Digital photo frame angle adjusting method and system based on human body movement tracking

Abstract

The invention discloses a digital photo frame angle adjusting method and system based on human body movement tracking, and the method comprises the steps: obtaining gravity center distribution data in the movement process of a user, processing the gravity center distribution data, and obtaining a gravity center distribution data set; generating an initial data set, and determining gravity center distribution characteristics; the intermittent static duration in the user moving process is analyzed, segmented labeling is carried out, and boundary point data are obtained; extracting moving intermittent features in the intermittent feature time sequence, and generating fluctuation mode distribution information; a segmented adjustment mechanism is adopted to generate an initial angle adjustment scheme; tracking the dynamic posture of the user in real time to obtain a dynamic posture estimation result; determining a final angle adjusting path; determining an angle adjustment opportunity based on the final angle adjustment path; and adjusting the pitch angle of the digital photo frame. According to the method, the pitching angle of the photo frame can be accurately adjusted in time at a proper adjusting time.

Description

Technical Field

[0001] This invention relates to the field of electronic photo frame technology, and in particular to an electronic photo frame angle adjustment method and system based on human body movement tracking. Background Technology

[0002] In the field of modern smart devices, the application of electronic photo frame angle adjustment technology based on human motion tracking is becoming increasingly widespread. This technology dynamically adjusts the display angle of the photo frame by sensing the user's position and posture to ensure the best visual effect, and its application potential in smart homes and personalized interactions cannot be ignored. Researching this area can not only promote the improvement of device intelligence, but also provide users with a more natural and comfortable interaction method.

[0003] However, existing technologies for adjusting the angle of electronic photo frames based on human motion tracking often overlook the dynamic changes during user movement. Especially in complex scenarios, it is difficult to accurately capture subtle fluctuations in human posture, resulting in insufficient precision in angle adjustment. Furthermore, when dealing with scenarios where the user alternates between movement and stillness, existing technologies often suffer from inaccurate adjustment timing or slow response due to a lack of detailed analysis of the transition states. Summary of the Invention

[0004] This invention provides a method and system for adjusting the angle of an electronic photo frame based on human movement tracking, in order to solve the technical problems of insufficient accuracy in angle adjustment, inaccurate adjustment timing, or slow response in the prior art.

[0005] Firstly, to address the aforementioned technical problems, this invention provides a method for adjusting the angle of an electronic photo frame based on human movement tracking, comprising: S10, acquiring center-of-gravity distribution data during user movement, processing the center-of-gravity distribution data to obtain a center-of-gravity distribution dataset; S20, determining center-of-gravity distribution characteristics based on the center-of-gravity distribution dataset; S30, analyzing the intermittent stillness duration during user movement based on the center-of-gravity distribution characteristics, segmenting and labeling the segments to form an intermittent feature time series, determining the boundary points between stillness and movement, and acquiring boundary point data; S40, extracting movement data from the intermittent feature time series. Based on the dynamic intermittent characteristics and the boundary point data, the center of gravity fluctuation pattern is identified, and fluctuation pattern distribution information is generated; S50, based on the fluctuation pattern distribution information, a segmented adjustment mechanism is adopted to generate an initial angle adjustment scheme; S60, based on the initial angle adjustment scheme, the user's dynamic posture is tracked in real time to obtain dynamic posture estimation results; S70, based on the dynamic posture estimation results, the final angle adjustment path is determined; S80, based on the final angle adjustment path, the angle adjustment timing is determined; S90, based on the angle adjustment timing and the final angle adjustment path, the pitch angle of the electronic photo frame is adjusted.

[0006] Optionally, step S10 includes: S110, collecting center of gravity distribution data during user movement based on multi-sensor fusion technology to obtain raw center of gravity signal data; S120, classifying different phases based on the raw center of gravity signal data to obtain phased center of gravity signal datasets; S130, denoising the phased center of gravity signal datasets based on a preset filtering method to obtain denoised center of gravity signal datasets; S140, extracting features from the denoised center of gravity signal datasets to determine initial center of gravity distribution data; S150, if the initial center of gravity distribution data does not match a preset threshold range, correcting the initial center of gravity distribution data based on a support vector machine algorithm to obtain the final center of gravity distribution dataset.

[0007] Optionally, step S30 includes: S310, obtaining relevant records of user movement based on the center of gravity signal dataset, wherein the relevant records include multiple location information and multiple timestamps during the user's movement; S320: Perform preliminary cleaning on multiple location information and multiple timestamps to remove outliers and obtain a cleaned motion trajectory dataset; S330: Based on the cleaned motion trajectory dataset, analyze the distribution of user movement and intermittent stillness. If the speed is lower than a preset threshold, it is determined to be a stillness phase, and an initial stillness phase division result is obtained; S340: Based on the initial stillness phase division result, obtain the duration of each stillness phase, perform duration analysis in combination with multiple timestamps corresponding to the duration, determine the start and end points of each stillness phase, and form a stillness duration sequence; S350: Based on the stillness duration sequence, segment and label the duration of each stillness phase to obtain duration distribution features, and use a clustering algorithm to group the duration distribution features to obtain an intermittent feature time series; S360: Based on the intermittent feature time series, analyze the switching points between the user's stillness phase and movement phase to determine the boundary points between stillness and movement, and obtain boundary point data.

[0008] Optionally, step S40 includes: S410, extracting moving intermittent features from the intermittent feature time series and preprocessing the moving intermittent features; S420, dividing the preprocessed moving intermittent features into multiple time segments based on a time window segmentation method to obtain an initial moving intermittent feature set; S430, based on the initial intermittent feature set, dynamically analyzing each time segment of the moving intermittent using a sliding window technique to determine the feature change trend of the moving intermittent; S440, based on the boundary point data, if the boundary point data and the change trend of the intermittent features overlap at a time point, recording the fluctuation intensity of the overlapping area to obtain preliminary distribution information of the center of gravity fluctuation; S450, based on the preliminary distribution information of the center of gravity fluctuation, analyzing the specific form of the fluctuation pattern, and using the K-means clustering algorithm to classify the fluctuation intensity to determine the fluctuation pattern category in different time segments, i.e., identifying the center of gravity fluctuation pattern; S460, generating fluctuation pattern distribution information based on the center of gravity fluctuation pattern.

[0009] Optionally, step S50 includes: S510, based on the fluctuation pattern distribution information, obtaining the characteristic distribution of fluctuation data, marking data with strong fluctuations, and recording the position interval containing the fluctuation mark; S520, based on the position interval containing the fluctuation mark, using an interval division tool to generate an adjustment window range, and obtaining the boundary data of the adjustment window; S530, based on the boundary data of the adjustment window, if the position of the fluctuation mark exceeds a preset threshold range, using a segmented adjustment mechanism to segment the pitch angle of the electronic photo frame, and determining the pitch angle interval that needs to be adjusted; S540, based on the pitch angle interval that needs to be adjusted, using a segmented adjustment mechanism to calculate each pitch angle interval data, and determining the initial pitch angle value; S550, based on the initial pitch angle value, obtaining adjustment parameters, if the deviation between the adjustment parameters and the initial pitch angle value exceeds a preset threshold, using a linear regression model to optimize the initial pitch angle value, and obtaining the optimized pitch angle value; S560, based on the optimized pitch angle value, generating an initial angle adjustment scheme.

[0010] Optionally, step S60 includes: S610, based on the initial angle adjustment scheme, real-time tracking of the user's dynamic posture using a sensor device, collecting initial angle data of the user's posture, and recording relevant information in real-time based on a preset acquisition frequency to obtain posture estimation data; S620, based on the posture estimation data, obtaining the deviation between the initial angle and the standard posture, and determining the direction and magnitude of the angle adjustment; S630, if the deviation between the initial angle and the standard posture exceeds a preset threshold, real-time acquisition of changes in the user's posture using the sensor device to determine whether it conforms to the expected trajectory of the adjustment scheme; S640, if the changes in the user's posture do not conform to the expected trajectory of the adjustment scheme, adjusting the acquisition parameters of the real-time tracking module based on a feedback mechanism; S650, based on the adjusted acquisition parameters, obtaining the dynamic posture estimation result.

[0011] Optionally, step S70 includes: S710, obtaining initial data based on the dynamic attitude estimation result, the initial data including angle change information; S720, extracting the angle change information to obtain an initial angle change sequence; S730, generating a preliminary angle change trend map based on the initial angle change sequence to determine the fluctuation nodes; S740, performing data analysis within the adjustment window based on the fluctuation nodes, and if the detected angle fluctuation exceeds a preset threshold, locally adjusting the data within the adjustment window to obtain an adjusted fluctuation sequence; S750, smoothing the adjusted fluctuation sequence by using a moving average method to generate a smoothed angle adjustment curve; S760, performing secondary smoothing on the angle change of the curve corresponding to the adjustment window in the angle adjustment curve to determine the final angle adjustment path.

[0012] Optionally, step S90 includes: S910, predicting the amplitude and frequency of angle adjustment based on the angle adjustment timing and the final angle adjustment path to obtain a control parameter combination; S920, adjusting the pitch angle of the electronic photo frame based on the control parameter combination, while monitoring the angle change in real time during the adjustment process. If the trend of angle change does not reach the preset stability standard, the control parameters are redefined; S930, continuously iteratively adjusting the pitch angle of the electronic photo frame based on the redefined control parameters, collecting the adjusted pitch angle data, analyzing the pitch angle data, and determining whether the angle adjustment timing is appropriate. If not, adjusting again until the final angle adjustment timing is determined; S940, adjusting the pitch angle of the electronic photo frame according to the final angle adjustment path at the final angle adjustment timing.

[0013] Secondly, this invention provides an electronic photo frame angle adjustment system based on human motion tracking, comprising: a data acquisition module for acquiring center-of-gravity distribution data during user movement, processing the center-of-gravity distribution data to obtain a center-of-gravity distribution dataset; a feature determination module for determining center-of-gravity distribution features based on the center-of-gravity distribution dataset; a boundary point determination module for analyzing the intermittent stillness duration during user movement based on the center-of-gravity distribution features, segmenting and labeling the segments to form an intermittent feature time series to determine the boundary points between stillness and movement, and acquiring boundary point data; and a pattern recognition module for extracting movement intermittent features from the intermittent feature time series, combined with... The boundary point data is used to identify the center of gravity fluctuation pattern and generate fluctuation pattern distribution information. A scheme generation module generates an initial angle adjustment scheme based on the fluctuation pattern distribution information and using a segmented adjustment mechanism. An attitude acquisition module tracks the user's dynamic attitude in real time based on the initial angle adjustment scheme and obtains dynamic attitude estimation results. A path determination module determines the final angle adjustment path based on the dynamic attitude estimation results. A noise reduction module determines the angle adjustment timing based on the final angle adjustment path. An angle adjustment module adjusts the pitch angle of the electronic photo frame based on the angle adjustment timing and the final angle adjustment path.

[0014] One or more technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages: This application acquires center of gravity distribution data by collecting data on the user's center of gravity distribution during movement; it then extracts movement interval features from the intermittent feature time series to generate fluctuation pattern distribution information, thereby generating an initial angle adjustment scheme; based on the initial angle adjustment scheme, it tracks the user's dynamic posture in real time to obtain dynamic posture estimation results; using the dynamic posture estimation results, it determines the final angle adjustment path; and finally, it adjusts the pitch angle of the electronic photo frame according to the angle adjustment timing and the final angle adjustment path. This method can accurately adjust the pitch angle of the photo frame at the appropriate adjustment time. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the electronic photo frame angle adjustment method based on human body movement tracking provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of the electronic photo frame angle adjustment system based on human body movement tracking provided in an embodiment of the present invention. Detailed Implementation

[0016] 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.

[0017] In the relevant descriptions of this embodiment, the terms "including," "containing," and "possessing" are all open terms and are generally understood to include but not be limited to; the term "at least one" is generally understood to mean one or more, where "multiple" refers to two or more; the term "at least one of the following" or similar expressions refer to any combination of these items, including any combination of single or multiple items, for example, "at least one of a, b, or c", or "at least one of a, b, and c", which can all mean: a, b, c, ab (i.e., a and b), ac, bc, or abc, where a, b, and c can be single or multiple; the symbol "A / B" is used to describe the selection relationship of associated objects, generally indicating an "or" relationship.

[0018] In the following description of the embodiments, the terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms "a" and "the" as used in the embodiments of this application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise.

[0019] Those skilled in the art should understand that, in the following description of the embodiments of this application, the sequence of numbers does not imply the order of execution. Some or all steps may be executed in parallel or sequentially. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0020] Those skilled in the art will understand that the numerical ranges in the embodiments of this application should be understood to specifically disclose each intermediate value between the upper and lower limits of the range. Any stated value or intermediate value within a stated range, as well as any other stated value or each smaller range between intermediate values ​​within a range, are also included within this invention. The upper and lower limits of these smaller ranges may be independently included or excluded from the range.

[0021] Unless otherwise stated, the technical / scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. While this application describes only preferred methods and materials, any methods and materials similar or equivalent to those described herein may be used in the implementation or testing of this application. All references to this specification are incorporated by way of citation to disclose and describe the methods and / or materials associated with those references. In the event of any conflict with any incorporated reference, the content of this specification shall prevail.

[0022] This invention provides a method for adjusting the angle of an electronic photo frame based on human motion tracking. This method can be executed by a computer. (See reference...) Figure 1 The method may include the following steps: S10, acquire center of gravity distribution data during user movement, process the center of gravity distribution data, and acquire center of gravity distribution dataset.

[0023] In an exemplary embodiment, step S10 may include the following steps: S110 collects center of gravity distribution data during user movement based on multi-sensor fusion technology to obtain raw center of gravity signal data; S120, Based on the original centroid signal data, perform classification processing for different phases to obtain phased centroid signal datasets; S130, the centroid signal set of the stages is denoised based on a preset filtering method to obtain the denoised centroid signal dataset. S140, Perform feature extraction on the denoised centroid signal dataset to determine the initial centroid distribution data; S150, if the initial centroid distribution data does not match the preset threshold range, the initial centroid distribution data is corrected based on the support vector machine algorithm to obtain the final centroid distribution dataset.

[0024] Specifically, the following implementation methods can be used to collect center of gravity distribution data during user movement based on multi-sensor fusion technology: Pressure and acceleration data are collected using an array of pressure sensors (e.g., 16 pressure sensors embedded in each shoe sole, with a sampling frequency of 100Hz) and an accelerometer (also with a sampling frequency of 100Hz) installed in the sole of the user's shoe. The pressure and acceleration data are then processed by a Kalman filter algorithm to reduce noise and obtain a smooth center of gravity position signal. The gait cycle is divided into three stages: standing, stepping, and supporting. The center of gravity distribution data for each stage is recorded in real time by using time window analysis (with a window size of 0.5 seconds) combined with threshold judgment (a center of gravity offset greater than 0.1 meters is considered a stepping stage).

[0025] Furthermore, the preset filtering method is a low-pass filtering algorithm, using a Butterworth low-pass filter with a cutoff frequency of 2Hz to filter out jitter noise above this frequency. The filter order is set to 4th order to ensure that the signal is smooth without excessive distortion. The calculation process is implemented through discrete Fourier transform, which decomposes the signal into the frequency domain, attenuates the high-frequency components, and then inversely transforms it back to the time domain to obtain the smoothed signal.

[0026] Furthermore, the specific steps for correcting the initial centroid distribution data using the support vector machine algorithm are as follows: Choose a suitable SVM (Support Vector Machine) model; Choose an appropriate kernel function; Set hyperparameters, such as normalization parameters; Train the SVM model; The initial centroid distribution data are predicted using a trained SVM model. The initial centroid distribution data is corrected based on the prediction results.

[0027] S20, Based on the centroid distribution dataset, determine the centroid distribution characteristics.

[0028] The center of gravity distribution characteristics include the concentration trend and dispersion of the center of gravity during user movement and intermittent periods.

[0029] S30, based on the aforementioned center of gravity distribution characteristics, analyze the duration of intermittent stillness during the user's movement process, and perform segmentation and labeling to form an intermittent feature time series, so as to determine the boundary points between stillness and movement, and obtain boundary point data.

[0030] In an exemplary embodiment, step S30 may include the following steps: S310, Based on the center of gravity signal dataset, obtain relevant records of user movement, the relevant records including multiple location information and multiple timestamps during the user's movement; S320, perform preliminary cleaning on the multiple location information and multiple timestamps, remove outliers, and obtain the cleaned motion trajectory dataset; S330, based on the cleaned motion trajectory dataset, analyzes the distribution of user movement and intermittent stillness. If the speed is lower than the preset threshold, it is determined to be a still stage, and the initial still stage division results are obtained. S340, based on the initial static phase division results, obtain the duration of each static phase, combine the multiple timestamps corresponding to the duration for duration analysis, determine the start and end points of each static phase, and form a static duration sequence. S350, based on the static duration sequence, the duration of each static phase is segmented and labeled to obtain duration distribution features, and a clustering algorithm is used to group the duration distribution features to obtain an intermittent feature time series; S360, based on the intermittent characteristic time series, analyze the switching points between the user's stationary phase and the moving phase to determine the boundary points between stationary and moving phases, and obtain boundary point data.

[0031] Specifically, the process of determining the boundary points between stationary and moving states is as follows: First, multiple location information and timestamps during the user's movement are obtained through a center of gravity signal dataset. Then, a dataset is constructed using these location and timestamp data. Assuming a GPS device records the user's location once per second, the dataset includes timestamps and latitude / longitude information. For example, a user generates 600 data points within 10 minutes. Continuous time periods with latitude / longitude changes of less than 0.0001 degrees (approximately 10 meters) are initially defined as stationary phases. Next, the duration of each stationary phase is segmented and labeled using a time threshold method. Stationary periods less than 30 seconds are classified as brief stationary (e.g., waiting at a red light), 30 seconds to 5 minutes as moderate stationary (e.g., short rest), and longer than 5 minutes as long stationary (e.g., eating or working). For example, a user whose position remains unchanged for 60 seconds, from the 100th to the 160th second, is labeled as moderately stationary. Subsequently, an intermittent time series is generated. Based on the segmentation results described above, the user's daily activity data is organized chronologically into a sequence containing both stationary and moving states. For example, 8:00 to 8:10 is moving, 8:10 to 8:12 is moderately stationary, 8:12 to 8:20 is moving, and so on, generating a complete intermittent time series. Finally, the boundary points between stationary and moving states are determined using a velocity calculation method. Instantaneous velocity is calculated based on the positional changes of adjacent time points, with a velocity threshold set at 0.1 m / s. If the velocity is below this value and the duration exceeds 30 seconds, it is marked as the start point of stationary states; otherwise, it is marked as the start point of moving states. For example, if the velocity decreases from 0.5 m / s to 0.05 m / s at the 200th second and remains so for 40 seconds, then the 200th second is marked as the start boundary point of stationary states.

[0032] S40, extract the moving intermittent features from the intermittent feature time series, combine them with the boundary point data, identify the center of gravity fluctuation pattern, and generate fluctuation pattern distribution information.

[0033] In an exemplary embodiment, step S40 may include the following steps: S410, Extract moving intermittent features from the intermittent feature time series, and preprocess the moving intermittent features; S420, Based on the time window segmentation method, the preprocessed movement interval features are divided into multiple time segments to obtain an initial movement interval feature set; S430, Based on the initial set of intermittent features, dynamic analysis is performed on each time segment of the moving interval using the sliding window technique to determine the characteristic change trend of the moving interval; S440, based on the boundary point data, if the boundary point data and the change trend of the intermittent feature overlap at a time point, record the fluctuation intensity of the overlapping area to obtain preliminary distribution information of the center of gravity fluctuation; S450, based on the preliminary distribution information of the center of gravity fluctuation, analyze the specific form of the fluctuation pattern, and use the K-means clustering algorithm to classify the fluctuation intensity in order to determine the fluctuation pattern category in different time segments, that is, to identify the center of gravity fluctuation pattern. S460, Based on the center of gravity fluctuation pattern, generate fluctuation pattern distribution information.

[0034] In this embodiment, K-means clustering is a commonly used unsupervised learning algorithm used to divide a dataset into K distinct clusters. Data points within each cluster have high similarity, while data points between clusters have low similarity. Using K-means clustering to classify fluctuation intensity results in more accurate classification.

[0035] S50, based on the fluctuation pattern distribution information, adopts a segmented adjustment mechanism to generate an initial angle adjustment scheme.

[0036] In an exemplary embodiment, step S50 may include the following steps: S510, Based on the fluctuation pattern distribution information, obtain the characteristic distribution of fluctuation data, mark the data with strong fluctuations, and record the location interval containing the fluctuation mark; S520, Based on the position interval containing the fluctuation mark, an adjustment window range is generated using an interval division tool, and the boundary data of the adjustment window is obtained; S530, based on the boundary data of the adjustment window, if the position of the fluctuation mark exceeds the preset threshold range, the pitch angle of the electronic photo frame is segmented through the segmented adjustment mechanism to determine the pitch angle range that needs to be adjusted. S540, based on the pitch angle range that needs to be adjusted, a segmented adjustment mechanism is used to calculate the data for each pitch angle range to determine the initial pitch angle value. S550, based on the initial pitch angle value, an adjustment parameter is obtained. If the deviation between the adjustment parameter and the initial pitch angle value exceeds a preset threshold, the initial pitch angle value is optimized through a linear regression model to obtain an optimized pitch angle value. S560 generates an initial angle adjustment scheme based on the optimized pitch angle value.

[0037] Specifically, after obtaining the fluctuation pattern distribution information, it is necessary to obtain the characteristic distribution of the fluctuation data of the frame pitch angle and determine the position of the fluctuation marker. Before obtaining the characteristic distribution of the fluctuation data, the fluctuation data is processed first. The processing process will be described in detail below.

[0038] First, the frame's pitch angle fluctuation data is collected using a sensor (such as a MEMS tilt sensor). Assuming the collected fluctuation data is the pitch angle change value per minute, the range of the collected fluctuation data is between -5 degrees and 5 degrees, and the collection period is 10 minutes, a total of 10 data points are obtained as shown below: [-2.3, 1.5, -1.8, 3.2, -0.5, 2.1, -3.4, 0.8, 1.9, -1.1] degrees. Then, a segmented adjustment mechanism was used to analyze the data. The fluctuation data was divided into three segments in chronological order, with approximately three to four data points in each segment. The average fluctuation value of each segment was calculated. The average value of the first segment [-2.3, 1.5, -1.8] was -0.87 degrees, the average value of the second segment [3.2, -0.5, 2.1] was 1.6 degrees, and the average value of the third segment [-3.4, 0.8, 1.9, -1.1] was -0.45 degrees. Based on this, the adjustment requirements of each segment were analyzed. If the absolute value of the average value was greater than 1 degree, an angle adjustment was required. The adjustment amount was the opposite of the average value multiplied by a coefficient of 0.8. For example, the adjustment amount for the second segment was -1.6 * 0.8 = -1.28 degrees. Next, an adjustment window is set based on the fluctuation marking interval. Points with an absolute fluctuation value greater than 2 degrees are marked as high fluctuation points, such as 3.2 degrees and -3.4 degrees in the data. A time window of 1 minute before and after these points is set around them. The local average value is calculated by combining the data points of adjacent data points. For example, the average value of the data [1.5, 3.2, -0.5] in the window near 3.2 degrees is 1.4. The target angle value in the adjustment window is -1.4 * 0.8 = -1.12. Finally, an initial angle adjustment scheme is generated. Combining the results of segmented adjustment and window adjustment, the window adjustment value is used first. If there is no window adjustment, the segmented adjustment value is used to generate a time series adjustment scheme. For example, at the high fluctuation point of 3.2 degrees in the second segment, the adjustment is -1.12 degrees, and the remaining points are adjusted by -1.28 degrees according to the segmented average. Other segments are processed similarly, thus outputting the complete adjustment sequence shown below [-0.7, -0.7, -0.7, -1.12, -1.28, -1.28, 0.36, 0.36, 0.36, 0.36] degrees.

[0039] S60, based on the initial angle adjustment scheme, the user's dynamic posture is tracked in real time to obtain the dynamic posture estimation result.

[0040] In an exemplary embodiment, step S60 may include the following steps: S610, based on the initial angle adjustment scheme, the user's dynamic posture is tracked in real time by the sensor device, the initial angle data of the user's posture is collected, and relevant information is recorded in real time based on the preset collection frequency to obtain posture estimation data. S620, based on attitude estimation data, obtains the deviation between the initial angle and the standard attitude, and determines the direction and magnitude of angle adjustment; S630, if the deviation between the initial angle and the standard posture exceeds a preset threshold, the sensor device acquires the user posture change information in real time to determine whether it conforms to the expected trajectory of the adjustment scheme. S640, if the user posture change information does not conform to the expected trajectory of the adjustment scheme, adjust the acquisition parameters of the real-time tracking module based on the feedback mechanism; S650, based on the adjusted acquisition parameters, obtains the dynamic attitude estimation result.

[0041] Specifically, this exemplary embodiment obtains posture estimation values ​​based on the initial angle adjustment scheme. The acquisition process is as follows: First, the user's initial posture values ​​are obtained through a posture estimation model. Assuming the OpenPose algorithm based on deep learning is used, the coordinates of human key points are extracted from the video frames captured by the camera. For example, the shoulder coordinates are (200, 300), and the hip coordinates are (210, 500). By calculating the angle of the line connecting the two points, the initial spinal tilt angle is obtained as 12.5 degrees. Next, the user maintains a standard posture, and the coordinates of human key points are extracted again from the video frames captured by the camera, i.e., the standard key point coordinates. The angle of the line connecting the two standard key point coordinates is calculated, i.e., the standard posture angle. By comparing the human key point coordinates and the standard posture angle (assuming it is 12 degrees), it can be found that the initial posture has a slight forward tilt. Subsequently, the system enters the real-time tracking phase, using a Kalman filter algorithm to dynamically update the user's posture. Assuming an update frequency of 30Hz per frame, the system tracks changes in the coordinates of key points on the shoulder and hip. For example, in frame 10, the shoulder coordinates become (205, 305), and the hip coordinates become (215, 505), calculating an angle change of 0.3 degrees / second. Linear regression analysis of 10 consecutive frames shows a gradually increasing angle change trend, predicting a potential increase to 14.0 degrees within the next 5 seconds, indicating a further risk of forward tilting. Finally, based on the dynamic posture estimation results, the system automatically adjusts the feedback mechanism, combining trend data and preset thresholds (e.g., 15.0 degrees as a warning line). If the predicted value approaches the threshold, the algorithm generates correction suggestions, such as adjusting the center of gravity distribution parameter in the virtual reality scene from the current value of 0.6 to 0.8, to guide the user back to a safe angle range (10.0-12.0 degrees). The system records the angle change data after each adjustment, such as when the angle falls back to 11.8 degrees after adjustment, verifying the effectiveness of the adjustment.

[0042] Among them, the Kalman filter algorithm is a linear minimum mean square error (LMSSE) estimation algorithm. This algorithm fuses the system model and sensor data through a recursive prediction-update process, dynamically adjusts the weights of the predicted and observed values, and finally obtains the optimal estimate of the state. Therefore, it can dynamically update the user's attitude.

[0043] S70, based on the dynamic attitude estimation results, determine the final angle adjustment path.

[0044] In an exemplary embodiment, step S70 may include the following steps: S710, Initial data is obtained based on the dynamic attitude estimation results, and the initial data includes angle change information; S720, Extract the angle change information to obtain an initial angle change sequence; S730 generates a preliminary angle change trend map based on the initial angle change sequence and determines the fluctuation nodes; S740, based on the fluctuation node, data analysis is performed within the adjustment window. If the detected angle fluctuation exceeds the preset threshold, the data within the adjustment window is locally adjusted to obtain the adjusted fluctuation sequence. S750, The adjusted fluctuation sequence is smoothed by using a moving average method to generate a smoothed angle adjustment curve. S760, perform secondary smoothing on the angle change of the curve within the corresponding adjustment window in the angle adjustment curve to determine the final angle adjustment path.

[0045] Specifically, after obtaining initial data based on dynamic attitude estimation results, the initial data is analyzed. Assuming that shoulder joint angle data is collected at one reading per second within a certain time period, the data for 10 consecutive seconds is [30, 32, 35, 38, 40, 42, 45, 48, 50, 52]. Using this data, an initial angle adjustment curve is generated, and the rate of angle change between adjacent points is calculated using a linear interpolation algorithm. For example, the rate of change from the 1st second to the 2nd second is (32-30) / 1 degree / second = 2 degrees / second, and this rate of change sequence is calculated sequentially.

[0046] Furthermore, in this exemplary embodiment, when smoothing the adjusted fluctuation sequence, the adjustment window is set to 3 seconds, and a moving average algorithm is used to smooth the data. For example, at the 2nd second, the average value (30+32+35) / 3 = 32.3 degrees is calculated from the data (30, 32, 35) from the 1st to the 3rd second as the smoothed angle value. This process is repeated for the entire sequence to obtain the smoothed angle sequence [30, 32.3, 35, 38.3, 40.3, 42.3, 45.3, 48, 50, 52]. By analyzing the data before and after smoothing, it is found that the angle change is more gradual, avoiding the risk of mechanical jitter caused by abrupt changes. Finally, the final angle adjustment path is determined. Based on the smoothed sequence and the response time of the equipment control system (assuming a response time of 0.5 seconds), the angle increment every 0.5 seconds is calculated. For example, from the 2nd second to the 2.5th second, the angle increases from 32.3 degrees to 33.65 degrees. The final angle adjustment path is generated to ensure that the equipment is gradually adjusted to the target angle according to the final angle adjustment path (i.e., the smoothed path).

[0047] S80, based on the final angle adjustment path, determine the timing of angle adjustment.

[0048] The analysis process is described in detail below. First, a filtering algorithm is used to process the center-of-gravity distribution data. Assuming the original data sampling frequency is 50Hz, the duration of intermittent stillness is calculated. Stillness is defined as a center-of-gravity displacement velocity less than 0.1 m / s and a duration exceeding 2 seconds. By traversing the data points, time periods meeting this condition are recorded. For example, in 1000 sampling points, three stillness durations were found: 3.2 seconds, 4.1 seconds, and 2.8 seconds, with an average stillness duration of 3.37 seconds. Simultaneously, the movement interval characteristics are analyzed, calculating the movement time and displacement distance between adjacent stillness segments. Assuming a movement speed threshold of 0.5 m / s, the average duration of the movement interval is found to be 1.5 seconds, and the average displacement distance is 0.8 meters. Subsequently, the jitter signal strength is detected, and the jitter threshold is set to an acceleration change of 0.05 m / s. Fourier transform is performed on the acceleration data to extract frequency components. If the dominant frequency component is below the threshold of 0.05, it is determined to be a low-jitter state, triggering an adjustment opportunity. For example, if an acceleration dominant frequency of 0.03 m / s is detected in a certain data segment, below the threshold, the system automatically marks this moment as the adjustment starting point. Finally, the optimal angle adjustment opportunity is determined. Based on a comprehensive evaluation of the deviation angle of the center of gravity trajectory and the duration of stillness, an angle deviation threshold of 5 degrees is set. Combining the weight of the stillness duration (weight set to 0.6) and the weight of the angle deviation (weight set to 0.4), a comprehensive score is calculated. Assuming a certain moment has a stillness duration of 3.5 seconds and an angle deviation of 6 degrees, the score is calculated as 3.5 × 0.6 + 6 × 0.4 = 4.5. If the score exceeds the preset value of 4.0, the system automatically determines this moment as the optimal angle adjustment opportunity.

[0049] S90, based on the angle adjustment timing and the final angle adjustment path, adjust the pitch angle of the electronic photo frame.

[0050] In an exemplary embodiment, step S90 may include the following steps: S910, based on the angle adjustment timing and the final angle adjustment path, predict the amplitude and frequency of the angle adjustment to obtain a combination of control parameters; S920, Based on the combination of control parameters, the tilt angle of the electronic photo frame is adjusted, and the angle change is monitored in real time during the adjustment process. If the trend of the angle change does not reach the preset stability standard, the control parameters are re-determined. S930, based on the redefined control parameters, continuously iteratively adjusts the pitch angle of the electronic photo frame, collects the adjusted pitch angle data, analyzes the pitch angle data, and determines whether the timing of the angle adjustment is appropriate. If it is not appropriate, it adjusts again until the final timing of the angle adjustment is determined. S940, at the final angle adjustment timing, the tilt angle of the electronic photo frame is adjusted according to the final angle adjustment path.

[0051] In the above embodiments of this application, center of gravity distribution data is obtained by collecting data on the user's center of gravity distribution during movement; movement interval features are extracted from the intermittent feature time series to generate fluctuation pattern distribution information, thereby generating an initial angle adjustment scheme; the user's dynamic posture is tracked in real time using the initial angle adjustment scheme to obtain dynamic posture estimation results; the final angle adjustment path is determined based on the dynamic posture estimation results; and the pitch angle of the electronic photo frame is adjusted based on the angle adjustment timing and the final angle adjustment path. This method can accurately adjust the pitch angle of the photo frame at the appropriate adjustment time.

[0052] Based on the above embodiments, this application also provides an electronic photo frame angle adjustment system based on human body movement tracking. Figure 2 This is a schematic diagram of the structure of an electronic photo frame angle adjustment system based on human body movement tracking, as described in an embodiment of the present invention. Figure 2 As shown, the electronic photo frame angle adjustment system 200 based on human motion tracking may include: Data acquisition module 210 acquires center of gravity distribution data during user movement, processes the center of gravity distribution data, and acquires center of gravity distribution dataset; Feature determination module 220 is used to determine the centroid distribution features based on the centroid distribution dataset; The boundary point determination module 230 is used to analyze the duration of intermittent stillness during the user's movement based on the center of gravity distribution characteristics, and to perform segmentation and labeling to form an intermittent feature time series, so as to determine the boundary points between stillness and movement, and to obtain boundary point data. The pattern recognition module 240 is used to extract the moving intermittent features in the intermittent feature time series, combine them with the boundary point data, identify the center of gravity fluctuation pattern, and generate fluctuation pattern distribution information. The scheme generation module 250 is used to generate an initial angle adjustment scheme based on the fluctuation pattern distribution information and using a segmented adjustment mechanism. The attitude acquisition module 260 is used to track the user's dynamic attitude in real time based on the initial angle adjustment scheme and obtain the dynamic attitude estimation result. The path determination module 270 is used to determine the final angle adjustment path based on the dynamic attitude estimation result; The noise reduction module 280 is used to determine the timing of angle adjustment based on the final angle adjustment path; Angle adjustment module 290 is used to adjust the pitch angle of the electronic photo frame based on the angle adjustment timing and the final angle adjustment path.

[0053] In the description of this application, it should be noted that the terms "first", "second", and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0054] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0055] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interface; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0056] The units described as separate components may or may not be physically separate. 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 units can be selected to achieve the purpose of this embodiment according to actual needs.

[0057] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0058] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0059] In the description of this application, it should be noted that the terms "first", "second", and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0060] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0061] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interface; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0062] The units described as separate components may or may not be physically separate. 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 units can be selected to achieve the purpose of this embodiment according to actual needs.

[0063] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0064] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0065] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the technical scope disclosed in this application. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

[0066] Furthermore, although the operations of the method of this application are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

Claims

1. An electronic photo frame angle adjustment method based on human body movement tracking, characterized in that, The method comprises the following steps: S10, obtaining the center of gravity distribution data in the user movement process, processing the center of gravity distribution data, and obtaining the center of gravity distribution data set; S20, determining the center of gravity distribution characteristics based on the center of gravity distribution data set; S30, based on the center of gravity distribution characteristics, analyzing the intermittent static time length in the user movement process, and performing segmentation labeling to form an intermittent feature time sequence to determine the boundary points of static and movement and obtain the boundary point data; S40, extracting the movement intermittent features in the intermittent feature time sequence, combining the boundary point data, identifying the center of gravity fluctuation mode, and generating the fluctuation mode distribution information; S50, based on the fluctuation mode distribution information, using a segmentation adjustment mechanism to generate an initial angle adjustment scheme; S60, based on the initial angle adjustment scheme, tracking the dynamic posture of the user in real time to obtain a dynamic posture estimation result; S70, based on the dynamic posture estimation result, determining a final angle adjustment path; S80, based on the final angle adjustment path, determining an angle adjustment opportunity; S90, based on the angle adjustment opportunity and the final angle adjustment path, adjusting the pitch angle of the electronic photo frame.

2. The method of claim 1, wherein, Step S10 comprises: S110, based on a multi-sensor fusion technology, collecting the center of gravity distribution data in the user movement process to obtain original center of gravity signal data; S120, based on the original center of gravity signal data, classifying and processing different postures to obtain a phased center of gravity signal data set; S130, based on a preset filtering method, denoising the phased center of gravity signal set to obtain a denoised center of gravity signal data set; S140, extracting features from the denoised center of gravity signal data set to determine initial center of gravity distribution data; S150, if the initial center of gravity distribution data does not match the preset threshold range, correcting the initial center of gravity distribution data based on a support vector machine algorithm to obtain a final center of gravity distribution data set.

3. The method of claim 1, wherein, Step S30 comprises: S310, based on the center of gravity signal data set, obtaining relevant records of user movement, the relevant records comprising a plurality of position information and a plurality of time stamps in the user movement process; S320, preliminarily cleaning the plurality of position information and the plurality of time stamps to eliminate outliers to obtain a cleaned movement trajectory data set; S330, based on the cleaned movement trajectory data set, analyzing the distribution of user movement and intermittent static, if the speed is lower than a preset threshold, determining as a static stage to obtain an initial static stage division result; S340, based on the initial static stage division result, obtaining the duration of each static stage, combining the corresponding plurality of time stamps within the duration to analyze the time length and determine the start and end points of each static stage to form a static time length sequence; S350, based on the static time length sequence, segmenting and labeling the duration of each static stage to obtain a time length distribution characteristic, and using a clustering algorithm to group the time length distribution characteristics to obtain an intermittent feature time sequence; S360, based on the intermittent feature time sequence, analyzing the switching points of the user's static and moving stages to determine the boundary points of static and movement and obtain boundary point data.

4. The method of claim 3, wherein, Step S40 includes: S410, extracting a moving intermittent feature from the intermittent feature time sequence, and preprocessing the moving intermittent feature; S420, based on a time window segmentation method, dividing the preprocessed moving intermittent feature into multiple time segments to obtain an initial moving intermittent feature set; S430, based on the initial intermittent feature set, dynamically analyzing each time segment of the moving intermittent by a sliding window technique to determine the trend of the feature change of the moving intermittent; S440, based on the boundary point data, if the boundary point data and the trend of the intermittent feature overlap in time points, recording the fluctuation intensity of the overlapping area to obtain preliminary distribution information of the gravity fluctuation; S450, based on the preliminary distribution information of the gravity fluctuation, analyzing the specific form of the fluctuation pattern, and using a K-means clustering algorithm to classify the fluctuation intensity to determine the fluctuation pattern category in different time segments to identify the gravity fluctuation pattern; S460, based on the gravity fluctuation pattern, generating fluctuation pattern distribution information.

5. The method of claim 1, wherein, Step S50 includes: S510, based on the fluctuation pattern distribution information, obtaining the feature distribution of the fluctuation data, marking the data with strong fluctuation, and recording the position interval with the fluctuation mark; S520, based on the position interval with the fluctuation mark, using an interval division tool to generate an adjustment window range to obtain the boundary data of the adjustment window; S530, based on the boundary data of the adjustment window, if the position of the fluctuation mark exceeds the preset threshold range, using a segmented adjustment mechanism to segment the pitch angle of the electronic photo frame, and determining the pitch angle interval that needs to be adjusted; S540, based on the pitch angle interval that needs to be adjusted, using a segmented adjustment mechanism to calculate each pitch angle interval data to determine an initial pitch angle value; S550, based on the initial pitch angle value to obtain an adjustment parameter, if the deviation of the adjustment parameter and the initial pitch angle value exceeds the preset threshold, using a linear regression model to optimize the initial pitch angle value to obtain an optimized pitch angle value; S560, based on the optimized pitch angle value, generating an initial angle adjustment scheme.

6. The method of claim 1, wherein, Step S60 includes: S610, based on the initial angle adjustment scheme, using a sensor device to track the user's dynamic posture in real time, collecting initial angle data of the user's posture, and recording relevant information in real time based on a preset collection frequency to obtain posture estimation data; S620, based on the posture estimation data, obtaining the deviation of the initial angle and the standard posture to determine the direction and amplitude of the angle adjustment; S630, if the deviation of the initial angle and the standard posture exceeds the preset threshold, using the sensor device to obtain the change information of the user's posture in real time to determine whether it meets the expected trajectory of the adjustment scheme; S640, if the change information of the user posture does not conform to the expected trajectory of the adjustment scheme, adjusting the acquisition parameters of the real-time tracking module based on a feedback mechanism; S650, based on the adjusted acquisition parameters, obtaining a dynamic posture estimation result.

7. The method of claim 1, wherein, Step S70 includes: S710, obtaining initial data based on the dynamic posture estimation result, the initial data including angle change information; S720, extracting the angle change information to obtain an initial angle change sequence; S730, generating a preliminary angle change trend graph based on the initial angle change sequence, and determining a fluctuation node; S740, based on the fluctuation node, performing data analysis within an adjustment window, if the angle fluctuation is detected to exceed a preset threshold, performing local adjustment on the data within the adjustment window to obtain an adjusted fluctuation sequence; S750, performing smoothing processing on the adjusted fluctuation sequence, and generating a smoothed angle adjustment curve using a moving average method; S760, performing secondary smoothing processing on the angle change of the curve within the corresponding adjustment window in the angle adjustment curve to determine a final angle adjustment path.

8. The method of claim 1, wherein, Step S90 includes: S910, based on the angle adjustment timing and the final angle adjustment path, predicting the amplitude and frequency of angle adjustment to obtain a control parameter combination; S920, based on the control parameter combination, adjusting the pitch angle of the electronic photo frame, and simultaneously monitoring the angle change in real time during the adjustment, if the trend of the angle change does not reach a preset stability standard, re-determining the control parameters; S930, based on the re-determined control parameters, continuously iteratively adjusting the pitch angle of the electronic photo frame, and collecting the adjusted pitch angle data, analyzing the pitch angle data to determine whether the angle adjustment timing is appropriate, if not, adjusting again until the final angle adjustment timing is determined; S940, at the final angle adjustment timing, adjusting the pitch angle of the electronic photo frame according to the final angle adjustment path.

9. An electronic photo frame angle adjustment system based on human body movement tracking, characterized in that, It includes: A data acquisition module acquires the center of gravity distribution data during user movement, processes the center of gravity distribution data, and acquires a center of gravity distribution data set; A feature determination module is configured to determine a center of gravity distribution feature based on the center of gravity distribution data set; A boundary point determination module is configured to analyze the intermittent stationary duration during user movement based on the center of gravity distribution feature, perform segmentation labeling, form an intermittent feature time sequence, determine the boundary points of stationary and movement, and acquire boundary point data; A pattern recognition module is configured to extract the movement intermittent features in the intermittent feature time sequence, identify the center of gravity fluctuation pattern in combination with the boundary point data, and generate fluctuation pattern distribution information; A scheme generation module is configured to generate an initial angle adjustment scheme based on the fluctuation pattern distribution information using a segmented adjustment mechanism; A posture acquisition module is configured to track the dynamic posture of a user in real time based on the initial angle adjustment scheme, and obtain a dynamic posture estimation result; A path determination module is configured to determine a final angle adjustment path based on the dynamic posture estimation result. A denoising module is configured to determine an angle adjustment opportunity based on the final angle adjustment path. An angle adjustment module is configured to adjust the pitch angle of the electronic photo frame based on the angle adjustment opportunity and the final angle adjustment path.

Images