 2 Schematic diagram of a standby image processing method of a Micro LED display screen according to the present invention. DETAILED DESCRIPTION

[0007] To further understand the content of the present invention, the present invention is described in detail with reference to the accompanying drawings and examples. The present application is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are intended only to illustrate the relevant invention and are not intended to limit the invention. It should also be noted that, for ease of description, only portions relevant to the invention are shown in the accompanying drawings.

[0008] like Figure 1-2 In this embodiment, a method for processing a standby image of a Micro LED display screen may specifically include: In step S101, when a user's finger touches the surface of the Micro LED display, the screen switches from the standby screen to the active screen and displays the touch response interface. At the same time, multi-touch data is acquired, the coordinate position and timestamp information of each touch point are analyzed, and the overlapping area range of different finger tracks is identified to obtain the track overlap analysis result.

[0009] When a user's finger touches the MicroLED display surface, the touch signal is detected and the pressure and contact area of ​​the initial contact point are immediately recorded. The interface switching process is also initiated, and the rendering process from the idle screen to the active screen is completed within a preset response time threshold. Once the active screen is displayed, the corresponding touch response interface is loaded based on the pressure and contact area. The sampling frequency is determined based on the pressure and contact area of ​​the initial contact point. Raw capacitance change data for all touch points on the display is acquired. The capacitance change is calculated by subtracting the capacitance values ​​of adjacent sampling points. Valid touch points are identified based on a preset capacitance threshold. For each valid touch point, its X- and Y-axis position in the screen coordinate system is extracted and the corresponding system timestamp is recorded, forming a touch data sequence containing position and time information. Based on the continuity of timestamps and the spatial proximity of coordinate points in the touch data sequence, a clustering method based on Euclidean distance is used to group touch points belonging to the same finger. For each touch point sequence, the least squares method is used to fit the equation for the finger's movement trajectory, and the set of boundary points of the area swept by each trajectory curve on the two-dimensional plane is calculated. For the boundary point sets of different finger trajectories, the intersection area is identified by judging whether the boundary points are inside other trajectory areas. If there is an intersection, all points located inside two or more trajectory areas are extracted as the overlapping area. The geometric center point coordinates and area values ​​of the overlapping area are calculated. The overlap ratio is obtained by dividing the overlapping area by the area of ​​each trajectory area. Combined with the overlap duration of the trajectories in the time dimension, the trajectory overlap analysis result including the overlap area range and overlap ratio is obtained.

[0010] Specifically, during the multi-touch interaction process of the MicroLED display, the acquisition of pressure value and contact area is the basis of the entire touch response system.

[0011] Specifically, when a finger touches the screen surface, a change in capacitance occurs in the contact area, with the intensity of this change proportional to the pressure applied by the finger. Pressure values ​​typically range from 0 to 1024, with a light touch likely producing a value of around 200 and a heavy press potentially exceeding 800. The contact area is calculated by calculating the number of pixels at the effective touch point. The average adult fingertip's contact area is approximately 100 to 150 pixels. These two parameters together determine the sensitivity and accuracy of the subsequent interface response.

[0012] It should be noted that the capacitance change is calculated using the differential method of adjacent sampling points because the ambient capacitance may have basic noise.

[0013] For example, if the capacitance value of a touch point at time t1 is 520 and the capacitance value at time t2 is 680, the capacitance change is 160. When this change exceeds a preset threshold, such as 100, it is determined to be a valid touch. This differential method can effectively eliminate environmental interference and improve the accuracy of touch detection.

[0014] In one possible implementation, the process of attributing discrete touch points to different fingers using a clustering method based on Euclidean distance needs to take ergonomic characteristics into consideration. The movement distance of the same finger between two adjacent sampling moments usually does not exceed 50 pixels, so the distance threshold is set to this value. If two touch points are adjacent in time and the spatial distance is less than the threshold, they are classified as the same finger trajectory. This clustering method is particularly suitable for fast sliding operations and can accurately distinguish the independent trajectories of multiple fingers. The process of fitting the trajectory curve using the least squares method is achieved by minimizing the sum of the squares of the vertical distances from all touch points to the fitting curve.

[0015] For example, a finger draws an arc on the screen with 20 sampling points. By constructing and differentiating the error function, we can obtain the best-fit quadratic curve equation. This fitting method can smooth out jitter in the original touch data, resulting in a smoother trajectory curve.

[0016] Preferably, the overlap region is identified using a ray method to determine whether a point is inside a polygon. Rays are emitted from the point to be determined in any direction, and the number of intersections between the ray and the trajectory boundary is counted. An odd number of intersections indicates the point is inside, while an even number indicates it is outside. When the boundary point sets of two trajectories contain points that are mutually inclusive, these points constitute an overlap region. Calculating the overlap ratio quantifies the degree of interaction between different finger trajectories, providing an important basis for subsequent gesture recognition.

[0017] In step S102 , if the trajectory overlap analysis result shows that several touch point trajectories overlap, the overlapping trajectories are separated by a clustering algorithm, and the trajectory overlap degree and separation confidence are calculated to obtain a touch trajectory separation result.

[0018] If the trajectory overlap analysis results indicate that several touch point trajectories overlap, the timestamp and coordinate sequences of each touch point within the overlapping region are obtained. The spatiotemporal distance value is calculated by calculating the Euclidean distance between two touch points and multiplying it by the normalization coefficient of the time difference. A touch point similarity matrix is ​​constructed based on the spatiotemporal distance values ​​between all pairs of touch points. Each element in the matrix represents the degree of similarity between two touch points. The number of clusters, K, is determined based on the number of connected components in the touch point similarity matrix. The touch point coordinates within the overlapping region are grouped using the K-means clustering algorithm. The input is the 2D coordinate sequence of each touch point. The K cluster center positions are iteratively updated until the sum of the distances from each point to its cluster center remains constant. The cluster label and cluster center coordinates of each touch point are output. For each cluster label and cluster center coordinate, the average Euclidean distance from all touch points within each cluster to the cluster center is calculated as the cluster closeness. The minimum Euclidean distance between cluster centers is calculated as the cluster separation. The cluster separation confidence value is calculated by dividing the cluster separation by the cluster closeness. The separation confidence value is compared with the preset confidence threshold. If the separation confidence is higher than the threshold, the touch point sequences with the same cluster label are recombined into independent tracks. The track overlap is calculated as the ratio of the number of touch points contained in each independent track to the total number of touch points in the original overlapping area. The touch track separation result is output, which includes the coordinate sequence of each independent track and the corresponding track overlap.

[0019] Specifically, the calculation of spatiotemporal distance values ​​is the key basis for separating overlapping trajectories.

[0020] Specifically, when two touch points are 20 pixels apart in space and 0.1 seconds apart in time, the differences in these two dimensions need to be considered comprehensively. Spatial distance is calculated directly using Euclidean distance, while the temporal difference needs to be normalized. The normalization coefficient is typically set to the inverse of the sampling period to make the temporal and spatial dimensions comparable.

[0021] For example, if the sampling period is 0.016 seconds, the normalization coefficient is 62.5, and the time difference of 0.1 seconds is normalized to 6.25, which is multiplied by the spatial distance of 20 to obtain a spatiotemporal distance value of 125. This calculation method can effectively distinguish between fast-moving trajectories and slow-moving trajectories.

[0022] It's important to note that the connected components in the touch point similarity matrix reflect the natural grouping of trajectories. A connected component is a set of interconnected elements in the matrix. When the similarity between two touch points exceeds a threshold, they are considered connected in the matrix. By traversing the matrix using a depth-first search, all connected components can be identified, each corresponding to an independent finger trajectory. This approach avoids the subjectivity of manually setting the number of clusters, making the determination of the K value more objective and accurate.

[0023] In one possible implementation, the iterative process of the K-means clustering algorithm includes several key steps. Initially, K touch points are randomly selected as cluster centers. The Euclidean distance from each touch point to each cluster center is calculated, and the touch point is assigned to the nearest cluster center. The center position of each cluster is recalculated as the average of the coordinates of all points within that cluster. The assignment and update process is repeated until the cluster center positions no longer change significantly. This iterative optimization process can find a locally optimal clustering solution. The calculation of cluster compactness and cluster separation provides quantitative evaluation criteria for the separation results.

[0024] For example, a cluster contains 15 touch points, and the distances from these points to the cluster center are 8, 12, 10, and so on, with an average distance of 10 pixels. This is the compactness of the cluster. The distance between the cluster center and the nearest cluster center is 50 pixels, which is the cluster separation. The separation confidence is calculated as 50 divided by 10, which equals 5. This indicates that the separation between clusters is five times greater than the compactness within the clusters, indicating good separation.

[0025] Preferably, the track overlap calculation quantifies the contribution of each track to the overlapping area. For example, if there are 100 touch points in the overlapping area, and after separation, the first track contains 65 points and the second track contains 35 points, the overlap of the two tracks is 0.65 and 0.35, respectively. This quantitative metric not only reflects the relative importance of the tracks but also provides a weighting reference for subsequent gesture recognition and interactive response, allowing the system to more accurately understand the user's touch intent.

[0026] In step S103, if the trajectory overlap analysis result shows that there are multiple touch point trajectories overlapping, the position density distribution of each touch point is analyzed, adjacent touch points are identified and classified as the same finger operation, and the overlap degree and separation reliability between each trajectory are calculated to obtain the touch trajectory separation result.

[0027] If the trajectory overlap analysis results indicate that multiple touch point trajectories overlap, the two-dimensional coordinates of each touch point within the overlapping area are obtained. The display area is divided into a preset grid size, and the number of touch points within each grid cell is counted. The position density value of each grid cell is calculated by dividing the number of touch points by the grid area, forming a position density distribution matrix. Based on the position density distribution matrix, the sum of the absolute values ​​of the density differences between each grid cell and its eight adjacent grid cells is calculated as the density gradient value of that grid cell. Grids with gradient values ​​below a preset threshold and density values ​​above a preset threshold are marked as density peak centers. Starting from each density peak center, adjacent grids with density values ​​above the threshold are gradually incorporated into the same area until a grid with density values ​​below the threshold is encountered. This results in multiple independent high-density regions and the touch point sets contained therein. For each touch point set within a high-density region, the Euclidean distance between any two touch points within the set is calculated. If the distance is less than a preset proximity threshold, a connection relationship is established between the two points. A depth-first search is performed to traverse all connection relationships, and touch points that can be reached through these connections are grouped into the same touch point group. Each touch point group corresponds to an independent trajectory. According to the attribution of each touch point group, the number of grids located at the intersection of two or more high-density areas is counted as the overlap value between the trajectories. The standard deviation of the distances from all points in each touch point group to the center point of the group is calculated as the spatial dispersion. The separation reliability is obtained by dividing the minimum distance between different groups by the average spatial dispersion. The coordinate sequence of each touch point group and the corresponding overlap value and separation reliability value are output as the touch trajectory separation result.

[0028] Specifically, the construction of the position density distribution matrix is ​​the basis for identifying overlapping trajectories.

[0029] Specifically, when the display is divided into a 10×10 pixel grid, each grid cell becomes a statistical unit. If a grid contains five touch points and the grid area is 100 square pixels, the position density value of that grid is 0.05. This gridding method converts discrete touch points into a continuous density field, facilitating subsequent region identification and analysis. The density value directly reflects the concentration of touch activity.

[0030] It should be noted that the density gradient value is calculated using the eight-neighborhood difference method. For any grid, it is necessary to calculate the density difference between it and its eight neighboring grids, and then add up the absolute values ​​of these eight differences.

[0031] For example, if the center grid density is 0.08 and the surrounding eight grid densities are 0.02, 0.03, 0.07, 0.06, 0.02, 0.01, 0.05, and 0.04, respectively, the density gradient is 0.34. A larger gradient indicates that the location is at the boundary of density variation; a smaller gradient indicates that the location is in a uniform density region.

[0032] In one possible implementation, the region growing process starting from the center of the density peak is similar to water wave diffusion. First, grids with gradient values ​​less than 0.1 and density values ​​greater than 0.05 are identified as seed points. These points are usually located in the center area of ​​the finger contact. Then the adjacent grids of the seed point are checked. If the density value of the adjacent grid is still higher than the threshold of 0.03, it is included in the same area. This process continues until a grid with a density value lower than the threshold is encountered. In this way, the high-density touch area formed by each finger can be accurately identified. Depth-first search plays a key role in judging the connectivity of touch points. When the Euclidean distance between two touch points is less than 15 pixels, they are considered to be connected. Starting from any touch point, all reachable touch points are visited along the connection relationship. These points constitute a connected component.

[0033] For example, touch points A and B are 12 pixels apart, and B and C are 14 pixels apart, but A and C are 28 pixels apart. Through depth-first search, A, B, and C are still classified into the same group because they form a connection chain through B.

[0034] Preferably, the calculation of spatial dispersion reflects the degree of clustering of the touch point group. First, the geometric center of all touch points in the group is calculated, and then the distance from each point to the center is calculated. The standard deviation of these distances is the spatial dispersion. A small dispersion indicates that the touch points are tightly clustered, usually corresponding to a stable touch of a single finger; a large dispersion may be a complex trajectory formed by a fast sliding finger or multiple fingers crossing. The separation reliability is measured by the ratio of the minimum distance between groups to the average dispersion. The larger the ratio, the better the separation effect of different trajectories and the higher the accuracy of touch recognition.

[0035] Step S104 : combining the pressure mutation detection result with the touch point coordinate change, calculating the touch offset by weighted average, obtaining a touch offset correction parameter, and adjusting the touch coordinate mapping relationship according to the correction parameter.

[0036] The pressure value sequence for each sampling moment during the touch process is acquired, and the difference between adjacent pressure values ​​is calculated. If the difference exceeds a preset mutation threshold, it is marked as a pressure mutation point. The timestamp of the mutation point and the corresponding pressure change amplitude are recorded to form the pressure mutation detection result. Based on the timestamp in the pressure mutation detection result, the coordinates of the touch point at the mutation moment and the previous moment are extracted. The difference between the two coordinates is calculated to obtain a coordinate offset vector. The pressure change amplitude is divided by the preset maximum pressure value to obtain a normalization coefficient as a weight coefficient. The weight coefficient is multiplied by the coordinate offset vector to obtain a weighted offset vector for the mutation point. For the weighted offset vectors of all pressure mutation points, the arithmetic mean of the horizontal component and the arithmetic mean of the vertical component are calculated. These two averages are combined into a two-dimensional vector as the overall touch offset, which is the touch offset correction parameter. Based on the touch offset correction parameter, the horizontal component of each original touch coordinate is subtracted from the horizontal component, and the vertical component of the correction parameter is subtracted from the vertical component to obtain the corrected touch coordinates. A linear transformation relationship is established from the original coordinates to the corrected coordinates as the adjusted touch coordinate mapping relationship.

[0037] Specifically, pressure mutation detection plays a key role in touch accuracy correction.

[0038] Specifically, when a user's finger suddenly switches from light touch to heavy pressure, the pressure value may increase dramatically from 200 units to 800 units. This 600-unit change far exceeds the normal gradual range of touch. The pressure mutation threshold is typically set at three times the statistical average of the pressure difference between adjacent sampling points, which is approximately 150 units. When the pressure difference exceeds this threshold, the system records that moment as the mutation point. This mutation is often accompanied by a sharp change in the finger's contact area, causing the touch center point to shift.

[0039] It should be noted that the calculation of the coordinate offset vector needs to accurately capture the position changes before and after the pressure mutation.

[0040] For example, at time t1, the pressure is 200, and the coordinates are (100, 150). At time t2, the pressure suddenly changes to 800, and the coordinates become (105, 148). The coordinate offset vector is (5, -2), indicating that the touch point is offset 5 pixels to the right and 2 pixels upward. This offset is mainly because as finger pressure increases, the contact area expands, and the touch center calculated by the system changes accordingly.

[0041] In one possible implementation, the weight coefficient is determined using a normalization method to reflect the relative degree of pressure change. Assuming the system's preset maximum pressure value is 1024 units, when the pressure change amplitude is 600 units, the normalization coefficient is 0.586. This coefficient is used as a weight and multiplied by the coordinate offset vector (5, -2) to obtain the weighted offset vector (2.93, -1.17). The introduction of the weight coefficient reduces the impact of sudden changes in pressure on the final correction parameters, while sudden changes in pressure with greater force have a greater weight. The calculation of the touch offset correction parameters uses statistical averaging to eliminate the influence of individual outliers.

[0042] For example, the system detected five pressure mutation points, with weighted offset vectors of (2.93, -1.17), (3.2, -1.5), (2.8, -1.0), (3.1, -1.3), and (2.9, -1.2). The average horizontal component is 2.99, and the average vertical component is -1.23. Combined, they yield the correction parameter (2.99, -1.23). This two-dimensional vector represents the average offset trend throughout the entire touch process.

[0043] Preferably, the establishment of a linear transformation relationship makes the touch coordinate correction process simple and efficient. For any original touch coordinate (x, y), the corrected coordinate is (x-2.99, y+1.23). This simple subtraction operation can be completed in real time without adding additional computational burden to the system. The corrected mapping relationship ensures that the position recognized by the system remains consistent when the user touches with different pressures, significantly improving the accuracy of touch positioning, especially in application scenarios that require precise operation.

[0044] Step S105 , using the touch offset correction result and the change in the user's finger contact angle and contact area, adopting gesture recognition technology to identify the user's touch gesture pattern, extracting finger gesture feature parameters, and generating user touch gesture data.

[0045] The touch offset correction results are used to obtain a corrected touch coordinate sequence. All boundary points of the touch area are identified, and the minimum circumscribed ellipse encompassing all boundary points is calculated. The major and minor axis lengths of the ellipse are extracted, and the angle between the major axis and the horizontal baseline of the display is calculated as the finger contact angle. The number of valid pixels in the touch area at each sampling moment is counted as the contact area, resulting in a contact angle sequence and a contact area sequence. Based on the contact angle sequence and the pressure values ​​of each point within the touch area, the weighted average coordinate of the pressure values ​​of all touch points is calculated as the pressure center of gravity position. The abscissa is equal to the sum of the products of the abscissas and the pressure values ​​of each point divided by the total pressure values. The ordinate is calculated in the same way. The finger tilt angle is obtained by taking the angle between the line connecting the pressure center of gravity and the ellipse center and the vertical direction. Based on the finger tilt angle values ​​and pressure center of gravity positions, the tilt angle is classified into preset angle intervals. The Euclidean distance between the pressure center of gravity and the ellipse center is calculated. If the distance exceeds a preset threshold, the touch gesture is determined to be tilted; otherwise, it is determined to be vertical, and the touch gesture category identifier is obtained. According to the touch posture category identifier and tilt angle value, the touch area is divided into multiple rectangular sub-areas according to the preset grid size. The average pressure value of all touch points in each sub-area is counted, and the pressure distribution vector is formed according to the spatial arrangement order of the sub-areas. Combined with the tilt angle value, touch posture category identifier and pressure distribution vector, user touch posture data containing finger posture feature parameters is generated.

[0046] Specifically, the calculation of the minimum circumscribed ellipse is a key step in identifying finger contact features.

[0047] Specifically, when a finger touches the screen at a certain angle, the touch area takes on an irregular elliptical shape. By scanning all boundary points of the touch area, the set of points farthest from the center can be found. These boundary points are fitted using the least squares method to determine an ellipse that contains all points and has the smallest area. The major axis of the ellipse usually extends along the longitudinal direction of the finger and is approximately 2 to 3 times the length of the minor axis. The angle between the major axis and the horizontal reference line directly reflects the tilt direction of the finger, and the range is usually between 0 and 90 degrees.

[0048] It should be noted that the calculation of the pressure center of gravity uses the concept of center of mass in physics. Each touch point can be regarded as a point mass, and its "mass" is the pressure value at that point.

[0049] For example, a touch area contains 100 touch points, with the coordinates of point i being (xi, yi) and the pressure value being pi. The horizontal coordinate of the pressure center is calculated as the sum of the products of all xi and pi divided by the sum of all pi. When the finger presses vertically, the pressure center roughly coincides with the geometric center of the ellipse. However, when pressing at an angle, the pressure is concentrated at the tip of the finger, causing the pressure center to deviate significantly from the geometric center.

[0050] In one possible implementation, the classification of touch gestures is based on the degree of eccentricity of the pressure distribution. The distance between the center of gravity of the pressure and the center of the ellipse reflects the unevenness of the pressure distribution. When this distance is less than 10% of the length of the minor axis of the ellipse, it can be considered as a uniform vertical press; when it exceeds 25%, it indicates that there is a significant tilt. The interval division of the tilt angle is usually 15 degrees as a gear, such as 0-15 degrees for a slight tilt, 15-30 degrees for a moderate tilt, and more than 30 degrees for a severe tilt. This classification method can adapt to the operating habits of different users. The construction of the pressure distribution vector requires the touch area to be gridded.

[0051] For example, the touch area is divided into a 4×4 grid, forming 16 sub-areas. Each sub-area calculates the average pressure of all touch points within it. If there are no touch points in a sub-area, the pressure value is recorded as 0. The 16 pressure values ​​are arranged into a one-dimensional vector from left to right and from top to bottom. With vertical touch, the pressure distribution is relatively uniform, with minimal variation among the elements in the vector. With tilted touch, the pressure is concentrated in a specific direction, and the vector exhibits a distinct gradient.

[0052] Preferably, the user touch posture data contains three core elements: the tilt angle value provides a quantitative description of the finger direction, the touch posture category identifier realizes fast pattern matching, and the pressure distribution vector retains the detailed pressure space characteristics. The multi-dimensional posture description method enables the system to accurately distinguish different touch intentions, such as distinguishing whether the user is performing precise clicks or quick sliding, thereby providing a more intelligent and personalized interactive experience.

[0053] Step S106 , based on the touch offset correction result and the pressure mutation detection result, combined with the finger posture characteristic parameters, the user's personalized touch habits are modeled through posture analysis to obtain touch sensitivity adjustment parameters.

[0054] Based on the touch offset correction results, the mean and standard deviation of the user's historical touch offset are calculated. The number of mutations per unit time is extracted from the pressure mutation detection results as the mutation frequency, and the average of the pressure difference before and after the mutation is used as the mutation amplitude. The occurrence ratio of each tilt angle interval in the finger posture feature parameters is calculated, and the offset mean, offset standard deviation, mutation frequency, mutation amplitude, and tilt angle ratio are combined to form a five-dimensional feature vector. The K-means clustering algorithm is used to classify touch patterns based on the five-dimensional feature vector. The number of clusters K is set to 3, corresponding to the three user types of light touch, standard touch, and heavy pressure. The current user's feature vector is input, and the cluster center is iteratively updated until convergence. The Euclidean distance between the feature vector and the three cluster centers is calculated, and the user is classified into the touch habit category represented by the cluster center with the smallest distance. Based on the touch habit category index value, the built-in parameter mapping relationship is queried to obtain the corresponding baseline parameter set. The baseline pressure for light touch is low, the response time is short, and the touch radius is small, while the opposite is true for heavy pressure. The sum of the absolute values ​​of the differences between the current user's feature vector and the cluster center of the category to which it belongs is calculated and divided by the number of feature vector dimensions to obtain the average deviation. The personalized adjustment coefficient is obtained by adding 1 to the normalized value of the average deviation. The baseline parameter set is modified using the personalized adjustment coefficient. The baseline pressure value is multiplied by the adjustment coefficient to obtain the pressure threshold suitable for the user. The baseline response time is divided by the adjustment coefficient to obtain a faster or slower response speed parameter. The baseline touch radius is multiplied by the adjustment coefficient to obtain the expanded or reduced touch area parameter. The touch sensitivity adjustment parameter containing the pressure threshold, response speed parameter, and touch area parameter is output.

[0055] Specifically, the construction of the five-dimensional feature vector reflects the multi-dimensional characteristics of user touch behavior.

[0056] Specifically, the mean offset reflects the systematic deviation of user touch. For example, the average historical offset of a user is 3.2 pixels, which means that their touch points are generally biased to the right. The standard deviation of the offset measures the stability of the touch. Users with a standard deviation of 1.5 have more stable touch than users with a standard deviation of 5.0. The mutation frequency is obtained by counting the number of times the pressure change exceeds the threshold per minute. Frequent mutations indicate that the user tends to click quickly. The mutation amplitude takes the average of the pressure difference in all mutation events, reflecting the intensity of the user's force. The tilt angle ratio records the frequency of occurrence of three intervals: 0-15 degrees, 15-30 degrees, and above 30 degrees, to form a portrait of the user's posture preference.

[0057] It should be noted that the application of the K-means clustering algorithm to touch habit classification is based on the natural grouping of user behavior. The algorithm first randomly selects three initial cluster centers, representing light-touch, standard-touch, and heavy-touch users, respectively. In each iteration, the Euclidean distance from all user feature vectors to each cluster center is calculated, and the user is assigned to the closest category. The mean of all user feature vectors within each category is then recalculated as the new cluster center. After multiple iterations, the cluster centers stabilize, forming a clear division into user types. Light-touch users typically exhibit a combination of low pressure, high frequency, and small offset.

[0058] In one possible implementation, the design of the parameter mapping relationship takes into account the operating characteristics of different user types. The baseline pressure threshold for light-touch users is set to a lower value, such as 100 units, because such users are accustomed to gentle touch; the response time is set to a shorter value, such as 30 milliseconds, to meet their fast operation needs; the touch radius is set to a smaller value, such as 20 pixels, to adapt to their precise clicking characteristics. On the contrary, the baseline pressure threshold for heavy-pressure users can reach 300 units, the response time is extended to 80 milliseconds to filter out false touches, and the touch radius is expanded to 40 pixels to accommodate their larger contact area. The calculation of the personalized adjustment coefficient realizes refined adaptation from category to individual.

[0059] For example, a user classified as standard has a eigenvector of [2.5, 1.8, 3.2, 180, 0.7], while the standard cluster center is [2.0, 2.0, 3.0, 200, 0.6]. The absolute values ​​of the differences in each dimension are 0.5, 0.2, 0.2, 20, and 0.1, respectively. Normalization is required to make the data of different dimensions comparable. After normalization, the sum of the differences is divided by 5, resulting in an average deviation of 0.15, which results in an adjustment factor of 1.15. This means that this user is slightly more stressed than standard.

[0060] Optimally, the parameter correction process ensures personalized optimization of the touch experience. Multiplying the standard baseline pressure of 200 by 1.15 yields a pressure threshold of 230, enabling the system to accurately identify the user's effective touch. The response time of 50 milliseconds divided by 1.15 yields 43 milliseconds, reducing false touches while ensuring response speed. The touch radius of 30 pixels multiplied by 1.15 yields 34.5 pixels, expanding the effective touch area. This dynamic adjustment mechanism enables each user to obtain the touch parameters that best suit their operating habits, significantly improving touch accuracy and comfort.

[0061] Step S107, based on the current interactive scene requirements, generates adaptive guidance animation data, combines the touch sensitivity adjustment parameters, and generates interactive response data by optimizing the animation playback rhythm and the priority order of user operation response. When the user completes the touch operation, the system maintains the current screen display state.

[0062] Based on the current interactive scenario, the user's operation stage is identified, the preset interface animation duration and animation type for this stage are obtained, the total duration of the animation playback is determined based on the number of interface elements, and the animation type and total duration are combined to generate guide animation data containing playback duration parameters. According to the playback duration parameters in the guide animation data, combined with the response speed value in the touch sensitivity adjustment parameters, the actual playback interval is obtained by dividing the playback duration by the response speed value. If the actual playback interval is less than the preset minimum interval, the minimum interval is used to determine the optimized animation playback rhythm. For the optimized animation playback rhythm, the time slice allocation ratio for user touch event processing is set to be higher than the time slice allocation ratio for animation rendering. The touch feedback strength value is calculated based on the ratio of the pressure threshold in the touch sensitivity adjustment parameters to the actual touch pressure. The current brightness of the interface element is obtained and multiplied by the touch feedback strength value to obtain the target highlight brightness. The display properties of the touched interface elements are updated through the target highlight brightness, the touch feedback intensity value is converted into a specific value within the preset vibration duration range, and a translucent circular confirmation icon is superimposed and displayed at the touch position. When the touch end signal is detected, the current display state of all interface elements is maintained, and interactive response data including the touch feedback intensity value, target highlight brightness and confirmation icon display parameters is output.

[0063] Specifically, the identification of interactive scenarios is the basis for generating appropriate guiding animations.

[0064] Specifically, when the user is in the application startup phase, the interface needs to display the brand logo and loading progress, and the animation duration at this time is usually set to 2-3 seconds. In the function selection phase, users need to quickly browse options, and the animation duration is shortened to 0.5-1 second. The number of interface elements directly affects the complexity of the animation. For example, when the main interface contains 20 icons, the gradual appearance animation of each icon takes 50 milliseconds, and the total duration is 1 second. This method of dynamic adjustment based on the scene avoids the monotony brought by fixed animation.

[0065] It's important to note that optimizing animation playback rhythm requires a balance between smoothness and responsiveness. The responsiveness value reflects the user's operating habits. A fast user might have a responsiveness value of 1.5, while a cautious user might only have a responsiveness value of 0.8. By dividing the playback duration by the responsiveness value, a personalized playback interval can be calculated.

[0066] For example, if the original playback duration is 1000 milliseconds and the response speed value is 1.5, the actual interval is 667 milliseconds. However, to prevent the animation from being too fast and causing visual confusion, the system sets a minimum interval of 500 milliseconds, so 667 milliseconds is ultimately used as the playback interval.

[0067] In one possible implementation, a time-slicing mechanism ensures prioritized touch responses. The system divides each rendering cycle into multiple time slices, with touch event processing occupying 70% and animation rendering occupying 30%. This allocation ensures timely response to user touch operations, even during complex animations. When a touch event is detected, the system immediately interrupts the current animation rendering, prioritizes touch logic processing, and resumes animation playback after processing is complete. The calculation of touch feedback intensity reflects refined pressure sensing. When the user's actual touch pressure is 400 units and their personalized pressure threshold is 200 units, the ratio is 2.0, indicating significant pressure. The system uses this ratio as the feedback intensity value to adjust various feedback effects. The current brightness of an interface element is typically between 0.6 and 0.8. Multiplying this by the feedback intensity value of 2.0 brings the target highlight brightness to a range of 1.2-1.6, creating a noticeable visual emphasis.

[0068] Preferably, the duration mapping of the vibration feedback adopts a piecewise linear approach. When the feedback intensity value is in the range of 0.5-1.0, the vibration duration is 10-30 milliseconds of mild feedback; the range of 1.0-2.0 corresponds to 30-80 milliseconds of moderate feedback; and more than 2.0 provides 80-150 milliseconds of strong feedback. The confirmation icon adopts a translucent design with a transparency set to 0.7 and a diameter of 1.5 times the touch area to ensure that the user can clearly see the operation results. When the touch ends, the system maintains the screen state by locking the rendering buffer, avoiding unexpected interface changes and providing users with a stable and reliable interactive experience.

[0069] The above description is only a specific implementation method of this specification. Those skilled in the art will clearly understand that for the convenience and brevity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here. It should be understood that the scope of protection of this specification is not limited to this. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed in this specification, and these modifications or replacements should be included in the scope of protection of this specification.

Claims

1. A method for processing a standby image of a Micro LED display, characterized in that: The method comprises: When a user's finger touches the surface of the Micro LED display, the device switches from the standby screen to the active screen, displays the touch response interface, obtains multi-touch data, analyzes the coordinate position and timestamp information of each touch point, identifies the overlapping area of ​​different finger trajectories, and obtains the trajectory overlap analysis result; If the track overlap analysis result shows that the touch point tracks overlap, the overlapping tracks are separated by a clustering algorithm, and the track overlap degree and separation confidence are calculated to obtain the touch track separation result; By analyzing the position density distribution of the touch points, identifying adjacent touch points, classifying them as the same finger operation, calculating the overlap degree and separation confidence between each track, and obtaining the touch track separation result; Combine the pressure mutation detection results with the touch point coordinate changes to calculate the touch offset, obtain the touch offset correction parameters, and adjust the touch coordinate mapping relationship; Through the touch offset correction results and the changes in finger contact angle and contact area, the user's touch posture pattern is identified, the finger posture feature parameters are extracted, and the user's touch posture data is generated; Combining the touch offset correction result, the pressure mutation detection result and the finger posture characteristic parameter, modeling the user's personalized touch habits to obtain a touch sensitivity adjustment parameter; Generate adaptive guidance animation data based on the current interactive scene requirements, combine the touch sensitivity adjustment parameters, optimize the animation playback rhythm and user operation response priority, generate interactive response data, and maintain the current screen display state.

2. The method for processing a standby image of a Micro LED display according to claim 1, wherein: When a user's finger touches the surface of the Micro LED display, the screen switches from the standby screen to the active screen, displays the touch response interface, obtains multi-touch data, analyzes the coordinate position and timestamp information of each touch point, identifies the overlapping area of ​​different finger tracks, and obtains track overlap analysis results, including: After detecting the touch signal, the pressure value and contact area of ​​the initial contact point are recorded, the interface switching program is started, the rendering from the standby screen to the active screen is completed, and the corresponding touch response interface is loaded; according to the pressure value and contact area of ​​the initial contact point, the sampling frequency is determined, the capacitance value change data of the touch point is obtained, the valid touch point is judged, and the screen coordinate position and time stamp of each valid touch point are extracted to form a touch data sequence; according to the timestamp continuity and coordinate space proximity of the touch data sequence, the touch points are classified using a clustering method, the movement trajectory curve of each group of touch points is fitted, and the boundary point set of the trajectory area is calculated; by judging whether the boundary point set is within other trajectory areas, the overlapping area is identified, the geometric center coordinates and area value of the overlapping area are calculated, and the trajectory overlap analysis result including the overlapping area range is obtained.

3. The method for processing a standby image of a Micro LED display according to claim 1, wherein: The touch track separation result is obtained by analyzing the position density distribution of the touch points, identifying adjacent touch points, classifying them as the same finger operation, calculating the overlap degree and separation confidence between each track, and including: Obtain the two-dimensional coordinates of the touch points in the overlapping area, divide the display screen area into grid sizes, count the number of touch points in each grid, calculate the position density value, and form a position density distribution matrix; based on the position density distribution matrix, calculate the density difference between each grid and its adjacent grids, mark the density peak center, and include adjacent high-density grids in the same area to obtain a touch point set in the high-density area; for the touch point set, calculate the Euclidean distance between the touch points, establish a connection relationship, and classify the touch point groups through a search algorithm, with each group corresponding to an independent trajectory; count the number of grids at the junction of the high-density area, calculate the spatial dispersion of the touch point group and the minimum distance between groups, obtain the overlap degree and separation confidence, and output the touch trajectory separation result.

4. The method for processing a standby image of a Micro LED display according to claim 1, wherein: If the track overlap analysis result shows that the touch point tracks overlap, separating the overlapping tracks using a clustering algorithm, calculating the track overlap degree and the separation confidence, and obtaining the touch track separation result, including: The coordinate sequence of the touch points in the overlapping area is obtained, the spatiotemporal distance values ​​between the touch points are calculated, and a similarity matrix is ​​constructed. The number of clusters is determined based on the similarity matrix, and the touch points are grouped using a clustering algorithm, and cluster labels and center coordinates are output. The average distance from the touch points corresponding to the cluster labels to the cluster centers and the minimum distance between the cluster centers are calculated to obtain the separation confidence. Based on the separation confidence, the touch point sequences with the same cluster label are combined, the trajectory overlap is calculated, and the touch trajectory separation result is output.

5. The method for processing a standby image of a Micro LED display according to claim 1, wherein: The method combines the pressure mutation detection result with the touch point coordinate change, calculates the touch offset, obtains the touch offset correction parameter, and adjusts the touch coordinate mapping relationship, including: Obtain a sequence of pressure values, calculate the pressure difference between adjacent moments, mark the pressure mutation point, and record the timestamp and pressure change amplitude; extract the coordinates of the pressure mutation point, calculate the coordinate offset vector, and calculate the weighted offset vector based on the pressure change amplitude; for the weighted offset vector, calculate the average of the horizontal and vertical components to obtain the touch offset; adjust the touch coordinates based on the touch offset to generate a corrected coordinate mapping relationship.

6. The method for processing a standby image of a Micro LED display according to claim 1, wherein: The method of identifying the user's touch gesture pattern, extracting the finger gesture feature parameters, and generating the user's touch gesture data by using the touch offset correction result and the finger contact angle and contact area changes includes: The corrected coordinate sequence is obtained through the touch offset correction result, the minimum circumscribed ellipse of the touch area boundary points is calculated, the lengths of the major axis and minor axis and the angle between the major axis and the touch area are extracted, the number of pixel points in the touch area is counted, and a contact angle sequence is obtained; based on the contact angle sequence and the pressure value, the pressure center position is calculated to determine the finger tilt angle; based on the finger tilt angle, the touch posture is classified, and user touch posture data containing posture feature parameters is generated.

7. The method for processing a standby image of a Micro LED display according to claim 1, wherein: The method of combining the touch offset correction result, the pressure mutation detection result and the finger posture characteristic parameter to model the user's personalized touch habits and obtain the touch sensitivity adjustment parameter includes: Based on the touch offset correction result, the offset statistics are calculated, the mutation frequency and amplitude are extracted from the pressure mutation detection result, the inclination angle distribution of the finger posture characteristic parameters is statistically analyzed, and a feature vector is generated; the feature vector is clustered to determine the touch habit category; based on the touch habit category, the parameter mapping relationship is queried, the baseline parameters are corrected, and the touch sensitivity adjustment parameters are output.

8. The method for processing a standby image of a Micro LED display according to claim 1, wherein: The method of generating adaptive guidance animation data based on the current interaction scene requirements, optimizing the animation playback rhythm and user operation response priority in combination with the touch sensitivity adjustment parameters, and generating interaction response data includes: Identify the operation phase of the interactive scenario and generate guidance animation data including animation duration and type; adjust parameters based on the guidance animation data and the touch sensitivity to determine the animation playback interval; set the touch event priority, calculate the touch feedback strength, update the interface element brightness, display a confirmation icon, and output the interactive response data.

Images