A high-precision mouse direction control method and system
By integrating mouse movement data, user operation frequency, and physiological state information, and using the Kalman filtering algorithm to optimize mouse control parameters, the problem of insufficient dynamic adaptation in existing technologies is solved, improving the accuracy and stability of mouse control and enhancing the user experience.
Patent Information
- Application Number
- CN202511742239.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-25
- Publication Date
- 2026-11-17
AI Technical Summary
Existing mouse control technology cannot dynamically adapt to user operation behavior and physiological state, and the deviation correction accuracy is insufficient, resulting in hypersensitive or sluggish response in high-precision operations, and inability to effectively identify and correct subtle directional deviations.
By acquiring mouse movement direction deviation data, user operation frequency data, and user status information, the Kalman filtering algorithm is used for smoothing and data optimization. Combined with the periodic characteristics of user operation frequency and fatigue index, the deviation correction weight is dynamically adjusted to generate a personalized control configuration file and optimize mouse control parameters in real time.
It achieves multi-dimensional data-driven deviation identification, significantly improving the positioning accuracy and deviation correction stability in high deviation areas, and enhancing control precision and user experience in precision operation scenarios.
Smart Images

Figure CN123114282A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of computer input device control technology, and in particular to a high-precision mouse direction control method and system. Background Technology
[0002] Currently, in modern human-computer interaction scenarios, the mouse serves as a core input device, and its control precision directly impacts the quality of task completion and user experience in scenarios such as design and drawing, precision operations, and e-sports competitions. With the increasing demand for high-precision operations, users are placing higher demands on the finesse, stability, and adaptability of mouse directional control. Motion perception, as the core element of the mouse in capturing operational intentions, directly determines the quality of control effectiveness.
[0003] Current mouse control technologies often rely on fixed parameter configurations or simple deviation correction logic for motion sensing, failing to adequately consider dynamic changes during user operations. On one hand, traditional solutions ignore the correlation between periodic fluctuations in user operation frequency and directional deviation, making it impossible to dynamically adjust control parameters based on high-frequency operations and frequent fine-tuning. On the other hand, they do not consider the impact of user physiological states (such as heart rate changes and fatigue levels) on operational accuracy, leading to issues like overly sensitive or sluggish mouse responses during prolonged use or when the user is not in a good state, making it difficult to match the user's real-time operating rhythm. Furthermore, existing technologies lack effective noise suppression and iterative optimization mechanisms for processing deviation data. In precision operation scenarios, subtle directional deviations cannot be accurately identified and corrected, further reducing control accuracy.
[0004] Therefore, existing technologies have the problems of being unable to dynamically adapt to user operation behavior and physiological state, and insufficient accuracy in deviation correction. Summary of the Invention
[0005] This invention provides a high-precision mouse direction control method and system to solve the problems of the inability to dynamically adapt to user operation behavior and physiological state, and insufficient deviation correction accuracy in existing technologies.
[0006] In a first aspect, to solve the above-mentioned technical problems, the present invention provides a high-precision mouse direction control method, comprising: Acquire mouse movement direction deviation data, user operation frequency data, and user status information; analyze the direction deviation data to mark high deviation areas and obtain a preliminary classified direction deviation dataset. The initial classification of the orientation deviation dataset is smoothed using the Kalman filtering algorithm, and the user state information and the user operation frequency data are fused together to optimize the data, resulting in an optimized deviation data sequence. Analyze the periodicity of the operation frequency in the optimized deviation data sequence, mark the abnormal intervals and adjust the deviation correction weights to obtain the frequency-adjusted deviation correction value; The deviation correction value after frequency adjustment is combined with the fatigue index in the user status information to obtain a real-time parameter optimization scheme by matching the operation rhythm. The directional gain setting of the mouse control system is updated according to the real-time parameter optimization scheme, and the mouse response under precision operation scenarios is calibrated to obtain the adjusted control parameter set. Based on the adjusted control parameter set, monitor the direction deviation data of subsequent mouse movements. If it does not match the real-time parameter optimization scheme, iteratively apply the Kalman filtering algorithm for optimization until it matches or reaches the maximum number of iterations, and then determine the final deviation correction model. Based on the final deviation correction model, the user operation frequency data and the user status information are fused together, and a personalized mouse control configuration file is generated through adaptive gain and applied in real time.
[0007] Secondly, the present invention provides a high-precision control system for mouse orientation, comprising: The data acquisition module acquires mouse movement direction deviation data, user operation frequency data, and user status information, analyzes the direction deviation data to mark high deviation areas, and obtains a preliminary classified direction deviation dataset. The filtering and optimization module uses the Kalman filtering algorithm to smooth the initially classified orientation deviation dataset, and integrates the user state information and the user operation frequency data to optimize the data, thereby obtaining the optimized deviation data sequence. The frequency correction module analyzes the periodic characteristics of the operating frequency in the optimized deviation data sequence, marks abnormal intervals, and adjusts the deviation correction weights to obtain the frequency-adjusted deviation correction value. The gain calculation module combines the frequency-adjusted deviation correction value with the fatigue index in the user status information to obtain a real-time parameter optimization scheme by matching the operation rhythm. The setting update module updates the directional gain setting of the mouse control system according to the real-time parameter optimization scheme, calibrates the mouse response in precision operation scenarios, and obtains the adjusted control parameter set. The iterative optimization module monitors the direction deviation data of subsequent mouse movements based on the adjusted control parameter set. If it does not match the real-time parameter optimization scheme, it iteratively applies the Kalman filtering algorithm for optimization until it matches or reaches the maximum number of iterations, and then determines the final deviation correction model. The configuration generation module integrates the user operation frequency data and the user status information according to the final deviation correction model, generates a personalized mouse control configuration file through adaptive gain, and applies it in real time.
[0008] Compared with the prior art, the present invention has the following beneficial effects: (1) By integrating mouse movement direction deviation data, user operation frequency data and user status information, this invention breaks through the limitation of traditional action perception relying on a single parameter, realizes deviation recognition driven by multi-dimensional data, and greatly improves the accuracy of high deviation area positioning.
[0009] (2) This invention uses the smoothing and iterative optimization mechanism of the Kalman filtering algorithm, combined with the dynamic adjustment of filtering parameters based on user status, to effectively suppress data noise interference. At the same time, it optimizes the correction weight by analyzing the periodic characteristics of user operation frequency, which significantly improves the stability and fineness of deviation correction.
[0010] (3) Based on the user fatigue index and other state information, the present invention dynamically adjusts the adaptive gain coefficient to generate a personalized control configuration file, realizes the real-time matching of control parameters with user operation rhythm and physiological state, and greatly improves control accuracy and user operation experience in precision operation scenarios. Attached Figure Description
[0011] Figure 1 This is a schematic flowchart of the high-precision mouse direction control method provided in the first embodiment of the present invention; Figure 2 This is a schematic diagram of the high-precision mouse direction control system provided in the second embodiment of the present invention. Detailed Implementation
[0012] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0013] Reference Figure 1 The first embodiment of the present invention provides a high-precision mouse direction control method, comprising the following steps: S11, acquire mouse movement direction deviation data, user operation frequency data and user status information, analyze the direction deviation data to mark high deviation areas, and obtain a preliminary classified direction deviation dataset; S12, the Kalman filtering algorithm is used to smooth the preliminary classified orientation deviation dataset, and the user state information and user operation frequency data are fused to optimize the data, resulting in an optimized deviation data sequence; S13, Analyze the periodic characteristics of the operation frequency in the optimized deviation data sequence, mark the abnormal intervals and adjust the deviation correction weight to obtain the frequency-adjusted deviation correction value. S14, combine the frequency-adjusted deviation correction value with the fatigue index in the user status information to perform operation rhythm matching and obtain a real-time parameter optimization scheme. S15, update the directional gain setting of the mouse control system according to the real-time parameter optimization scheme, calibrate the mouse response in precision operation scenarios, and obtain the adjusted control parameter set; S16. Monitor the direction deviation data of subsequent mouse movement according to the adjusted control parameter set. If it does not match the real-time parameter optimization scheme, iteratively apply the Kalman filtering algorithm for optimization until it matches or reaches the maximum number of iterations, and then determine the final deviation correction model. S17, Based on the final deviation correction model, the user operation frequency data and the user status information are fused together, and a personalized mouse control configuration file is generated through adaptive gain and applied in real time.
[0014] In step S11, it is necessary to acquire mouse movement direction deviation data, user operation frequency data, and user status information, analyze the direction deviation data to mark high deviation regions, and obtain a preliminary classified direction deviation dataset, including: The sensor collects real-time data on the direction deviation of mouse movement and the frequency of user operation, and obtains user physiological status data from the heart rate monitoring device as user status information. The directional deviation data is compared with a preset deviation threshold. If the deviation exceeds the threshold, it is marked as a high deviation region, thus obtaining a preliminary directional deviation dataset.
[0015] It's important to note that directional deviation data refers to the quantified offset between the actual mouse movement trajectory and the user's intended target trajectory. This data originates from the mouse's built-in optical or laser sensor, which captures the mouse's displacement coordinates on a two-dimensional plane (such as a desktop or mouse pad) in real time and calculates the difference between these coordinates and the target coordinates corresponding to the user's intended action. The purpose is to accurately quantify the degree of deviation in mouse control, providing a data foundation for subsequent calibration. User operation frequency data refers to the number of mouse operations or movement frequencies per unit of time, including click frequency (left and right clicks) and movement frequency (distance or number of mouse movements per unit of time). This data originates from the mouse's operation trigger module and reflects the density and rhythm of user operations. Physiological state data in the user status information primarily refers to heart rate data. Heart rate changes are strongly correlated with user fatigue, tension, and other states, providing support for interpreting the deviation data from the user's actual state and enabling control optimization that better reflects the user's actual condition.
[0016] The preset deviation threshold is a baseline value determined by stratified sampling and probability distribution percentile method based on all historical measured user sample data covering different operation scenarios (document editing, drawing, e-sports, etc.) and operation habits. First, deviation data under each scenario is collected in stratified manner. Then, a normal distribution is fitted to the full data and the 95th percentile is taken as the preset deviation threshold (such as ±8 pixels under 1920×1080 resolution) to distinguish between normal deviation and abnormal deviation. It can be adaptively adjusted according to different screen resolutions and operation scenarios.
[0017] In this step, the sensor's data acquisition function is first activated. The mouse's built-in optical sensor acquires movement trajectory data in real time at a resolution of 1200 DPI, recording the mouse's two-dimensional coordinates (x-axis and y-axis) every 10 milliseconds. Simultaneously, the mouse's microcontroller counts the number of clicks and movement distance per unit time, generating user operation frequency data. The acquisition frequency is set to 1 time / second to ensure real-time data accuracy. Next, real-time heart rate data of the user is acquired from a heart rate monitoring device via Bluetooth or USB communication protocol, with an acquisition frequency of 1 time / 5 seconds. This data is timestamped and synchronized with the mouse operation data using a unified system clock reference. Both the mouse sensor and the heart rate monitoring device read the computer / controller's UTC timestamp (accurate to the millisecond level), and this timestamp is synchronously appended when acquiring data. Subsequent timestamp alignment completes the integration of multi-source data, ensuring the time consistency of directional deviation data, user operation frequency data, and user status information. This integration forms an initial dataset with the data format (timestamp, x-axis coordinate, y-axis coordinate, number of clicks, movement distance, heart rate value). The orientation deviation data is analyzed by calculating the Euclidean distance between the current mouse coordinates and the target coordinates (identified based on the user's operation context, such as the location of the clicked icon or the location of the dragged target) to obtain the orientation deviation data at each time point. The calculated orientation deviation data is compared with a preset deviation threshold (±8 pixels). If the absolute value of the deviation is greater than 8 pixels, the deviation data corresponding to that time point is marked as high deviation area data. All marked datasets are integrated to form a preliminary classified orientation deviation dataset.
[0018] It's worth noting that the preset deviation threshold is not a fixed value. It can be customized by the user or adaptively adjusted by the system according to the operating scenario. For example, it can be adjusted to ±5 pixels in a precision drawing scenario and ±10 pixels in a normal browsing scenario, improving the applicability of the method. In another possible implementation, in addition to the heart rate device, fatigue can be estimated based on user operation frequency data. A 10-minute sliding window is used (theoretically, 10 minutes of data can ensure the authenticity of the results, smooth out accidental fluctuations, and will not consume too many computing resources; in practical applications, it can be dynamically fine-tuned according to different scenarios and devices). The average user operation frequency data within the window is statistically analyzed, and the average user operation frequency of the first 30% of the time is taken as the baseline value. The frequency decay rate is obtained by calculating the ratio of the difference between the two to the baseline value, which is used to replace the heart rate data for parameter adjustment.
[0019] For example, suppose a user is editing a document on a computer with a screen resolution of 1920×1080, and their goal is to click on a specific paragraph of text (target coordinates x=800 pixels, y=500 pixels). The mouse's optical sensor captures the movement trajectory in real time. At a certain timestamp (1680000000000 milliseconds), the actual coordinates recorded are x=810 pixels, y=505 pixels. The deviation value is calculated using the Euclidean distance formula. The value is greater than the preset deviation threshold of 8 pixels, therefore the deviation data at this time point is marked as high deviation area data. At the same time, the operation frequency data corresponding to this timestamp is 2 clicks per second, 30 pixels of movement distance, and 85 beats / minute heart rate data. These data together constitute a record in the initial dataset and are included in the preliminary classification of the orientation deviation dataset.
[0020] In another example, when a user browses a webpage, the actual coordinates of the mouse movement are x=1200 pixels, y=300 pixels, and the target coordinates are x=1205 pixels, y=303 pixels. The deviation value is calculated using the formula. Pixels smaller than the preset deviation threshold are not marked as high deviation area data.
[0021] In step S12, the initial classified orientation deviation dataset needs to be smoothed using the Kalman filter algorithm, and the user state information and user operation frequency data are fused for data optimization to obtain an optimized deviation data sequence, including: The Kalman filtering algorithm is used to correct the orientation deviation dataset of the initial classification point by point to obtain a smoothed deviation data set. For the smoothed deviation data set, the heart rate data in the user status information is compared with a preset heart rate threshold. If the threshold is exceeded, the deviation data for the corresponding time period is weighted to obtain the adjusted deviation data set. Based on the adjusted set of deviation data, the changing trend of the user operation frequency data is integrated, and the deviation data showing abnormal fluctuations is corrected a second time to obtain an optimized deviation data sequence.
[0022] It should be noted that the Kalman filtering algorithm is an algorithm based on the state equation of a linear system. It uses observed data from the system's input and output to optimally estimate the system state. Its core function is to suppress noise interference in the data (such as sensor acquisition errors and coordinate fluctuations caused by ambient light interference), making the deviation data sequence smoother and better reflecting the true deviation trend. The preset heart rate threshold is determined based on the resting heart rate range of adults (60-100 beats / minute) combined with the characteristics of the operating scenario. It is divided into a normal threshold range (60-85 beats / minute) and an abnormal threshold range (<60 beats / minute or >85 beats / minute). When the heart rate exceeds the normal range, the user may be in a state of fatigue, tension, or distraction. In this case, the weight of the deviation data needs to be adjusted to highlight the deviation characteristics under these conditions.
[0023] The trend of user operation frequency data refers to the increase or decrease of operation frequency data analyzed by the sliding window method, including a stable trend, an upward trend (high frequency mode), and a downward trend (low frequency mode). The fluctuation of operation frequency is related to the anomaly of deviation data. Users are more likely to experience operation fatigue during high-frequency operation, which leads to an increase in deviation. Its function is to help identify abnormal fluctuation characteristics in deviation data.
[0024] In this step, point-by-point correction of the Kalman filter is first performed to establish the linear state equation and observation equation for the orientation deviation data. The state equation is as follows: , Where X(k) is the state vector at time k (the true deviation value to be estimated), A is the state transition matrix (set to 1, because the time correlation of the deviation data is linear), B is the control matrix (set to 0, with no external control input), U(k) is the control vector (set to 0, with no external control input), and W(k) is the process noise (following a Gaussian distribution with a mean of 0 and a variance Q of 0.01).
[0025] The observation equation is, Where Z(k) is the observation value at time k (the raw deviation data collected by the sensor), H is the observation matrix (set to 1), and V(k) is the observation noise (following a Gaussian distribution with a mean of 0 and a variance R of 0.02).
[0026] The orientation bias data of the initial classification is processed point by point through the prediction and update steps of Kalman filtering. The prediction step calculates the prior state estimate. and prior error covariance The Kalman gain is calculated using the current observation value Z(k) collected by the sensor. / ( By fusing predictions and observations, the smoothed deviation value at time k is obtained. Simultaneously update the error covariance Repeat the above steps for all deviation data points to finally form a smoothed deviation data set.
[0027] Further, weighted processing of the heart rate data is performed. The heart rate data corresponding to each data point in the smoothed deviation data set is extracted and compared with a preset heart rate threshold (60-85 beats / minute). If the heart rate data is within the normal threshold range, the deviation data remains unchanged; if the heart rate is below 60 beats / minute or above 85 beats / minute, a corresponding time period is determined (the time interval corresponding to two data points before and after the data point). A weighting coefficient is calculated based on the degree of heart rate deviation: when the heart rate is 50-60 beats / minute or 85-95 beats / minute, the weighting coefficient is 1.2; when the heart rate is 40-50 beats / minute or 95-105 beats / minute, the weighting coefficient is 1.3; and when the heart rate is below 40 beats / minute or above 105 beats / minute, the weighting coefficient is 1.5. All deviation data within this time period are multiplied by the corresponding weighting coefficient to obtain the adjusted deviation data set. The weighting coefficients are determined based on the quantitative correlation statistics of the degree of heart rate deviation in all historical measured user sample data, and the average value of the coefficients is taken according to the ideal test effect under different coefficients for each heart rate.
[0028] Finally, a secondary correction is performed on the trend of operation frequency changes. A 5-second sliding window is used to segment the user operation frequency data, calculating the average operation frequency within each window. The trend is analyzed by comparing the differences between the averages of adjacent windows. If the absolute value of the difference between the average of the later window and the average of the earlier window is less than 5%, it indicates a stable trend; if the difference is greater than or equal to 5%, it indicates an upward trend; and if the difference is less than or equal to -5%, it indicates a downward trend. For the adjusted deviation data set, if the operation frequency in the corresponding time period shows an upward trend (high-frequency mode) and the fluctuation amplitude of the deviation data (the difference between the maximum and minimum values) is greater than 5 pixels, then a secondary correction is performed on the deviation data in that time period. In one implementation, the correction formula can be designed by someone skilled in the art. If the operation frequency shows a downward trend and the deviation data fluctuation range is greater than 5 pixels, then the correction formula is as follows: Where b is the corrected deviation value, a is the original deviation value, and c is the fluctuation range. If it is a stable trend, the deviation data remains unchanged, and finally an optimized deviation data sequence is formed.
[0029] It is worth noting that the process noise and observation noise variance of the Kalman filter in this step are statistically determined based on the noise characteristics of a large amount of sensor-collected data. This effectively suppresses the interference of random noise on the deviation data while preserving the true deviation trend. The weighted processing time period is set to two data points before and after because the changes in user state have a certain temporal continuity. This setting can cover the deviation data interval corresponding to the state change, improving the rationality of the weighted processing. The analysis of the operation frequency change trend adopts the sliding window method. The sliding window size of 5 seconds and the difference for judging anomalies of 5% can be dynamically adjusted according to the actual application scenario. The values here are only examples. The design of the secondary correction formula fully considers the different effects of high-frequency and low-frequency operations on the deviation data. During high-frequency operations, the user's operation intensity is high, and the deviation fluctuation may be more significant. Therefore, a positive correction coefficient is used. During low-frequency operations, the user's operation intensity is low, and the deviation fluctuation is relatively gentle. A mild reverse correction coefficient is used. Other correction formulas that can achieve the same effect are also within the scope of protection of the embodiments in this specification.
[0030] For example, suppose the deviation data in the smoothed deviation data set for a certain time period (timestamp 1680000000000 milliseconds to 1680000000040 milliseconds) are 10 pixels, 11 pixels, 12 pixels, 11 pixels, and 10 pixels respectively. The corresponding heart rate data are 92 beats / minute, 93 beats / minute, 95 beats / minute, 94 beats / minute, and 93 beats / minute respectively, all exceeding the preset heart rate threshold (85 beats / minute), and the heart rate is in the range of 85-95 beats / minute. Therefore, the weighting coefficient is set to 1.2. Multiplying the deviation data within this time period by 1.2 respectively yields the adjusted deviation data: 12 pixels, 13.2 pixels, 14.4 pixels, 13.2 pixels, and 12 pixels, forming the adjusted deviation data set. Further analysis of the operation frequency data corresponding to this time period revealed that the average operation frequency within the sliding window (5 seconds) was 18 times / minute, 20 times / minute, 23 times / minute, 22 times / minute, and 21 times / minute, respectively. The differences in the average values between adjacent windows were 11.1%, 15%, -4.3%, and -4.5%, respectively. The first three windows showed a clear upward trend (high-frequency mode), and the fluctuation range of the adjusted deviation data within this time period was 14.4-12=2.4 pixels, which is less than 5 pixels. Therefore, no secondary correction was required, and the deviation data for this time period was directly included in the optimized deviation data sequence.
[0031] In another example, the adjusted deviation data set for a certain period contains deviation data of 15 pixels, 18 pixels, 16 pixels, 20 pixels, and 17 pixels, with a fluctuation range of 5 pixels. The corresponding operation frequency trend is upward (the average value increases from 25 times / minute to 35 times / minute). Then, according to the correction formula, the corrected deviation values are calculated to be 15×(1+5 / 100)=15.75 pixels, 18×1.05=18.9 pixels, 16×1.05=16.8 pixels, 20×1.05=21 pixels, and 17×1.05=17.85 pixels, forming the deviation data after secondary correction, which is then incorporated into the optimized deviation data sequence.
[0032] In step S13, it is necessary to analyze the periodicity characteristics of the operation frequency in the optimized deviation data sequence, mark abnormal intervals, and adjust the deviation correction weights to obtain the frequency-adjusted deviation correction value, including: Based on the optimized deviation data sequence, the time interval is segmented to obtain the distribution characteristics of the operation frequency and determine the fluctuation range of the periodic characteristics. If the fluctuation range of the periodic feature exceeds the preset fluctuation threshold, the corresponding time interval is marked to obtain an abnormal interval identifier. For the deviation data corresponding to the abnormal interval identifier, and in conjunction with the user status information, the correction weight is adjusted, and the adjusted weight configuration is determined. Based on the adjusted weight configuration and the preset frequency adjustment logic rules, the deviation data corresponding to the abnormal interval identifier is subjected to secondary correction processing to obtain the frequency-adjusted deviation correction value.
[0033] It should be noted that time interval segmentation refers to dividing the optimized deviation data sequence into continuous intervals of fixed length, with the time length set to 1 minute (the periodic characteristics of user operation frequency are usually presented on a minute-level time scale). This segmentation method can capture periodic characteristics while avoiding feature fragmentation caused by excessively short intervals or feature ambiguity caused by excessively long intervals. The distribution characteristics of operation frequency include parameters such as frequency mean, peak value, trough value, and cycle length. By statistically analyzing the above parameters of the operation frequency data within each time interval, the periodic pattern of frequency changes can be identified. For example, in drawing operations, users may exhibit a periodic pattern of "high-frequency operation for 30 seconds - low-frequency operation for 30 seconds". The fluctuation range of the periodic characteristic refers to the difference between the maximum and minimum values of the operation frequency within each cycle, used to quantify the severity of periodic changes.
[0034] The preset fluctuation threshold is determined based on frequency fluctuation statistics under normal operating scenarios, and is generally set to 30% of the normal frequency average (it can be dynamically adjusted according to the actual application scenario). It is used to distinguish between normal periodic fluctuations and abnormal periodic fluctuations. The abnormal interval identifier refers to the marking of the time interval where the fluctuation range exceeds the preset threshold, including the start time of the interval, the end time, and the fluctuation degree level (mild abnormality: exceeding the threshold by less than 10%; moderate abnormality: exceeding the threshold by 10%-30%; severe abnormality: exceeding the threshold by more than 30%). Its purpose is to clarify the key interval for subsequent correction.
[0035] Deviation correction weight refers to the coefficient used to adjust the degree of deviation correction. The base weight is set to 1.0, and it is dynamically adjusted based on user status information and frequency characteristics. The adjustment range is 0.8-1.5 (the maximum reasonable range determined by relevant technical personnel based on a large number of experimental sample data, which can be dynamically adjusted according to actual application scenarios). Its core function is to apply different degrees of correction to deviation data under different scenarios, thereby improving the targeting of the correction. The preset frequency adjustment logic rules are a set of rules formulated based on the correlation between operation frequency and deviation correction. For example, "increase the weight in high-frequency mode and decrease the weight in low-frequency mode" and "increase the weight of high-frequency operations when the user is fatigued." Its function is to standardize the execution logic of weight adjustment and secondary correction, ensuring the consistency and rationality of the correction process.
[0036] In this step, the time interval is segmented and periodic characteristics are analyzed first. The optimized deviation data sequence is divided into several intervals with a time length of 1 minute. For example, starting from the timestamp 1680000000000 milliseconds, each interval contains all deviation data and corresponding operation frequency data within 60 seconds. The operation frequency data within each interval is statistically analyzed to calculate the frequency mean (μ), peak value (Fmax), and trough value (Fmin). The period length is determined (calculated by the time interval between adjacent peaks or troughs), and the fluctuation range of the periodic characteristics ΔF = Fmax - Fmin is calculated.
[0037] Furthermore, abnormal intervals are labeled by comparing the fluctuation range ΔF of each interval with a preset fluctuation threshold (30% of the normal frequency average). Assuming the normal frequency average of this interval is μ = 20 times / minute and the preset fluctuation threshold is 6 times / minute, if ΔF = 15 times / minute > 6 times / minute, then this interval is determined to be an abnormal interval. The abnormality level is calculated based on the excess amplitude: (15-6) / 6 = 150% > 30%, therefore it is marked as a severely abnormal interval. An example of an abnormal interval label could include (start time: 1680000000000 milliseconds, end time: 1680000000600 milliseconds, abnormality level: severe).
[0038] In one implementation, a correction weight adjustment is also required. This involves extracting user status information (heart rate data) corresponding to the abnormal interval identifier and dynamically adjusting the correction weight based on the abnormality level. The base weight is 1.0. Exemplary adjustment rules are as follows: For mildly abnormal intervals, if the heart rate is within the normal range (60-85 beats / minute), the weight is increased to 1.1; if the heart rate exceeds the normal range, the weight is increased to 1.2. For moderately abnormal intervals, if the heart rate is normal, the weight is increased to 1.2; if the heart rate is abnormal, the weight is increased to 1.3. For severely abnormal intervals, if the heart rate is normal, the weight is increased to 1.3; if the heart rate is abnormal, the weight is increased to 1.5. For example, if the heart rate data corresponding to the severely abnormal interval is 90 beats / minute (exceeding the normal range), the adjusted weight configuration is 1.5. The weight value is determined based on the quantitative correlation statistics of the heart rate deviation degree of all historical measured user sample data, and the average of the coefficients is taken based on the ideal test effect for each heart rate corresponding to different coefficients.
[0039] Finally, a secondary correction and deviation correction value calculation are performed, incorporating preset frequency adjustment logic rules to perform secondary correction on the deviation data within the abnormal range. The frequency adjustment logic rules include: when the operation frequency is in high-frequency mode (above 120% of the interval average), the correction value is the product of the original deviation value and the adjusted weight, multiplied by a proportional coefficient slightly greater than 1 (e.g., 1.05; this coefficient is used to increase the deviation value, and different coefficients can be selected according to different application scenarios, as long as the target effect is achieved within a reasonable range; this is just an example); when the operation frequency is in low-frequency mode (below 80% of the interval average), the correction value is the product of the original deviation value and the adjusted weight, multiplied by a proportional coefficient slightly less than 1 (e.g., 0.95; this coefficient is used to decrease the deviation value, and different coefficients can be selected according to different application scenarios, as long as the target effect is achieved within a reasonable range; this is just an example); when the operation frequency is in normal mode (between 80% and 120%), the correction value is the product of the original deviation value and the adjusted weight. All corrected data constitute the set of deviation correction values after frequency adjustment.
[0040] It's worth noting that the segment length of the time interval can be flexibly adjusted according to different operation scenarios. For example, in high-frequency operation scenarios such as e-sports, the segment length can be adjusted to 30 seconds to more accurately capture rapidly changing periodic features; in low-frequency operation scenarios such as document editing, it can be adjusted to 2 minutes to reduce the division of invalid intervals. The calculation base (normal frequency average) for the preset fluctuation threshold uses the frequency average of each interval itself, rather than a globally fixed value, which can adapt to the differences in operation frequency among different users and improve the accuracy of abnormal interval identification.
[0041] For example, suppose the operation frequency data for a certain 1-minute time interval (1680000000600 ms to 1680000001200 ms) in the optimized deviation data sequence are 10 times / minute, 35 times / minute, 12 times / minute, 33 times / minute, 15 times / minute, and 30 times / minute, with a mean frequency μ = 22.5 times / minute, a peak Fmax = 35 times / minute, a trough Fmin = 10 times / minute, and a fluctuation range ΔF = 25 times / minute. The preset fluctuation threshold is 22.5 × 30% = 6.75 times / minute. ΔF far exceeds the threshold, and the calculated excess is (25 - 6.75) / 6.75 ≈ 270%, which is marked as a severely abnormal interval. The corresponding heart rate data for this interval is 88 beats / minute (outside the normal range), and the adjusted weight configuration is 1.5. If a deviation data point within the interval is 16 pixels, the corresponding operation frequency is 35 times / minute (120% higher than the average, high-frequency mode), then the correction value is 16 × 1.5 × 1.05 = 25.2 pixels; if a deviation data point is 8 pixels, the corresponding operation frequency is 10 times / minute (80% lower than the average, low-frequency mode), then the correction value is 8 × 1.5 × 0.95 = 11.4 pixels, forming the deviation correction value after frequency adjustment.
[0042] In step S14, the frequency-adjusted deviation correction value needs to be combined with the fatigue index in the user status information to perform operation rhythm matching and obtain a real-time parameter optimization scheme, including: Based on the user status information, the corresponding fatigue index is obtained from the pre-established monitoring database, and the fatigue index is processed by time segmentation to obtain the assessment result of the user's current energy level. If the energy level assessment result is lower than the preset energy level threshold, then the operation rhythm range that adapts to the current state is matched through logical judgment; By combining the operation rhythm range with the deviation correction value after frequency adjustment, the control parameter configuration is dynamically calculated, and the user operation frequency data is integrated to generate a real-time parameter optimization scheme.
[0043] It should be noted that the fatigue index refers to comprehensive data used to quantify the user's fatigue level. It is calculated using a preset fatigue assessment formula (such as a weighted scoring formula based on heart rate variability and operation time), with a value range of 0-100. The higher the score, the lower the fatigue level and the higher the energy level. The data comes from a pre-established monitoring database. The pre-established monitoring database consists of the user's historical initial dataset, using lightweight structured storage (such as SQLite). It stores the user's historical operation scene identifier, timestamp, operation frequency, deviation value, physiological state (heart rate / fatigue index), operation area, preset thresholds (deviation / heart rate, etc.) by scene partition index. User data is added in real time, and the fatigue assessment model and threshold parameters are updated weekly using incremental training. The monitoring database is mainly used to store real-time and historical time-series data, including raw and processed directional deviation, user operation frequency, user physiological state (heart rate), and calculated fatigue index. The database supports real-time data writing and high-frequency querying for energy level assessment, behavioral habit analysis, etc.
[0044] The preset energy level threshold is a critical value determined based on research into the physiology of human fatigue. A reasonable and feasible setting is 70 points. A score below this indicates that the user is fatigued, while a score above or equal to this indicates that the user is in good condition. The operation rhythm range refers to the appropriate range of operation frequencies for the user at the current energy level. It is dynamically matched based on the energy level assessment results, providing a rhythm adaptation benchmark for the calculation of control parameters. The directional gain coefficient is a key parameter determining the mouse's response sensitivity, with a value ranging from 0.8 to 1.5. The higher the gain coefficient, the more sensitive the mouse response.
[0045] In this step, fatigue index data and energy level assessment are first performed. User status information (heart rate data, operation time data) from the 10 minutes prior to the current time point is extracted from a pre-established monitoring database. A fatigue index score is then calculated for each time point using a fatigue assessment formula. One possible implementation involves collecting physiological data (heart rate, reaction time, etc.) and subjective fatigue scores from a large population, and using regression analysis to fit the fatigue assessment formula. Where m represents heart rate, with a baseline value of 75 beats / minute, and n represents operation duration in minutes (the maximum calculation duration is 120 minutes; any duration exceeding this is calculated as 120 minutes). The fatigue index score within 10 minutes is divided into 5-minute segments, and the mean of each segment is calculated to obtain the energy level assessment result.
[0046] Next, the operation rhythm range is matched. The current energy level assessment result is compared with the preset energy level threshold (70 points). If it is less than the threshold, the user is determined to be in a low energy level state. The corresponding operation rhythm range is matched through logical judgment and combined with the deviation correction value after frequency adjustment to dynamically calculate the directional gain coefficient. In one possible implementation, the default base gain for all scenarios is 1.0. Logically, when the energy level is ≥90 points, the operation rhythm range is 15-25 times / minute, and the gain is increased by 0.3; when the energy level is 70 points ≤ energy level <90 points, the operation rhythm range is 12-20 times / minute, and the gain is increased by 0.2; when the energy level is 50 points ≤ energy level <70 points, the operation rhythm range is 10-15 times / minute, and the gain remains unchanged; when the energy level is <50 points, the operation rhythm range is 8-12 times / minute, and the gain is decreased by 0.2.
[0047] Furthermore, fine-tuning is performed based on the deviation correction value: for deviation correction values less than 5 pixels, the gain is increased by 0.1; for deviation correction values greater than 5 pixels but less than 15 pixels, the gain remains unchanged; and for deviation correction values greater than 15 pixels, the gain is decreased by 0.1. Specifically, the directional gain coefficient... The final calculation formula can be expressed as: ,in, Sets the scene's base gain (default is 1.0). This is a gain adjustment based on the energy level (e.g., +0.3 when the energy level is ≥90). This is the gain fine-tuning amount based on the deviation correction value (e.g., +0.1 when the deviation value is <5 pixels); each adjustment amount is independently determined based on historical data verification and is linearly superimposed in this formula to jointly form the final directional gain coefficient.
[0048] The adjustment values involved are determined by stratified sampling and control effect verification based on all historical measured user sample data. First, the data are collected in three dimensions: "scenario, user status, and deviation range". Then, the control logic is matched and verified for each layer of data to select the optimal fine-tuning range that "neither amplifies the deviation nor fails to ensure the feel of operation".
[0049] Finally, a real-time parameter optimization scheme is generated, which integrates the current user operation frequency data (assuming the current operation frequency is 10 times / minute, which is within the matching rhythm range). The calculated control parameter configuration is integrated into a real-time parameter optimization scheme, which also includes the parameter's effective time (immediate effect) and adjustment cycle (recalculated every 30 seconds).
[0050] For example, suppose a user's energy level assessment score is 75, the matching operation rhythm range is 12-20 times / minute, the average deviation correction value after current frequency adjustment is 8 pixels, and the current operation frequency is 13 times / minute. The directional gain coefficient is calculated as 1.0 + 0.2 = 1.2, forming the corresponding real-time parameter optimization scheme.
[0051] In step S15, the directional gain setting of the mouse control system needs to be updated according to the real-time parameter optimization scheme to calibrate the mouse response under precision operation scenarios, resulting in an adjusted set of control parameters, including: According to the real-time parameter optimization scheme, the directional gain data corresponding to the current precision operation scenario is obtained from the pre-established database and segmented according to the operation area to determine the appropriate gain range; Using the adapted gain range and combined with dynamically changing scene characteristics, the mouse response is initially calibrated to obtain a response baseline value; If the response benchmark value deviates from the preset benchmark threshold, the control parameters are fine-tuned through logical judgment, and the final gain configuration scheme is generated by data fusion to obtain the adjusted control parameter set.
[0052] It should be noted that the pre-established database is mainly used to store model parameters and configuration templates, including gain ranges segmented by operation area under different precision operation scenarios, scene feature weights, preset benchmark thresholds, and the initial values of optimal Kalman filter parameters and gain adjustments learned from historical data. These data are obtained through a large number of user operation tests and effect feedback statistics in different scenarios. User operation habits and accuracy requirements vary in different operation areas. The central area is usually the main operation area with higher accuracy requirements and relatively sensitive gain settings, while the edge and corner areas have lower operation frequency and relatively lower accuracy requirements, and relatively stable gain settings. The database collects gain adaptation data for new scenarios or new users in real time, optimizes the gain range and benchmark thresholds through incremental training every month, and updates synchronously. The database supports indexed queries by scenario and region, providing a benchmark for parameter optimization.
[0053] The adapted gain range refers to the range of gain coefficients extracted from the database for the current precision operation scenario and area. This range is used to constrain the adjustment range of the directional gain coefficient in the real-time parameter optimization scheme, preventing the gain coefficient from exceeding the reasonable range and causing operational loss of control. Dynamically changing scene characteristics include operation speed, operation trajectory complexity, and target accuracy requirements. These characteristics are obtained through mouse sensor data and the application's scene recognition module.
[0054] The response baseline value refers to a benchmark value obtained by quantifying the mouse response sensitivity based on the adapted gain range and scene characteristics. It reflects the proportional relationship between the mouse movement distance and the screen cursor movement distance. In one possible implementation, the response baseline value is the product of the directional gain coefficient and the scene feature weights (where the scene feature weights are the sum of the operation speed weight, trajectory complexity weight, and target precision weight). The preset baseline threshold is the optimal response baseline value for each precision operation scenario. It is determined through testing with a large number of experimental sample data and is used to judge whether the response baseline value after preliminary calibration meets the scenario requirements.
[0055] In this step, the adaptation gain range is first determined. Assuming the current precision operation scenario is a drawing and design scenario, the scenario type is determined by the scenario recognition module of the application (such as the scenario identifier sent by the drawing software to the system), and the directional gain data corresponding to the scenario is extracted from the pre-established database. For example, the gain data for a drawing and design scene is segmented by the operating area: the gain range for the center area of the screen (screen coordinates x∈[480,1440], y∈[270,810], for a 1920×1080 resolution) is 1.2-1.5; the gain range for the edge area of the screen (x∈[0,480)∪(1440,1920] or y∈[0,270)∪(810,1080]) is 0.8-1.0; and the gain range for the four corner areas of the screen (x∈[0,240]∪[1680,1920] and y∈[0,135]∪[945,1080]) is 0.6-0.8. The current operating area is determined to be the center area of the screen (x=900 pixels, y=500 pixels) based on mouse sensor data; therefore, the appropriate gain range is extracted to be 1.2-1.5.
[0056] Next, preliminary calibration of the response baseline is performed to obtain dynamically changing scene features. The operation speed, calculated from mouse sensor data, is 100 pixels / second, which is considered a medium-speed operation, corresponding to a weight of 0.3; the operation trajectory complexity is a curved trajectory, corresponding to a weight of 0.4; the target precision requirement is high precision (line thickness 1 pixel), corresponding to a weight of 0.3; the total weight of scene features is 1.0. The directional gain coefficient in the real-time parameter optimization scheme is 1.33, therefore the response baseline value is 1.33 × (0.3 + 0.4 + 0.3) = 1.33 × 1.0 = 1.33.
[0057] It is worth noting that the values of operation speed weight, trajectory complexity weight, and target accuracy weight here are determined based on the priority of accuracy requirements in precision operation scenarios and the correlation analysis of all historical measured user sample data. Higher priority values have higher weights and can be dynamically adjusted according to the requirements of different scenario types, as long as the sum of the three values is 1.0. The values here are just examples.
[0058] Furthermore, logical judgments are made to fine-tune the control parameters. The preset baseline threshold for the drawing and design scenario is 1.3. The deviation between the response baseline value and the preset baseline threshold is calculated: 1.33 - 1.3 = 0.03, and the deviation rate is 0.03 / 1.3 ≈ 2.3%. In one implementation, the logical judgment fine-tuning rules are as follows: if the absolute value of the deviation rate is <3%, the fine-tuning amplitude is the deviation value multiplied by 0.5; if 3% ≤ absolute value of the deviation rate is <5%, the fine-tuning amplitude is the deviation value multiplied by 0.8; if the absolute value of the deviation rate is ≥5%, the fine-tuning amplitude remains unchanged. The threshold and the proportional coefficient of the fine-tuning amplitude used for judgment in the logical judgment fine-tuning rules are verified by collecting adjustment effect data of different proportional coefficients based on all historical measured user sample data. The optimal fine-tuning amplitude is selected and can be adjusted according to the accuracy requirements in different scenarios. Other logical judgment fine-tuning rules that can achieve the same effect are all within the scope of protection of this specification.
[0059] Finally, the adjusted control parameter set is generated. Using data fusion technology, the fine-tuned directional gain coefficient of 1.32 is integrated with the gain range constraint segmented by the operation area (1.2-1.5 in the central area, and the current coefficient of 1.32 is within the range). At the same time, scene adaptation parameters (such as trajectory smoothing coefficient of 0.9 and target adsorption threshold of 3 pixels) are added to form the adjusted control parameter set. The parameter set may also include scene identifier (drawing design), operation area identifier (central area), and effective time (immediate effect).
[0060] For example, assuming the current precision operation scenario is an e-sports aiming scenario, the operation area is the center area of the screen (the core aiming area), and the adaptation gain range extracted from the database is 1.4-1.8. The directional gain coefficient in the real-time parameter optimization scheme is 1.6. The scenario characteristics are: operation speed 200 pixels / second (high-speed operation), weight 0.4; operation trajectory complexity is a straight line trajectory, weight 0.2; target accuracy requirement is ultra-high precision (aiming point diameter 1 pixel), weight 0.4; the response baseline value is 1.6×(0.4+0.2+0.4)=1.6. The preset baseline threshold for the e-sports aiming scenario is 1.55, the deviation is 1.6-1.55=0.05, the deviation rate is 3.23%, the fine-tuning amplitude is 0.05×0.8=0.04, and the fine-tuned directional gain coefficient is 1.6-0.04=1.56. The integrated and adjusted set of control parameters includes a directional gain coefficient of 1.56, a trajectory smoothing coefficient of 0.8, a target adsorption threshold of 1 pixel, a scene identifier of e-sports aiming, and an operation area identifier of the center area.
[0061] In step S16, it is necessary to monitor the direction deviation data of subsequent mouse movement according to the adjusted control parameter set. If it does not match the real-time parameter optimization scheme, the Kalman filtering algorithm is applied iteratively for optimization until it matches or reaches the maximum number of iterations, and then the final deviation correction model is determined.
[0062] It should be noted that the subsequent mouse movement direction deviation data refers to the new deviation data collected in real time by the mouse sensor after applying the adjusted control parameter set. The collection method and calculation method are consistent with the direction deviation data in step S11. The matching judgment between the deviation data and the optimization scheme involves comparing the newly collected deviation data with the preset deviation target range in the real-time parameter optimization scheme. The deviation target range is determined based on the adjusted control parameter set and scenario requirements. For example, the deviation target range for a drawing and design scenario is ±3 pixels, and for an e-sports aiming scenario it is ±1 pixel. The judgment criterion is that if the deviation values of 5 consecutive data points are all within the target range, it is considered a match; otherwise, it is considered a mismatch.
[0063] Iterative application of the Kalman filtering algorithm refers to re-executing the Kalman filtering process in step S12 on the mismatched bias data. During the iteration process, the process noise and observation noise variance of the filtering algorithm are dynamically adjusted. For example, after each iteration, the process noise variance Q decreases by 10%, and the observation noise variance R decreases by 5%, until the bias data matches or the number of iterations reaches the preset upper limit of 5. Based on all historical measured user sample data, 3-4 iterations can cover the matching requirements of 90% of the scenarios. 5 iterations can further improve the adaptation rate. Beyond that, the marginal benefit drops sharply, and it will lead to excessive system resource consumption. 5 iterations can ensure the optimization effect while controlling system overhead. If there is still no match after reaching the maximum number of iterations, a fallback strategy is executed, and a global adaptation parameter set is activated. This involves monitoring the average parameter value of all historical user operation data stored in the database to stabilize operations. At the same time, abnormal data is marked for subsequent model optimization. The next time the system is started or the operation scenario is switched, the abnormal data is automatically called for model retraining, and the initial values of the Kalman filtering parameters and the gain coefficient adaptation rules are fine-tuned to improve the subsequent matching success rate.
[0064] The final deviation correction model refers to a mathematical model that, after multiple iterations of optimization, can stably match the deviation data to the preset target range. The model input consists of Kalman filter parameters (original deviation data, process noise variance, observation noise variance), directional gain coefficient, and deviation correction weight. The output is the deviation correction value, which is equal to the Kalman filter (original deviation data, process noise variance, observation noise variance) multiplied by the directional gain coefficient and the deviation correction weight. This model provides a stable model foundation for the subsequent generation of personalized configuration files.
[0065] In this step, the deviation data monitoring function is first activated. Using the adjusted control parameter set, the mouse sensor continues to collect movement trajectory data at a resolution of 1200 DPI, calculating the direction deviation data every 10 milliseconds. Five data points are continuously collected for matching. If the deviation data is determined to be mismatched with the optimization scheme, an iterative optimization process needs to be initiated. The Kalman filtering algorithm is applied again to these five deviation data points to adjust the process noise variance and observation noise variance. The filtering process in step S12 is followed for point-by-point correction, resulting in filtered deviation data that is then matched again. To ensure stability, the next set of five data points is monitored. If the newly collected data is still determined to be mismatched, a second iteration is performed. The process noise variance and observation noise variance are further adjusted, and the new deviation data is filtered. If all data points are within the target range, and two consecutive sets of data (a total of 10 data points) match the target range, the deviation data is determined to match the iteratively generated real-time parameter optimization scheme.
[0066] Finally, the final deviation correction model was determined by integrating the parameters of the Kalman filter after two iterations (process noise variance 0.0081, observation noise variance 0.01805), the adjusted control parameters (directional gain coefficient 1.32), deviation correction weight (1.2), etc., to establish a mathematical model. The deviation correction value is equal to the Kalman filter (original deviation data, process noise variance, observation noise variance) multiplied by the directional gain coefficient and the deviation correction weight. This model is the final deviation correction model.
[0067] For example, suppose that in an e-sports aiming scenario, after the adjusted control parameter set is applied, the first five deviation data collected are 1.2 pixels, 1.5 pixels, 0.9 pixels, 1.6 pixels, and 1.1 pixels. The preset deviation target range is ±1.5 pixels. The 1.6 pixel is outside the range and is judged to be mismatched, and the first iteration is started. The process noise variance (original 0.008, adjusted to 0.0072) and observation noise variance (original 0.015, adjusted to 0.01425) were adjusted. The filtered deviation data were 1.2 pixels, 1.4 pixels, 0.9 pixels, 1.5 pixels, and 1.1 pixels, all within the target range. The next set of data was monitored: 1.3 pixels, 1.4 pixels, 1.2 pixels, 1.5 pixels, and 1.0 pixels. All of them matched, and the requirements were met. The final deviation correction model was determined. The deviation correction value was Kalman filter (original deviation data, 0.0072, 0.01425) × 1.56 × 1.3. This model can stably control the deviation data within ±1.5 pixels, meeting the high precision requirements of e-sports aiming.
[0068] In step S17, the user operation frequency data and the user state information need to be fused according to the final deviation correction model to generate a personalized mouse control configuration file through adaptive gain and apply it in real time, including: The user operation frequency data and the user status information are processed by regional segmentation, and the correlation comparison is performed through data integration to obtain the range of user behavior habits; If the user's behavioral habits exceed the preset behavior threshold, the adaptive parameters are adjusted through logical comparison to determine the appropriate personalized configuration scheme. The mouse trajectory data is collected in real time, and the personalized configuration scheme is dynamically corrected through a feedback mechanism. Combined with changes in user status information, adaptive gain optimization is used to generate a personalized mouse control configuration file and apply it to the mouse control system.
[0069] It should be noted that correlation comparison refers to analyzing the correlation between operation frequency data and user status information (heart rate, fatigue index) within the same region and time segment, and calculating the correlation coefficient (ranging from -1 to 1). A positive correlation coefficient indicates that the operation frequency increases with the increase of heart rate / fatigue index, while a negative correlation coefficient indicates that the operation frequency decreases with the increase of heart rate / fatigue index. This is used to identify the association pattern between user behavior and status, such as "in the central region - morning time period, fatigue index and operation frequency are strongly negatively correlated".
[0070] The preset behavior threshold is a baseline range determined by establishing a probabilistic model based on the maximum amount of behavioral habit data of ordinary users that can be collected in the corresponding scenario, covering 90% of the sample data. This range is used to determine whether the current user's behavior is within the normal range; for example, the operation frequency habit threshold is 8-25 times / minute, and the heart rate habit threshold is 60-90 beats / minute. Adaptive parameters refer to dynamic parameters used to adjust the personalized configuration scheme. The core parameter is the gain adaptive coefficient, which dynamically adjusts the configuration scheme parameters based on the degree of deviation between the user's behavior habits and the preset behavior threshold. Adaptive gain optimization refers to adjusting the gain coefficient in real time based on changes in user state information (such as increased heart rate and decreased fatigue indicators). In one possible embodiment, the optimized gain is obtained by calculating the sum of 1 and the proportion of heart rate change, the sum of 1 and the proportion of fatigue indicator change, and the current gain, and finally multiplying these three to ensure that the gain coefficient always matches the user's current state.
[0071] A personalized mouse control configuration file is a lightweight JSON file that contains personalized configuration schemes, adaptive parameters, feedback correction rules, etc., making it easy for the mouse control system to read and apply. The file also contains metadata such as user identifier, configuration generation time, and effective period. The configuration file is transmitted to the mouse's built-in microcontroller via USB or Bluetooth.
[0072] In this step, the data is first segmented and compared for correlation. User operation frequency data and user status information from the past hour are segmented into three time intervals (20 minutes each) and three operation areas (center, edge, and corners), resulting in nine area-time segments. Statistical analysis is performed on the data within each segment. For example, in the "Central Area - Interval 1 (0-20 minutes)," the average operation frequency is 20 times / minute, the average heart rate is 75 beats / minute, and the average fatigue index is 80 points. Correlation coefficients are calculated: the correlation coefficient between operation frequency and heart rate is -0.2 (weak negative correlation), and the correlation coefficient between operation frequency and fatigue index is 0.3 (weak positive correlation). In the "Edge Area - Interval 2 (20-40 minutes)," the average operation frequency is 12 times / minute, the average heart rate is 80 beats / minute, and the average fatigue index is 70 points, with correlation coefficients of -0.1 and 0.2, respectively. Through data integration, the range of user behavior habits was obtained: operation frequency in the central area was 18-22 times / minute, heart rate was 70-80 beats / minute, and fatigue index was 75-85 points; operation frequency in the peripheral area was 10-14 times / minute, heart rate was 75-85 beats / minute, and fatigue index was 65-75 points; operation frequency in the four corner areas was 8-10 times / minute, heart rate was 80-90 beats / minute, and fatigue index was 60-70 points.
[0073] Next, behavioral threshold comparison and adaptive parameter adjustment are performed. The user's behavioral habit range is compared with the preset behavioral thresholds. The preset behavioral thresholds are: central area operation frequency 15-25 times / minute, heart rate 60-90 beats / minute, fatigue index 60-90 points; edge area operation frequency 8-16 times / minute, heart rate 60-90 beats / minute, fatigue index 50-80 points; four corner areas operation frequency 6-12 times / minute, heart rate 60-90 beats / minute, fatigue index 50-80 points. The current user's behavioral habit range is within the preset thresholds, so there is no need to adjust the adaptive parameters. The adaptive parameters remain at their default values (adaptive gain coefficient 1.0). The personalized configuration scheme is determined as follows: central area directional gain coefficient 1.32, deviation correction weight 1.2; edge area directional gain coefficient 0.9, deviation correction weight 1.0; four corner areas directional gain coefficient 0.7, deviation correction weight 0.9.
[0074] Furthermore, the feedback mechanism is dynamically adjusted. The feedback mechanism is activated to collect mouse trajectory data in real time, calculating the deviation rate between the actual deviation data and the expected deviation target (center area ±3 pixels, edge area ±5 pixels, corner area ±7 pixels) every minute. For example, the center area deviation data collected in the first minute is 2.8 pixels, 3.0 pixels, 2.7 pixels, 3.2 pixels, and 2.9 pixels, with an expected target of ±3 pixels. The average deviation rate is (|2.8-0|+|3.0-0|+|2.7-0|+|3.2-0|+|2.9-0|) / 5 / 3×100%≈95.3%, which is within an acceptable range (deviation rate ≤110%), requiring no configuration adjustment. The edge area deviation data collected in the second minute is 5.2 pixels, 4.8 pixels, 5.5 pixels, 4.9 pixels, and 5.1 pixels. The expected target is ±5 pixels, and the average deviation rate is (5.2+4.8+5.5+4.9+5.1) / 5 / 5×100%=102.8%, which is slightly higher than the acceptable range. The deviation correction weight of the edge area is finely adjusted by logical judgment. The original weight is 1.0, and after fine adjustment it is 1.0×(1-(102.8%-100%))=0.972≈0.97. The corrected edge area deviation data is recalculated as 5.0 pixels, 4.7 pixels, 5.3 pixels, 4.8 pixels, and 5.0 pixels, with an average deviation rate of ≈99.3%, which meets the requirements.
[0075] Finally, adaptive gain optimization and configuration file generation are performed. Considering changes in user status information, assuming the user's heart rate increases from 80 beats / minute to 85 beats / minute in the third minute, and the fatigue index decreases from 70 to 65, adaptive gain optimization is performed on the edge regions. The optimized gain coefficient is 0.9 × (1 + (85-80) / 100) × (1 - (70-65) / 100) ≈ 0.90. The final parameters, adaptive parameters, and feedback correction rules for all regions are integrated to generate a personalized mouse control configuration file. The file contains metadata such as user identifier, configuration generation time, and effective period. The configuration file is sent to the mouse control system via USB or Bluetooth communication. The mouse control system parses the file and applies the parameters in real time, completing the deployment of personalized control.
[0076] It should be noted that the generated personalized mouse control configuration file (JSON format) is transmitted to the mouse's microcontroller unit (MCU) via the HID (Human Interface Device) API provided by the host operating system or the manufacturer's specific device communication protocol. The mouse MCU firmware integrates a parsing module that can read the parameters in the configuration file and dynamically adjust its internal sensor data processing algorithm and cursor movement mapping logic, thereby enabling the real-time application of control parameters.
[0077] It is worth noting that the coefficient values and expected deviation targets involved in the correlation comparison in this embodiment are examples used to explain the principle and can be dynamically adjusted according to different application scenarios. Other values and calculation methods that can achieve the same effect are all within the protection scope of this specification.
[0078] For example, suppose that in a user's personalized configuration, the initial parameter for the four corner regions is a directional gain coefficient of 0.7. The collected deviation data for the four corner regions are 7.2 pixels, 7.5 pixels, 6.8 pixels, 7.3 pixels, and 7.1 pixels, with an expected target of ±7 pixels. The average deviation rate is approximately 104.3%. After fine-tuning the deviation correction weight to 0.93, the optimized deviation data are 6.9 pixels, 7.2 pixels, 6.7 pixels, 7.0 pixels, and 6.9 pixels, with an average deviation rate of approximately 98.6%. Subsequently, the user's heart rate decreases from 85 beats / minute to 75 beats / minute, and the fatigue index increases from 65 points to 75 points. After adaptive gain optimization, the gain coefficient for the four corner regions is 0.7 × (1 + (75-85) / 100) × (1 + (75-65) / 100) = 0.7 × 0.9 × 1.1 = 0.693 ≈ 0.69. This is ultimately incorporated into the personalized configuration file and applied in real time, achieving accurate adaptation of the four corner region operations.
[0079] In summary, this invention discloses a high-precision mouse direction control method, which includes collecting mouse movement direction deviation, user operation frequency data, and user status information to initially classify high-deviation areas; employing a Kalman filtering algorithm to smooth the deviation data, adjusting filtering parameters, analyzing periodic features to increase deviation correction weights, and dynamically calculating the control parameter gain coefficient to match the operation rhythm through adaptive gain, determining a real-time parameter optimization scheme and updating the directional gain settings of this invention, calibrating the response under precise operation scenarios, and iterating twice to form a final deviation correction model if mismatched, and finally fusing the data to generate a personalized mouse control configuration file for real-time application. This invention overcomes the limitations of traditional motion perception relying on a single parameter by fusing mouse movement direction deviation data, user operation frequency data, and user status information, achieving multi-dimensional data-driven deviation identification and significantly improving the accuracy of high-deviation area positioning. Simultaneously, this invention effectively suppresses data noise interference through the smoothing and iterative optimization mechanism of the Kalman filtering algorithm, combined with dynamic adjustment of filtering parameters based on user status, and significantly improves the stability and finesse of deviation correction by optimizing correction weights through periodic feature analysis of operation frequency. Furthermore, this invention dynamically adjusts the adaptive gain coefficient based on user fatigue index and other state information to generate a personalized control configuration file, thereby achieving real-time matching of control parameters with user operation rhythm and physiological state, significantly improving control accuracy and user operation experience in precision operation scenarios.
[0080] Reference Figure 2The second embodiment of the present invention provides a high-precision control system for mouse direction, comprising: The data acquisition module acquires mouse movement direction deviation data, user operation frequency data, and user status information, analyzes the direction deviation data to mark high deviation areas, and obtains a preliminary classified direction deviation dataset. The filtering and optimization module uses the Kalman filtering algorithm to smooth the initially classified orientation deviation dataset, and integrates the user state information and the user operation frequency data to optimize the data, thereby obtaining the optimized deviation data sequence. The frequency correction module analyzes the periodic characteristics of the operating frequency in the optimized deviation data sequence, marks abnormal intervals, and adjusts the deviation correction weights to obtain the frequency-adjusted deviation correction value. The gain calculation module combines the frequency-adjusted deviation correction value with the fatigue index in the user status information to obtain a real-time parameter optimization scheme by matching the operation rhythm. The setting update module updates the directional gain setting of the mouse control system according to the real-time parameter optimization scheme, calibrates the mouse response in precision operation scenarios, and obtains the adjusted control parameter set. The iterative optimization module monitors the direction deviation data of subsequent mouse movements based on the adjusted control parameter set. If it does not match the real-time parameter optimization scheme, it iteratively applies the Kalman filtering algorithm for optimization until it matches or reaches the maximum number of iterations, and then determines the final deviation correction model. The configuration generation module integrates the user operation frequency data and the user status information according to the final deviation correction model, generates a personalized mouse control configuration file through adaptive gain, and applies it in real time.
[0081] It should be noted that the high-precision mouse direction control system provided in this embodiment of the invention is used to execute all the process steps of the high-precision mouse direction control method in the above embodiment. The working principle and beneficial effects of the two are one-to-one, so they will not be described again.
[0082] It should be noted that the system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the system embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.
[0083] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.
Claims
1. A high-precision mouse direction control method, characterized in that, include: Acquire mouse movement direction deviation data, user operation frequency data, and user status information; analyze the direction deviation data to mark high deviation areas and obtain a preliminary classified direction deviation dataset. The initial classification of the orientation deviation dataset is smoothed using the Kalman filtering algorithm, and the user state information and the user operation frequency data are fused together to optimize the data, resulting in an optimized deviation data sequence. Analyze the periodicity of the operation frequency in the optimized deviation data sequence, mark the abnormal intervals and adjust the deviation correction weights to obtain the frequency-adjusted deviation correction value; The deviation correction value after frequency adjustment is combined with the fatigue index in the user status information to obtain a real-time parameter optimization scheme by matching the operation rhythm. The directional gain setting of the mouse control system is updated according to the real-time parameter optimization scheme, and the mouse response under precision operation scenarios is calibrated to obtain the adjusted control parameter set. Based on the adjusted control parameter set, monitor the direction deviation data of subsequent mouse movements. If it does not match the real-time parameter optimization scheme, iteratively apply the Kalman filtering algorithm for optimization until it matches or reaches the maximum number of iterations, and then determine the final deviation correction model. Based on the final deviation correction model, the user operation frequency data and the user status information are fused together, and a personalized mouse control configuration file is generated through adaptive gain and applied in real time.
2. The high-precision mouse direction control method according to claim 1, characterized in that, The process involves acquiring mouse movement direction deviation data, user operation frequency data, and user status information; analyzing the direction deviation data to mark high-deviation regions; and obtaining a preliminary classified direction deviation dataset, including: The sensor collects real-time data on the direction deviation of mouse movement and the frequency of user operation, and obtains user physiological status data from the heart rate monitoring device as user status information. The directional deviation data is compared with a preset deviation threshold. If the deviation exceeds the threshold, it is marked as a high deviation region, thus obtaining a preliminary directional deviation dataset.
3. The high-precision mouse direction control method according to claim 1, characterized in that, The initial classification of the orientation deviation dataset is smoothed using the Kalman filter algorithm, and the user state information and user operation frequency data are fused for data optimization to obtain an optimized deviation data sequence, including: The Kalman filtering algorithm is used to correct the orientation deviation dataset of the initial classification point by point to obtain a smoothed deviation data set. For the smoothed deviation data set, the heart rate data in the user status information is compared with a preset heart rate threshold. If the threshold is exceeded, the deviation data for the corresponding time period is weighted to obtain the adjusted deviation data set. Based on the adjusted set of deviation data, the changing trend of the user operation frequency data is integrated, and the deviation data showing abnormal fluctuations is corrected a second time to obtain an optimized deviation data sequence.
4. The high-precision mouse direction control method according to claim 1, characterized in that, The process involves analyzing the periodicity of the operation frequency in the optimized deviation data sequence, marking abnormal intervals, adjusting the deviation correction weights, and obtaining the frequency-adjusted deviation correction value, including: Based on the optimized deviation data sequence, the time interval is segmented to obtain the distribution characteristics of the operation frequency and determine the fluctuation range of the periodic characteristics. If the fluctuation range of the periodic feature exceeds the preset fluctuation threshold, the corresponding time interval is marked to obtain an abnormal interval identifier. For the deviation data corresponding to the abnormal interval identifier, and in conjunction with the user status information, the correction weight is adjusted, and the adjusted weight configuration is determined. Based on the adjusted weight configuration, and by integrating the preset frequency adjustment logic rules, the deviation data corresponding to the abnormal interval identifier is subjected to secondary correction processing to obtain the frequency-adjusted deviation correction value.
5. The high-precision mouse direction control method according to claim 1, characterized in that, The step of combining the frequency-adjusted deviation correction value with the fatigue index in the user status information to obtain a real-time parameter optimization scheme by matching the operation rhythm includes: Based on the user status information, the corresponding fatigue index is obtained from the pre-established monitoring database, and the fatigue index is processed by time segmentation to obtain the assessment result of the user's current energy level. If the energy level assessment result is lower than the preset energy level threshold, then the operation rhythm range that adapts to the current state is matched through logical judgment. By combining the operation rhythm range with the deviation correction value after frequency adjustment, the control parameter configuration is dynamically calculated, and the user operation frequency data is integrated to generate a real-time parameter optimization scheme.
6. The high-precision mouse direction control method according to claim 1, characterized in that, The step of updating the directional gain setting of the mouse control system according to the real-time parameter optimization scheme, calibrating the mouse response under precision operation scenarios, and obtaining the adjusted control parameter set includes: According to the real-time parameter optimization scheme, the directional gain data corresponding to the current precision operation scenario is obtained from the pre-established database and segmented according to the operation area to determine the appropriate gain range; Using the adapted gain range and combined with dynamically changing scene characteristics, the mouse response is initially calibrated to obtain a response baseline value; If the response benchmark value deviates from the preset benchmark threshold, the control parameters are fine-tuned through logical judgment, and the final gain configuration scheme is generated by data fusion to obtain the adjusted control parameter set.
7. The high-precision mouse direction control method according to claim 1, characterized in that, The step of fusing the user operation frequency data and the user state information according to the final deviation correction model, generating a personalized mouse control configuration file through adaptive gain, and applying it in real time includes: The user operation frequency data and the user status information are processed by regional segmentation, and the correlation comparison is performed through data integration to obtain the range of user behavior habits; If the user's behavioral habits exceed the preset behavior threshold, the adaptive parameters are adjusted through logical comparison to determine the appropriate personalized configuration scheme. The mouse trajectory data is collected in real time, and the personalized configuration scheme is dynamically corrected through a feedback mechanism. Combined with changes in user status information, adaptive gain optimization is used to generate a personalized mouse control configuration file and apply it to the mouse control system.
8. A high-precision control system for mouse direction, characterized in that, include: The data acquisition module acquires mouse movement direction deviation data, user operation frequency data, and user status information, analyzes the direction deviation data to mark high deviation areas, and obtains a preliminary classified direction deviation dataset. The filtering and optimization module uses the Kalman filtering algorithm to smooth the initially classified orientation deviation dataset, and integrates the user state information and the user operation frequency data to optimize the data, thereby obtaining the optimized deviation data sequence. The frequency correction module analyzes the periodic characteristics of the operating frequency in the optimized deviation data sequence, marks abnormal intervals, and adjusts the deviation correction weights to obtain the frequency-adjusted deviation correction value. The gain calculation module combines the frequency-adjusted deviation correction value with the fatigue index in the user status information to obtain a real-time parameter optimization scheme by matching the operation rhythm. The setting update module updates the directional gain setting of the mouse control system according to the real-time parameter optimization scheme, calibrates the mouse response in precision operation scenarios, and obtains the adjusted control parameter set. The iterative optimization module monitors the direction deviation data of subsequent mouse movements based on the adjusted control parameter set. If it does not match the real-time parameter optimization scheme, it iteratively applies the Kalman filtering algorithm for optimization until it matches or reaches the maximum number of iterations, and then determines the final deviation correction model. The configuration generation module integrates the user operation frequency data and the user status information according to the final deviation correction model, generates a personalized mouse control configuration file through adaptive gain, and applies it in real time.
