Fall detection method and device, electronic equipment and storage medium

Abstract

The invention relates to a tumble detection method and device, electronic equipment and a storage medium, the tumble detection method is applied to wearable equipment, the tumble detection method comprises the following steps: obtaining original motion data, the original motion data being used for representing a motion direction and a motion acceleration; in response to the original motion data meeting a preset signal strength condition, awakening a fall detection mode; and in the fall detection mode, determining a fall detection result based on the original motion data. According to the tumble detection method, the tumble detection mode is awakened only when the original motion data meets the preset signal intensity condition, pre-screening of the original motion data is achieved, and whether the tumble detection mode needs to be awakened to determine the tumble detection result or not can be judged according to the signal intensity; the power consumption of the device when the fall detection mode is not awakened is reduced, the endurance time of the device is ensured, and the user experience is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of electronic devices, and particularly to a fall detection method, apparatus, electronic device, and storage medium. Background Art

[0002] In recent years, with the popularization of intelligent devices and the rapid development of related technologies, wearable devices with health monitoring functions have been widely used in people's daily lives. As one of the major hidden dangers affecting the health and safety of the elderly, falls need to be monitored in real time through wearable devices to provide timely assistance and rescue to users.

[0003] However, the fall detection methods using related technologies to identify fall actions have problems such as high device power consumption, resulting in shortened device battery life and poor user experience. Summary of the Invention

[0004] To overcome the problems existing in the related technologies, the present disclosure provides a fall detection method, apparatus, electronic device, and storage medium.

[0005] According to a first aspect of an embodiment of the present disclosure, a fall detection method is provided, which is applied to a wearable device. The fall detection method includes:

[0006] Obtain original motion data, where the original motion data is used to characterize the motion direction and motion acceleration;

[0007] In response to the original motion data satisfying a preset signal strength condition, wake up the fall detection mode;

[0008] In the fall detection mode, based on the original motion data, determine a fall detection result, where the fall detection result is used to characterize whether a fall has occurred.

[0009] In some embodiments of the present disclosure, the obtaining of the original motion data includes:

[0010] In response to the wearable device being in a worn state, collect the original motion data at a fixed preset acquisition frequency, where the preset acquisition frequency is lower than 50HZ.

[0011] In some embodiments of the present disclosure, the waking up of the fall detection mode in response to the original motion data satisfying a preset signal strength condition includes:

[0012] Calculate parameter values of a plurality of different signal strength parameters corresponding to the original motion data within a first preset time range, where each parameter value of the signal strength parameter has a corresponding parameter value threshold, and the signal strength parameter is used to characterize the signal strength of the original motion data;

[0013] In response to the parameter values of the preset number of the signal strength parameters all being greater than or equal to the corresponding parameter value thresholds, the fall detection mode is awakened, and the original motion data within the first preset time range is determined as target motion data.

[0014] In some embodiments of the present disclosure, the multiple different signal strength parameters include multiple first data statistical features, and different ones of the first data statistical features are used to characterize different data statistical methods for the original motion data.

[0015] In some embodiments of the present disclosure, determining the fall detection result based on the original motion data includes:

[0016] Feature extraction is performed on the target motion data to obtain statistical feature values corresponding to multiple second data statistical features, and different ones of the second data statistical features are used to characterize different data statistical methods for the target motion data;

[0017] Based on each of the statistical feature values and a pre-trained classification tree model, the fall detection result is determined.

[0018] In some embodiments of the present disclosure, the training samples of the pre-trained classification tree model include sample data corresponding to fall actions and sample data corresponding to false fall actions.

[0019] In some embodiments of the present disclosure, after determining the fall detection result based on the original motion data, the fall detection method further includes:

[0020] Switch the fall detection mode to the waiting-to-be-awakened state.

[0021] In some embodiments of the present disclosure, after determining the fall detection result based on the original motion data, the fall detection method further includes:

[0022] If the fall detection result indicates that a fall has occurred, in response to continuously being in a fallen state within a second preset time range after the fall, a fall call for help operation is initiated.

[0023] In some embodiments of the present disclosure, initiating a fall call for help operation in response to continuously being in a fallen state within a second preset time range after the fall includes:

[0024] Calculate the statistical feature values of multiple third data statistical features corresponding to the original motion data within the second preset time range, different ones of the third data statistical features are used to characterize different data statistical methods for the original motion data, and each of the third data statistical features has a corresponding statistical feature threshold;

[0025] In response to the statistical feature values of each of the third data statistical features being less than the corresponding statistical feature thresholds, initiate a fall call for help operation.

[0026] In some embodiments of the present disclosure, the original motion data includes three-axis acceleration data, and the three-axis acceleration data is detected by an acceleration sensor; and / or,

[0027] Before acquiring the original motion data, the fall detection method further includes:

[0028] In response to enabling the function of performing fall detection through the fall detection mode, the fall detection mode enters a waiting wake-up state.

[0029] According to a second aspect of the embodiments of the present disclosure, there is provided a fall detection device, which is applied to a wearable device. The fall detection device includes:

[0030] An acquisition module, which is used to acquire original motion data, and the original motion data is used to characterize the motion direction and motion acceleration;

[0031] A wake-up module, which is used to wake up the fall detection mode in response to the original motion data satisfying a preset signal strength condition;

[0032] A determination module, which is used to determine a fall detection result based on the original motion data in the fall detection mode, and the fall detection result is used to characterize whether a fall has occurred.

[0033] According to a third aspect of the embodiments of the present disclosure, there is provided an electronic device, and the electronic device includes:

[0034] A processor;

[0035] A memory for storing instructions executable by the processor;

[0036] Wherein, the processor is configured to:

[0037] Acquire original motion data, and the original motion data is used to characterize the motion direction and motion acceleration;

[0038] In response to the original motion data satisfying a preset signal strength condition, wake up the fall detection mode;

[0039] In the fall detection mode, determine a fall detection result based on the original motion data, and the fall detection result is used to characterize whether a fall has occurred.

[0040] According to a fourth aspect of the embodiments of the present disclosure, there is provided a non-transitory computer-readable storage medium. When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute a fall detection method, and the fall detection method includes:

[0041] Obtain original motion data, where the original motion data is used to characterize the motion direction and motion acceleration;

[0042] In response to the original motion data satisfying a preset signal strength condition, wake up the fall detection mode;

[0043] In the fall detection mode, based on the original motion data, determine a fall detection result, where the fall detection result is used to characterize whether a fall has occurred.

[0044] The technical solutions provided by the embodiments of the present disclosure may include the following beneficial effects: The fall detection method of the present disclosure wakes up the fall detection mode only when the original motion data satisfies a preset signal strength condition, realizing pre-screening of the original motion data, and being able to determine whether to wake up the fall detection mode for determining the fall detection result according to the signal strength, reducing the device power consumption when the fall detection mode is not woken up, ensuring the device battery life, and improving the user experience.

[0045] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present invention, and are used together with the specification to explain the principles of the present invention.

[0047] Figure 1 is a flowchart of a fall detection method shown according to an exemplary embodiment.

[0048] Figure 2 is a flowchart of waking up the fall detection mode in response to the original motion data satisfying a preset signal strength condition shown according to an exemplary embodiment.

[0049] Figure 3 is a flowchart of determining a fall detection result based on the original motion data shown according to an exemplary embodiment.

[0050] Figure 4 is a flowchart of initiating a fall call for help operation in response to remaining in a fallen state continuously within a second preset time range after a fall occurs shown according to an exemplary embodiment.

[0051] Figure 5 is a flowchart of a fall detection method shown according to another exemplary embodiment.

[0052] Figure 6 is a block diagram of a fall detection device shown according to an exemplary embodiment.

[0053] Figure 7 is a block diagram of an electronic device shown according to an exemplary embodiment.

[0054] In the figure:

[0055] 10 - Acquisition module; 20 - Wake - up module; 30 - Determination module; 101 - Processing component; 102 - Memory; 103 - Power component; 104 - Multimedia component; 105 - Audio component; 106 - Input / output interface; 107 - Sensor component; 108 - Communication component; 109 - Processor. Detailed implementation manners

[0056] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the present invention. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0057] In recent years, with the popularization of wearable devices such as watches and bracelets and the rapid development of their related technologies, wearable devices with health monitoring functions such as pulse and blood pressure monitoring have been widely used in people's daily lives. With the aggravation of the aging population, falls have become a major hidden danger affecting the health and safety of the elderly. Due to the characteristics of convenient wearing and signal acquisition, wearable devices can realize all - weather identification and monitoring of fall actions to provide timely rescue and assistance to users.

[0058] In the related art, corresponding motion data can be obtained through a variety of different sensors, and different algorithms or models can be selected to process and analyze the motion data. For example, different methods such as the hidden Markov model, time - series analysis, or support vector machine (SVM) can be used to identify and judge fall actions.

[0059] However, when using the above - mentioned detection methods to identify fall actions, continuous high - intensity data collection and operations are required, resulting in high overall power consumption of the wearable device. Limited by the volume of the wearable device battery, the battery life of the wearable device will be severely shortened, and the user experience is poor.

[0060] Based on this, an exemplary embodiment of the present disclosure provides a fall detection method. By acquiring raw motion data, it can wake up the fall detection mode when the raw data meets the preset signal strength condition, and determine the fall detection result according to the raw motion data in the fall detection mode, realizing the detection and recognition of fall actions. Only when the raw motion data meets the preset signal strength condition, the fall detection mode is woken up, realizing the pre-screening of the raw motion data. It can judge whether it is necessary to wake up the fall detection mode to determine the fall detection result according to the signal strength, reducing the device power consumption when the fall detection mode is not woken up, ensuring the device battery life, and improving the user experience.

[0061] In an exemplary embodiment, a fall detection method is provided, which is applied to a wearable device. The wearable device can be, for example, a device such as a watch, a bracelet, a ring, etc. that can be worn on a user's body part. Refer to Figure 1 As shown, the fall detection method includes:

[0062] S100. Acquire raw motion data, which is used to characterize the motion direction and motion acceleration.

[0063] In step S100, the raw motion data can be acquired in real time, for example, by a sensor in the wearable device. The raw motion data can characterize the motion direction and motion acceleration. When the wearable device is a device such as a watch or a bracelet worn on the user's wrist, the raw motion data can simultaneously characterize the motion direction and motion acceleration of the wearable device and the user's wrist. Therefore, it can be determined whether a fall has occurred by processing and analyzing the raw motion data subsequently. Exemplarily, the raw motion data can include, for example, acceleration data in three mutually perpendicular axial directions.

[0064] S200. In response to the raw motion data meeting the preset signal strength condition, wake up the fall detection mode.

[0065] In step S200, a preliminary analysis of the raw motion data can determine whether the raw motion data meets the preset signal strength condition, that is, whether the signal strength of the raw motion data meets the requirement for fall judgment. When the raw motion data meets the preset signal strength condition, the fall detection mode is woken up. The fall detection mode is used to judge whether a fall has occurred according to a preset algorithm. If the raw motion data does not meet the preset signal strength condition, it means that the signal strength of the raw motion data does not meet the requirement for fall judgment. At this time, it can be determined that no fall has occurred without subsequent judgment, and the fall detection mode is not woken up, so that the wearable device continues to operate with low power consumption until the newly acquired raw motion data meets the preset signal strength condition.

[0066] S300. In the fall detection mode, based on the original motion data, determine a fall detection result, which is used to characterize whether a fall has occurred.

[0067] In step S300, after waking up and entering the fall detection mode, the wearable device can determine a fall detection result according to the original motion data through an algorithm preset in the fall detection mode. The fall detection result is used to characterize whether a fall has occurred. Exemplarily, the fall detection result can be displayed on the display screen of the wearable device or presented to the user by means such as the indicator light flashing or emitting a prompt sound.

[0068] In this embodiment, by acquiring the original motion data, the fall detection mode can be woken up when the original data meets the preset signal strength condition, and the fall detection result can be determined according to the original motion data in the fall detection mode, realizing the detection and recognition of fall actions. Only waking up the fall detection mode when the original motion data meets the preset signal strength condition realizes the pre-screening of the original motion data, can judge whether it is necessary to wake up the fall detection mode to determine the fall detection result according to the signal strength, reduces the device power consumption when the fall detection mode is not woken up, ensures the device battery life, and improves the user experience.

[0069] In some embodiments, the original motion data includes three-axis acceleration data, which is detected by an acceleration sensor.

[0070] The original motion data includes three-axis acceleration data. At each data acquisition moment, the acceleration data in three axes are synchronously acquired by the acceleration sensor. The motion direction and motion acceleration can be characterized by the three-axis acceleration direction and subsequent judgment on whether a fall has occurred can be made. Exemplarily, the three axes can be, for example, the Z-axis direction along the altitude direction, the X-axis direction along the horizontal direction and perpendicular to the Z-axis direction, and the Y-axis direction along the horizontal direction and perpendicular to both the X-axis direction and the Z-axis direction.

[0071] In this embodiment, the original motion data includes three-axis acceleration data detected by the acceleration sensor, realizing the acquisition of the original motion data. Subsequently, the fall detection result can be determined through the three-axis acceleration data. Compared with the related art where it is necessary to simultaneously acquire the three-axis angular velocity data by a gyroscope, the acquisition efficiency of the original motion data and the subsequent processing efficiency are improved, and the power consumption generated by acquiring and processing the original motion data is reduced on the premise of ensuring the realization of fall recognition.

[0072] In some embodiments, acquiring the original motion data includes: in response to the wearable device being in a worn state, acquiring the original motion data at a fixed preset acquisition frequency, and the preset acquisition frequency is lower than 50HZ.

[0073] When the user is not wearing the wearable device, the original motion data obtained by the wearable device can only characterize the motion state of the wearable device and cannot simultaneously characterize the motion state of the user. At this time, obtaining the original motion data is meaningless and will cause waste of power consumption. Therefore, it is possible to detect whether the wearable device is in a worn state by detecting the optoelectronic signal or detecting the capacitance, and obtain the original motion data when the wearable device is in a worn state.

[0074] The original motion data is acquired at a fixed preset acquisition frequency, and the preset acquisition frequency is set to be lower than 50HZ, that is, the three-axis acceleration data is acquired 50 times per second. The original motion data can be obtained by using an acceleration sensor with a relatively low power consumption and a maximum data acquisition frequency of 50HZ, reducing the power consumption generated by obtaining the original motion data.

[0075] In this embodiment, the original motion data is only obtained when the wearable device is in a worn state, which can avoid misjudgment of the fall detection result in scenarios such as throwing and dropping when the wearable device is not worn, and prevent waste of power consumption caused by the continuous acquisition of the original motion data when the wearable device is in an unworn state. In addition, the original motion data is obtained at a preset acquisition frequency of up to 50HZ, and the original motion data can be obtained by using an acceleration sensor with a relatively low power consumption, further ensuring the battery life of the wearable device.

[0076] In some embodiments, as shown in Figure 2 response to the original motion data satisfying the preset signal strength condition, the fall detection mode is awakened, including:

[0077] S210. Calculate the parameter values of multiple different signal strength parameters corresponding to the original motion data within the first preset time range. Each parameter value of the signal strength parameter has a corresponding parameter value threshold, and the signal strength parameter is used to characterize the signal strength of the original motion data.

[0078] In step S210, the original motion data within a certain time window is selected to determine whether it satisfies the preset signal strength condition, and the original motion data within the first preset time range is calculated to obtain the parameter values of multiple different signal strength parameters corresponding thereto. The first preset time range can be, for example, 5 seconds. The signal strength parameter can characterize the signal strength of the original motion data. The greater the signal strength of the original motion data, the smaller the noise of this part of the original motion data and the more intense the motion it represents, and the greater the possibility of a fall.

[0079] It should be noted that if the original motion data includes three-axis acceleration data, the parameter values of multiple signal strength parameters corresponding to the acceleration data on each axis need to be calculated separately.

[0080] S220. In response to the parameter values of a preset number of signal strength parameters being greater than or equal to the corresponding parameter value thresholds, the fall detection mode is awakened, and the original motion data within the first preset time range is determined as target motion data.

[0081] The preset number can be adjusted according to the number of signal strength parameters. Exemplarily, if the number of signal strength parameters is n, the preset number can be set to m, where both n and m are positive integers and m ≤ n. By calculating the parameter values of n signal strength parameters from 250 three-axis acceleration data within 5 seconds, when the parameter values of m, m + 1,..., n signal strength parameters among the n signal strength parameters are all greater than their respective parameter value thresholds, it means that the signal strength of the original motion data, that is, the possibility of a fall, is large enough. Then it is considered that the preset signal strength condition is met at this time, the fall detection mode is awakened, and the original motion data within these 5 seconds is marked as target motion data, which is used for the subsequent determination of the fall detection result.

[0082] In this embodiment, the signal strength parameter can characterize the signal strength of the original motion data. By calculating the parameter values of multiple different signal strength parameters corresponding to the original motion data within the first preset time range, it is possible to determine whether the preset signal strength condition is met according to the number of strength parameters greater than or equal to the corresponding parameter value thresholds, thereby deciding whether to awaken the fall detection mode, ensuring the rationality when awakening the fall detection mode. Determining the original motion data within the first preset time range as target motion data marks the target motion data as part of the original motion data for subsequent determination of the fall detection result, providing a basis for the subsequent determination of the fall detection result.

[0083] In some embodiments, the multiple different signal strength parameters include multiple first data statistical features, and different first data statistical features are used to characterize different data statistical methods for the original motion data.

[0084] The different signal strength parameters can correspond to different first data statistical features. Different first data statistical features are used to characterize different data statistical methods for the original motion data. The statistical feature values of the first data statistical features can be obtained by calculating the original motion data according to different data statistical methods. The first data statistical feature can characterize the distribution or related features among the original motion data within the first preset time range, and thus can characterize the signal strength of the original motion data.

[0085] Exemplarily, the first data statistical feature may include, for example, data statistical methods such as maximum value, minimum value, difference between maximum value and minimum value, average value, and standard deviation. If p three-axis acceleration data (ax1, ay1, az1), (ax2, ay2, az2), …, (axp, ayp, azp) are obtained within the first preset time range, then the maximum value, minimum value, difference between maximum value and minimum value, average value, and standard deviation are calculated for the acceleration data (ax1, ax2, …, axp), (ay1, ay2, …, ayp), (az1, az2, …, azp) in three different axial directions respectively, and the parameter values of the maximum value, minimum value, difference between maximum value and minimum value, average value, and standard deviation of the acceleration in the X-axis, Y-axis, and Z-axis directions are obtained as the parameter values of 15 signal strength parameters. If the parameter values of a preset number of signal strength parameters among the 15 signal strength parameters are all greater than or equal to the corresponding parameter value thresholds, it is considered that the preset signal strength condition is satisfied at this time, the fall detection mode is awakened, and the three-axis acceleration data within this first preset time is marked as target motion data.

[0086] In this embodiment, different first data statistical features are used as different signal strength parameters, realizing the calculation of the parameter values of multiple signal strength parameters, and being able to characterize the signal strength of the original motion data within the first preset time through the first data statistical feature, providing a basis for the judgment of whether to awaken the fall detection mode.

[0087] In some embodiments, refer to Figure 3 As shown, based on the original motion data, determining the fall detection result includes:

[0088] S310. Extract features from the target motion data to obtain statistical feature values corresponding to multiple second data statistical features. Different second data statistical features are used to characterize different data statistical methods for the target motion data.

[0089] In step S310, feature extraction is performed on the target motion data determined when the fall detection mode is awakened, and statistical feature values corresponding to multiple second data statistical features are obtained. Different second data statistical features are used to characterize different data statistical methods for the target motion data. The statistical feature values of the second data statistical features can be obtained by calculating the target motion data according to different data statistical methods. The second data statistical features can characterize the distribution or related features among the respective target motion data. Exemplarily, when the target motion data is triaxial acceleration data within a first preset time range, feature extraction can be performed on the target motion data to calculate the statistical feature values corresponding to the second data statistical features. The second data statistical features can include, for example, data statistical methods such as triaxial correlation coefficients, kurtosis, skewness, the sum of data above the average, and the sum of data below the average. Among them, the triaxial correlation coefficients include the correlation coefficient between the X-axis and the Y-axis, the correlation coefficient between the X-axis and the Z-axis, and the correlation coefficient between the Y-axis and the Z-axis. Taking the correlation coefficient r(X,Y) between the X-axis and the Y-axis as an example, r(X,Y) can be calculated by the formula shown below:

[0090]

[0091] where Cov(X,Y) is the covariance between the acceleration data on the X-axis and the acceleration data on the Y-axis, Var[X] is the variance of the acceleration data on the X-axis, and Var[Y] is the variance of the acceleration data on the Y-axis.

[0092] S320. Determine the fall detection result based on each statistical feature value and the pre-trained classification tree model.

[0093] In step S320, the classification tree model, as a pre-trained supervised learning algorithm, is used to handle classification problems. By learning the potential relationship between the attribute set and the category in the training samples, the data is classified and predicted. The classification tree model can be, for example, a decision tree model. The decision tree model consists of decision nodes, branches, and leaf nodes. The decision nodes represent the attributes or attribute values that appear in the samples. The branches represent the test results of the attributes, and the leaf nodes store the categories corresponding to the nodes.

[0094] The fall detection result can include, for example, two classification results: a fall occurred and no fall occurred. Whether a fall occurred can be judged through the classification tree model based on the obtained statistical feature values corresponding to the second data statistical features, and the final fall detection result is output.

[0095] In this embodiment, by extracting features from the target motion data, statistical feature values corresponding to multiple second data statistical features are obtained, and the determination of the fall detection result can be achieved based on each statistical feature value and the pre-trained classification tree model. Extracting features from the target motion data enables the second data statistical features to characterize features such as the distribution pattern of the original motion data within the first preset time range and the linear relationship between different axes. Furthermore, the classification tree model can be used to classify and predict the motion state represented by the second data statistical features, accurately determining whether a fall has occurred, ensuring the accuracy of fall recognition, and improving the user experience.

[0096] In some embodiments, the training samples of the pre-trained classification tree model include sample data corresponding to fall actions and sample data corresponding to fake fall actions.

[0097] As mentioned above, the classification tree model is obtained through the pre-training method of learning the training samples. The training samples used by the classification tree model include sample data corresponding to fall actions and sample data corresponding to fake fall actions. The sample data can be, for example, data obtained by processing the sample motion data corresponding to fall actions and fake fall actions through the same feature extraction method.

[0098] The sample data corresponding to fall actions includes sample data corresponding to fall actions when falling occurs in various scenarios and various postures. Exemplarily, the various scenarios can include different scenarios such as standing fall, walking fall, jumping fall, and stair climbing fall, and the various postures can include different postures such as forward fall, backward fall, left fall, and right fall, enabling the classification tree model to cover the recognition of fall actions in various scenarios and postures.

[0099] A fake fall action is an action with a tendency to fall but without an actual fall. The fake fall action also needs to include sample data corresponding to various scenarios and postures, enabling the classification tree model to cover the recognition of fake fall actions in various scenarios and postures, avoiding misidentifying a fake fall action as a fall action, and preventing misjudgment when the user makes a fake fall action.

[0100] In this embodiment, the classification tree model is pre-trained using both the sample data corresponding to fall actions and the sample data corresponding to fake fall actions, enabling the classification tree model to cover the recognition of fall actions and fake fall actions in various scenarios and postures simultaneously, achieving the recognition and judgment of fall actions, avoiding misidentifying a fake fall action as a fall action, reducing or eliminating misjudgment caused by fake fall actions, further ensuring the accuracy of fall recognition, and improving the user experience.

[0101] In some embodiments, after determining the fall detection result based on the original motion data, the fall detection method further includes: switching the fall detection mode to the waiting wake-up state.

[0102] After determining the fall detection result, the fall detection mode has completed the judgment on whether a fall has occurred within the first preset time. If the fall detection mode continues to be kept in the wake-up state, it will cause waste of power consumption due to the continuous operation of relevant algorithm modules. At this time, it is necessary to switch the fall detection mode to the waiting-to-wake state until the subsequently acquired original motion data again meets the preset signal strength condition, and then re-wake up the fall detection mode.

[0103] In this embodiment, by switching the fall detection mode to the waiting-to-wake state after determining the fall detection result, and making the fall detection mode temporarily sleep after completing a judgment on whether a fall has occurred, it can avoid waste of power consumption caused by the continuous operation of relevant algorithm modules, further save the overall power consumption of the wearable device, ensure the device's battery life, and improve the user experience.

[0104] In some embodiments, after determining the fall detection result based on the original motion data, the fall detection method further includes: if the fall detection result indicates that a fall has occurred, in response to continuously being in a fallen state within the second preset time range after the fall, initiate a fall emergency call operation.

[0105] The fall detection result indicating that a fall has occurred means that the user made a falling motion within the first preset time range. If the user can get up or move shortly after the fall, it means that the falling motion has not had a significant impact on the user's health and safety. If the user has been in a fallen state without getting up or moving shortly after the fall, it means that the user's physical function has been affected after the fall and may have lost the ability to move. At this time, the wearable device initiates a fall emergency call operation to request help or rescue from others.

[0106] Exemplarily, the second preset time can be set to 10 seconds, for example. If the wearable device determines that the user has been in a fallen state without getting up or moving within 10 seconds after the fall, it initiates a fall emergency call operation. When initiating the fall emergency call operation, for example, send a distress request to, for example, a rescue platform, associated users, etc., and perform real-time positioning through the positioning function, and include the location information of the user when the fall occurred in the distress request. The rescue platform or associated users can go to the location where the user fell according to the distress request to rescue the user.

[0107] In this embodiment, when the fall detection result indicates that a fall has occurred, if the wearable device determines that the user has been continuously in a fallen state within the second preset time range after the fall, it initiates a fall emergency call operation, which can automatically send a rescue request when the user falls and loses the ability to move, realizes the emergency assistance function, and improves the efficiency and timeliness of rescue.

[0108] In some embodiments, refer to Figure 4As shown, in response to the user remaining in the fallen state within the second preset time range after a fall occurs, a fall alarm operation is initiated, including:

[0109] S410. Calculate the statistical feature values of multiple third data statistical features corresponding to the original motion data within the second preset time range. Different third data statistical features are used to represent different data statistical methods for the original motion data, and each third data statistical feature has a corresponding statistical feature threshold.

[0110] In step S410, the prerequisite for initiating the fall alarm operation is that the user remains in the fallen state without getting up or moving within the second preset time range after a fall occurs. Therefore, it is necessary to analyze the original motion data within the second preset time range. Calculate the statistical feature values of multiple third data statistical features corresponding to the original motion data within the second preset time range. Different third data statistical features are used to represent different data statistical methods for the original motion data. The statistical feature values of the third data statistical features can be obtained by calculating the original motion data within the second preset time range according to different data statistical methods. The third data statistical feature can characterize the distribution or related features among the original motion data within the second preset time range. Whether the user remains in the fallen state can be analyzed through the statistical feature values of the third data statistical features.

[0111] Exemplarily, the original motion data is triaxial acceleration data, the second preset time range is 10 seconds, and the third data statistical features can include data statistical methods such as maximum value, minimum value, and triaxial modulus coefficient. If q triaxial acceleration data (ax1, ay1, az1), (ax2, ay2, az2), …, (axq, ayq, azq) are obtained within the second preset time range, then the maximum value, minimum value, and triaxial modulus coefficient are calculated for the acceleration data in three different axes (ax1, ax2, …, axq), (ay1, ay2, …, ayq), (az1, az2, …, azq) respectively, to obtain the maximum value, minimum value of the acceleration in the X axis, Y axis, and Z axis, and the triaxial modulus coefficient of the XYZ axis. Among them, for the triaxial modulus coefficient, the average value of the acceleration data in each axis needs to be calculated first, and the average value of the acceleration in the X axis The average value of the acceleration in the Y axis And the average value of the acceleration in the Z axis And the triaxial modulus coefficient Z is further obtained according to the average value of the acceleration in each axis xyz , the triaxial modulus coefficient Z xyz Can be calculated through the formula shown as follows:

[0112]

[0113] S420. Initiate a fall emergency call operation in response to the statistical feature values of all the third data statistical features being less than their corresponding statistical feature thresholds.

[0114] In step S420, the statistical feature values of multiple calculated third data statistical features each have a corresponding statistical feature threshold. The statistical feature values of each third data statistical feature are compared with their corresponding statistical feature thresholds. If all the statistical feature values are less than their corresponding statistical feature thresholds, it indicates that the movement amplitude of the user represented by the original motion data within the second preset time is small, and the possibility of not getting up or moving is high. At this time, it can be considered that the user may have lost the ability to move, and then initiate a fall emergency call operation.

[0115] In this embodiment, by calculating the statistical feature values of multiple third data statistical features corresponding to the original motion data within the second preset time range, it is possible to determine whether the user is continuously in a continuous fall state based on the comparison results of the statistical feature values with their corresponding statistical feature thresholds, thereby providing a basis for initiating a fall emergency call operation. Each third data statistical feature can characterize the size of the movement amplitude of the user after a fall, ensuring the accuracy of initiating a fall emergency call operation and effectively preventing false triggering of the fall emergency call operation due to misjudgment.

[0116] In some embodiments, after initiating a fall emergency call operation, the fall detection method further includes: in response to the fall emergency call operation being answered or ended, continue to determine whether to wake up the fall detection mode.

[0117] As described above, after determining the fall detection result, the fall detection mode is switched to the waiting-to-wake-up state. If a fall occurs and the user remains in a fall state within the second preset time range after the fall, a fall emergency call operation is initiated. The emergency call operation can be answered by a rescue platform that is the recipient of the emergency call operation or an associated user, and the emergency call operation can also be actively ended in response to the user performing an emergency call end operation on the wearable device. In both cases, continue to determine whether to wake up the fall detection mode, that is, start to obtain new original motion data again, calculate the parameter values of the signal strength parameters corresponding to the updated original motion data and compare them with the corresponding parameter value thresholds, so that the judgment process of waking up the fall detection mode can continue to execute after a complete fall process ends.

[0118] In this embodiment, by continuing to determine whether to wake up the fall detection mode when the fall emergency call operation is answered or ended, the judgment process of waking up the fall detection mode can continue to execute after a complete fall process ends, avoiding the situation where a fall occurs again after the fall emergency call operation is answered or ended but is not detected because the fall detection mode fails to wake up, realizing continuous coverage of the fall detection function, being able to provide timely and effective help to the user every time a fall occurs, and improving the user experience.

[0119] In some embodiments, before acquiring the original motion data, the fall detection method further includes: in response to enabling the function of performing fall detection through the fall detection mode, the fall detection mode enters a waiting wake-up state.

[0120] The user can actively control the enabling of the fall detection function. When the user wants to perform fall detection and enables the function of performing fall detection through the fall detection mode, the fall detection mode enters a waiting wake-up state, so that the fall detection mode can be woken up when the original motion data meets the preset signal strength condition. Exemplarily, the fall detection function can be enabled, for example, by turning on the corresponding switch. The corresponding switch of the fall detection function can be a hardware switch provided on the wearable device or a control switch displayed on the display screen of the wearable device.

[0121] In this embodiment, the wearable device can respond to enabling the function of performing fall detection through the fall detection mode, so that the fall detection mode enters a waiting wake-up state, can actively enable the function of performing fall detection through the fall detection mode when fall monitoring is required, and save the power consumption of the wearable device when fall monitoring is not required, realizing the active control of enabling the fall detection function and improving the user experience.

[0122] In an exemplary embodiment, a fall detection method is provided, which is applied to a wearable device. Refer to Figure 5 As shown, the fall detection method includes:

[0123] S1. In response to enabling the function of performing fall detection through the fall detection mode, the fall detection mode enters a waiting wake-up state;

[0124] S2. In response to the wearable device being in a worn state, acquire the original motion data at a fixed preset acquisition frequency, and the preset acquisition frequency is lower than 50 HZ;

[0125] S3. Calculate the parameter values of a plurality of different signal strength parameters corresponding to the original motion data within a first preset time range. Each parameter value of the signal strength parameter has a corresponding parameter value threshold, and the signal strength parameter is used to characterize the signal strength of the original motion data;

[0126] S4. In response to the parameter values of the preset number of signal strength parameters being greater than or equal to the corresponding parameter value thresholds, wake up the fall detection mode and determine the original motion data within the first preset time range as the target motion data;

[0127] S5. Extract features from the target motion data to obtain the statistical feature values corresponding to a plurality of second data statistical features;

[0128] S6. Determine the fall detection result based on each statistical feature value and the pre-trained classification tree model;

[0129] S7. Switch the fall detection mode to the waiting-for-wake-up state;

[0130] S8. In response to the fall detection result indicating a fall, calculate the statistical feature values of multiple third data statistics corresponding to the original motion data within the second preset time range, and each third data statistic has a corresponding statistical feature threshold;

[0131] S9. In response to the statistical feature value of each third data statistic being less than the corresponding statistical feature threshold, initiate a fall calling-for-help operation;

[0132] S10. In response to the fall calling-for-help operation being answered or ended, continue to execute S3;

[0133] S11. In response to the fall detection result indicating no fall, continue to execute S3.

[0134] In this embodiment, by obtaining the original motion data, the fall detection mode can be awakened when the original data meets the preset signal strength condition, and the fall detection result can be determined based on the original motion data in the fall detection mode, realizing the detection and recognition of fall actions. Only when the original motion data meets the preset signal strength condition is the fall detection mode awakened, realizing the pre-screening of the original motion data, and it is possible to judge whether it is necessary to awaken the fall detection mode to determine the fall detection result according to the signal strength, reducing the device power consumption when the fall detection mode is not awakened, ensuring the device battery life, and improving the user experience.

[0135] In an exemplary embodiment, as shown in Figure 6 a fall detection device is provided, which is applied to a wearable device. The fall detection device includes an acquisition module 10, a wake-up module 20, and a determination module 30. The acquisition module 10 is used to acquire original motion data, and the original motion data is used to represent the motion direction and motion acceleration. The wake-up module 20 is used to wake up the fall detection mode in response to the original motion data meeting the preset signal strength condition. The determination module 30 is used to determine the fall detection result based on the original motion data in the fall detection mode, and the fall detection result is used to represent whether a fall has occurred.

[0136] In this embodiment, the original motion data is obtained by the obtaining module 10. When the original data meets the preset signal strength condition, the fall detection mode can be awakened by the awakening module 20, and the fall detection result can be determined by the determining module 30 according to the original motion data in the fall detection mode, realizing the detection and recognition of the fall action. Only when the original motion data meets the preset signal strength condition is the fall detection mode awakened, realizing the pre-screening of the original motion data, and being able to judge whether it is necessary to awaken the fall detection mode to determine the fall detection result according to the signal strength, reducing the device power consumption when the fall detection mode is not awakened, ensuring the device battery life, and improving the user experience.

[0137] In one embodiment, the obtaining module 10 is further configured to: in response to the wearable device being in a worn state, collect the original motion data at a fixed preset collection frequency, and the preset collection frequency is lower than 50HZ.

[0138] In one embodiment, the awakening module 20 is further configured to: calculate the parameter values of a plurality of different signal strength parameters corresponding to the original motion data within a first preset time range, and each parameter value of the signal strength parameter has a corresponding parameter value threshold, and the signal strength parameter is used to characterize the signal strength of the original motion data. In response to the parameter values of the preset number of signal strength parameters being greater than or equal to the corresponding parameter value thresholds, awaken the fall detection mode, and determine the original motion data within the first preset time range as the target motion data.

[0139] In one embodiment, the awakening module 20 is further configured to: in response to the fall call for help operation being answered or ended, continue to judge whether to awaken the fall detection mode.

[0140] In one embodiment, the plurality of different signal strength parameters include a plurality of first data statistical features, and different first data statistical features are used to characterize different data statistical methods for the original motion data.

[0141] In one embodiment, the determining module 30 is further configured to: extract features from the target motion data to obtain the statistical feature values corresponding to a plurality of second data statistical features, and different second data statistical features are used to characterize different data statistical methods for the target motion data; based on each statistical feature value and the pre-trained classification tree model, determine the fall detection result.

[0142] In one embodiment, the training samples of the pre-trained classification tree model include the sample data corresponding to the fall action and the sample data corresponding to the fake fall action.

[0143] In some embodiments, the fall detection device further includes a mode switching module, and the mode switching module is configured to switch the fall detection mode to the waiting awakening state after determining the fall detection result based on the original motion data.

[0144] In some embodiments, the fall detection device further includes a distress calling module. The distress calling module is configured to initiate a fall distress call operation if the fall detection result indicates a fall and the device remains in the fallen state continuously within a second preset time range after the fall occurs.

[0145] In one embodiment, the distress calling module is further configured to: calculate the statistical feature values of a plurality of third data statistical features corresponding to the original motion data within the second preset time range, where different third data statistical features are used to represent different data statistical methods for the original motion data, and each third data statistical feature has a corresponding statistical feature threshold; and initiate a fall distress call operation in response to the statistical feature values of each third data statistical feature being less than the corresponding statistical feature threshold.

[0146] In one embodiment, the original motion data includes triaxial acceleration data, which is detected by an acceleration sensor.

[0147] In some embodiments, the fall detection device further includes an enabling module. The enabling module is configured to enter the waiting wake-up state of the fall detection mode in response to enabling the function of performing fall detection in the fall detection mode before acquiring the original motion data.

[0148] In an exemplary embodiment, an electronic device is provided. The electronic device can be, for example, a wearable device such as a watch, a bracelet, or a ring.

[0149] Reference Figure 7 As shown, the electronic device may include one or more of the following components: a processing component 101, a memory 102, a power component 103, a multimedia component 104, an audio component 105, an input / output (I / O) interface 106, a sensor component 107, and a communication component 108.

[0150] The processing component 101 generally controls the overall operation of the electronic device, such as operations associated with display, telephone call, data communication, camera operation, and recording operation. The processing component 101 may include one or more processors 109 to execute instructions to complete all or part of the steps of the above method. In addition, the processing component 101 may include one or more modules to facilitate the interaction between the processing component 101 and other components. For example, the processing component 101 may include a multimedia module to facilitate the interaction between the multimedia component 104 and the processing component 101.

[0151] The memory 102 is configured to store various types of data to support the operation of the electronic device. Examples of such data include instructions for any application or method operating on the electronic device, contact data, phone book data, messages, pictures, videos, and the like. The memory 102 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, a magnetic disk, or an optical disk.

[0152] The power component 103 provides power to various components of the electronic device. The power component 103 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the electronic device.

[0153] The multimedia component 104 includes a screen that provides an output interface between the electronic device and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors can sense not only the boundaries of the touch or swipe actions but also detect the duration and pressure associated with the touch or swipe operation. In some embodiments, the multimedia component 104 includes a front camera and / or a rear camera. When the electronic device is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera can receive external multimedia data. Each of the front camera and the rear camera can be a fixed optical lens system or have a focal length and optical zoom capabilities.

[0154] The audio component 105 is configured to output and / or input audio signals. For example, the audio component 105 includes a microphone (MIC) that is configured to receive external audio signals when the electronic device is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals can be further stored in the memory 102 or transmitted via the communication component 108. In some embodiments, the audio component 105 further includes a speaker for outputting audio signals.

[0155] The I / O interface 106 provides an interface between the processing component 101 and a peripheral interface module, which can be a keyboard, a click wheel, buttons, and the like. These buttons can include, but are not limited to: a home button, a volume button, a power button, and a lock button.

[0156] The sensor assembly 107 includes one or more sensors for providing a status assessment of various aspects of the electronic device. For example, the sensor assembly 107 can detect the on / off state of the electronic device, the relative positioning of components, such as the display and keypad of the electronic device, the sensor assembly 107 can also detect a change in the position of the electronic device or a component of the electronic device, the presence or absence of user contact with the electronic device, the orientation or acceleration / deceleration of the electronic device, and the temperature change of the electronic device. The sensor assembly 107 can include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 107 can also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 107 can also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0157] The communication component 108 is configured to facilitate communication between the electronic device and other devices in a wired or wireless manner. The device can access a wireless network based on communication standards, such as WiFi, 2G, or 3G, or a combination thereof. In an exemplary embodiment, the communication component 108 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 108 further includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0158] In an exemplary embodiment, the electronic device can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components for performing the above-described fall detection method applied to the electronic device.

[0159] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as the memory 102 including instructions, and the above instructions can be executed by the processor 109 of the electronic device to complete the above-described fall detection method applied to the electronic device. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device, etc. When the instructions in the storage medium are executed by the processor 109 of the electronic device, the electronic device is enabled to execute the fall detection method shown in the above embodiments.

[0160] Other embodiments of the present invention will be readily apparent to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention, which follow the general principles of the invention and include known common general knowledge or conventional technical means in the technical field not disclosed in this disclosure. The specification and examples are only to be considered as exemplary, and the true scope and spirit of the invention are pointed out by the following claims.

[0161] It should be understood that the present invention is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. A fall detection method, applied to a wearable device, characterized in that The described fall detection method includes: Obtaining original motion data, where the original motion data is used to characterize the motion direction and motion acceleration; In response to the original motion data satisfying a preset signal strength condition, waking up the fall detection mode; In the fall detection mode, based on the original motion data, determining a fall detection result, where the fall detection result is used to characterize whether a fall has occurred.

2. The fall detection method according to claim 1, wherein The obtaining of the original motion data includes: In response to the wearable device being in a worn state, collecting the original motion data at a fixed preset acquisition frequency, where the preset acquisition frequency is lower than 50HZ.

3. The fall detection method according to claim 1, characterized in that, The waking up of the fall detection mode in response to the original motion data satisfying a preset signal strength condition includes: Calculating parameter values of a plurality of different signal strength parameters corresponding to the original motion data within a first preset time range, where each parameter value of the signal strength parameter has a corresponding parameter value threshold, and the signal strength parameter is used to characterize the signal strength of the original motion data; In response to the parameter values of a preset number of the signal strength parameters being greater than or equal to the corresponding parameter value thresholds, waking up the fall detection mode and determining the original motion data within this first preset time range as target motion data.

4. The fall detection method according to claim 3, wherein The plurality of different signal strength parameters include a plurality of first data statistical features, and different ones of the first data statistical features are used to characterize different data statistical methods for the original motion data.

5. The fall detection method according to claim 3, wherein The determining of the fall detection result based on the original motion data includes: Performing feature extraction on the target motion data to obtain statistical feature values corresponding to a plurality of second data statistical features, where different ones of the second data statistical features are used to characterize different data statistical methods for the target motion data; Based on each of the statistical feature values and a pre-trained classification tree model, determining the fall detection result.

6. The fall detection method according to claim 5, characterized in that, The training samples of the pre-trained classification tree model include sample data corresponding to fall actions and sample data corresponding to fake fall actions.

7. The fall detection method according to claim 1, characterized in that After the determining of the fall detection result based on the original motion data, the fall detection method further includes: Switching the fall detection mode to a waiting-to-wake state.

8. The fall detection method according to any one of claims 1-7, characterized in that, After the determining of the fall detection result based on the original motion data, the fall detection method further includes: If the fall detection result indicates that a fall has occurred, in response to continuously being in a fallen state within a second preset time range after the fall, initiating a fall calling-for-help operation.

9. The fall detection method according to claim 8, characterized in that, The initiating of the fall calling-for-help operation in response to continuously being in a fallen state within a second preset time range after the fall includes: Calculating statistical feature values of a plurality of third data statistical features corresponding to the original motion data within the second preset time range, where different ones of the third data statistical features are used to characterize different data statistical methods for the original motion data, and each of the third data statistical features has a corresponding statistical feature threshold; In response to the statistical feature values of each of the third data statistical features being less than the corresponding statistical feature thresholds, initiating a fall calling-for-help operation.

10. The fall detection method according to any one of claims 1-7, characterized in that, The original motion data includes three-axis acceleration data, and the three-axis acceleration data is detected by an acceleration sensor; And / or Before obtaining the original motion data, the fall detection method further includes: In response to enabling the function of fall detection through the fall detection mode, the fall detection mode enters a waiting wake-up state.

11. A fall detection device, applied to a wearable device, characterized in that, The fall detection device includes: An acquisition module, which is configured to acquire original motion data, and the original motion data is used to characterize the motion direction and motion acceleration; A wake-up module, which is configured to wake up the fall detection mode in response to the original motion data satisfying a preset signal strength condition; A determination module, which is configured to determine a fall detection result based on the original motion data in the fall detection mode, and the fall detection result is used to characterize whether a fall has occurred.

12. An electronic device, characterized in that, The electronic device includes: A processor; A memory for storing processor-executable instructions; Wherein, the processor is configured to: Acquire original motion data, and the original motion data is used to characterize the motion direction and motion acceleration; Wake up the fall detection mode in response to the original motion data satisfying a preset signal strength condition; In the fall detection mode, determine a fall detection result based on the original motion data, and the fall detection result is used to characterize whether a fall has occurred.

13. A non-transitory computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor of the electronic device, the electronic device is enabled to execute a fall detection method, and the fall detection method includes: Acquire original motion data, and the original motion data is used to characterize the motion direction and motion acceleration; Wake up the fall detection mode in response to the original motion data satisfying a preset signal strength condition; In the fall detection mode, determine a fall detection result based on the original motion data, and the fall detection result is used to characterize whether a fall has occurred.

Images