CN108831527B

CN108831527B - User motion state detection method and device and wearable device

Info

Publication number: CN108831527B
Application number: CN201810556612.7A
Authority: CN
Inventors: 陈志勇; 林晨曦
Original assignee: Xiamen Greentouch Co ltd
Current assignee: Xiamen Greentouch Co ltd
Priority date: 2018-05-31
Filing date: 2018-05-31
Publication date: 2021-06-04
Anticipated expiration: 2038-05-31
Also published as: CN108831527A

Abstract

The invention discloses a user motion state detection method, a device and wearable equipment, wherein the user motion state detection method comprises the following steps: acquiring personal information and/or environmental information of a user, and determining the type of motion data required to be acquired according to the personal information and/or the environmental information; acquiring primary motion data of a user according to the motion data type, and establishing a personal portrait model by combining personal information and/or environmental information; and determining whether to acquire secondary motion data according to the model to establish a secondary artificial intelligence model. By implementing the method and the device, the types of the motion data to be acquired and analyzed can be finely screened according to the personal information of the user and the environment information, and the acquisition amount and the calculation amount are reduced through a multi-level model contained in one model, so that the consumption of electric quantity is effectively reduced.

Description

User motion state detection method and device and wearable device

Technical Field

The invention relates to the technical field of intelligent wearing, in particular to a user motion state detection method and device and wearable equipment.

Background

The recognition of the human motion posture plays an important role in the fields of motion analysis, fall early warning, disease prevention, rehabilitation, identity recognition and the like. The pressure and intensity of the sole of the human body can be changed along with the functional disorder or pathological change of the foot structure of the human body and the change of the motion state of the human body.

The dynamic and kinetic characteristics of gait can be found by researching the distribution of human plantar pressure and pressure in a static state or a motion process. With the rapid development and popularization of intelligent wearable equipment, the mass center movement, energy consumption, movement positions, joint stress conditions and the like of a human body in a walking process can be calculated through the obtained ground reaction force and information such as the coordinates of the movement positions of the human body of the existing wearable foot tool. And real-time alarms such as falling early warning are carried out according to the obtained information. However, due to the limited electric quantity of the battery, the duration of the wearable foot tool is strictly limited, which may cause that the wearable foot tool cannot be used continuously in the time period most needing to be monitored, thereby affecting the monitoring of the motion state of the user, and even affecting the early warning of the health condition of the user at the critical moment.

Disclosure of Invention

In view of this, embodiments of the present invention provide a method and an apparatus for detecting a user motion state, and a wearable device, so as to solve the problems of too fast power consumption and short sustainable use time of the conventional wearable device.

According to a first aspect, an embodiment of the present invention provides a user motion state detection method, including: acquiring personal information and/or environmental information of a user, and determining the type of motion data required to be acquired according to the personal information and/or the environmental information; acquiring actual motion data of a user according to the motion data type; and analyzing the actual motion data to determine the motion state of the user.

In the embodiment of the invention, primary motion data of a user are collected according to the type of the motion data, and a personal portrait model is established by combining personal information and/or environmental information; whether secondary motion data are collected or not is determined according to the model to establish a secondary artificial intelligence model, the types of the motion data needing to be collected and analyzed can be finely screened according to personal information of a user and environment information of the user, and the collection amount and the calculation amount are reduced through a multi-stage model contained in the model, so that the electricity consumption is effectively reduced, and the sustainable service time of the wearable equipment is prolonged.

With reference to the first aspect, in a first implementation manner of the first aspect, the personal information includes: at least one of sex information, height information, weight information and age information of the user; the environment information includes: at least one of temperature information and season information.

With reference to the first aspect or the first implementation manner of the first aspect, in a second implementation manner of the first aspect, determining a type of motion data to be acquired according to the personal information and/or the environmental information includes: extracting health history information of the user from the personal information, and determining data needing to be collected as acceleration data or both the acceleration data and the angular velocity data according to the health history information; or extracting the health history information of the user from the personal information, and determining the data to be acquired as acceleration data or both the acceleration data and the angular velocity data according to the health history information and the environmental information; or determining the data to be acquired as acceleration data or both the acceleration data and the angular velocity data according to the environment information.

With reference to the second embodiment of the first aspect, in a third embodiment of the first aspect, the actual motion data includes acceleration data; the analyzing the actual motion data to determine the motion state of the user includes: acquiring acceleration change information in a plurality of motion directions according to the acceleration data; judging whether the user is in a static state or not according to the acceleration change information; and when the user is not in a static state, judging that the motion state of the user belongs to a circular motion state or a non-circular motion state according to the acceleration change information.

With reference to the third implementation manner of the first aspect, in the fourth implementation manner of the first aspect, if the motion state of the user belongs to a circular motion state, it is determined that the motion state of the user belongs to a walking, running or abnormal walking state according to the acceleration change information.

In embodiments of the present invention, in conjunction with the acquired environmental information (e.g., season, temperature, climate, etc.), the type of athletic data to be acquired may be filtered. In the subsequent data analysis process, the current motion state of the user can be judged by combining the environmental information and only through a preliminary analysis step, and data support can be provided for the subsequent health state analysis without further performing a detailed analysis process, so that the processing and calculation amount of the motion data is greatly reduced, the electric quantity consumption of corresponding equipment is reduced, and the cruising ability of the equipment is improved.

With reference to the third embodiment of the first aspect, in the fifth embodiment of the first aspect, the actual motion data further includes angular velocity data; the analyzing the actual motion data to determine the motion state of the user further comprises: if the motion state of the user belongs to the circular motion state, judging that the motion state of the user belongs to going upstairs, going downstairs, walking on flat ground or running according to the acceleration change information and the angular velocity change information; and if the motion state of the user belongs to the circular motion state, judging that the motion state of the user belongs to sitting, squatting, standing, jumping or falling according to the acceleration change information and the angular velocity change information.

With reference to the second, third, fourth or fifth implementation manner of the first aspect, in a sixth implementation manner of the first aspect, the user motion state detection method further includes: and determining the health state of the user according to the actual motion data of the user.

With reference to the sixth implementation manner of the first aspect, in the seventh implementation manner of the first aspect, the determining the health status of the user according to the actual exercise data of the user includes: determining a health state range to be detected according to the personal information and/or the environmental information; determining the degree of the gait of the user deviating from the normal sample crowd according to the health state range, the acceleration data and a pre-trained gait deviation model; if the degree of the gait of the user deviating from the normal sample crowd is larger than a preset threshold value, judging that the motion state of the user belongs to a normal state or a sick state according to the degree of the gait of the user deviating from the normal sample crowd and the actual motion data, otherwise, judging that the motion state of the user belongs to the normal state.

With reference to the seventh implementation manner of the first aspect, in the eighth implementation manner of the first aspect, when the health status range to be detected cannot be determined according to the personal information and/or the environmental information, the determining the health status of the user according to the actual exercise data of the user further includes: determining the degree of the gait of the user deviating from the normal sample crowd according to the acceleration data, the angular velocity data and a pre-trained gait deviation model; if the degree of the gait of the user deviating from the normal sample crowd is larger than a preset threshold value, judging that the motion state of the user belongs to a normal state or a sick state according to the degree of the gait of the user deviating from the normal sample crowd and the actual motion data, otherwise, judging that the motion state of the user belongs to the normal state.

With reference to the seventh or eighth implementation manner of the first aspect, in a ninth implementation manner of the first aspect, the method for detecting a motion state of a user further includes: verifying a judgment result for judging whether the motion state of the user belongs to a normal state or a sick state; and optimizing the weight set and the ill-conditioned threshold value respectively corresponding to each motion state determined by the SVM algorithm in advance under each ill condition according to the verification result.

With reference to the ninth implementation manner of the first aspect, in the tenth implementation manner of the first aspect, the verifying the determination result of determining that the motion state of the user belongs to a normal state or a sick state includes: re-determining the degree of the gait of the user deviating from the normal sample crowd according to the actual motion data and the updated gait deviation model; the updated gait deviation model is obtained according to the data of the re-screened sample population in the normal state; if the redetermined degree of the user gait deviating from the normal sample crowd is larger than the preset threshold, the motion state of the user is redetermined to be normal state or sick state according to the redetermined degree of the user gait deviating from the normal sample crowd and the actual motion data; and verifying the last judgment result according to the re-judged judgment result.

According to a second aspect, an embodiment of the present invention provides a user motion state detection apparatus, including: the information acquisition module is used for acquiring personal information and/or environmental information of a user and determining the type of the motion data required to be acquired according to the personal information and/or the environmental information; the actual motion data acquisition module is used for acquiring actual motion data of the user according to the motion data type; and the motion state determining module is used for analyzing the actual motion data and determining the motion state of the user.

With reference to the second aspect, in a first embodiment of the second aspect, the personal information includes: at least one of sex information, height information, weight information and age information of the user; the environment information includes: at least one of temperature information and season information.

With reference to the second aspect or the first implementation manner of the second aspect, in a second implementation manner of the second aspect, the information obtaining module is specifically configured to: extracting health history information of the user from the personal information, and determining data needing to be collected as acceleration data or both the acceleration data and the angular velocity data according to the health history information; or extracting the health history information of the user from the personal information, and determining the data to be acquired as acceleration data or both the acceleration data and the angular velocity data according to the health history information and the environmental information; or determining the data to be acquired as acceleration data or both the acceleration data and the angular velocity data according to the environment information.

With reference to the second aspect of the second embodiment, in a third embodiment of the second aspect, the actual motion data includes acceleration data; the motion state determination module comprises: the acceleration change information acquisition submodule is used for acquiring acceleration change information in a plurality of motion directions according to the acceleration data; the cyclic motion judgment submodule is used for judging whether the user is in a static state or not according to the acceleration change information; and when the user is not in a static state, judging that the motion state of the user belongs to a circular motion state or a non-circular motion state according to the acceleration change information.

With reference to the third embodiment of the second aspect, in a fourth embodiment of the second aspect, the motion state determination module further includes: and the motion state judgment sub-module judges that the motion state of the user belongs to a walking, running or abnormal walking state according to the acceleration change information if the motion state of the user belongs to a circular motion state.

With reference to the third embodiment of the second aspect, in a fifth embodiment of the second aspect, the actual motion data further includes angular velocity data; the motion state determination module further comprises: the motion state judgment sub-module judges that the motion state of the user belongs to ascending stairs, descending stairs, walking on flat ground or running according to the acceleration change information and the angular velocity change information if the motion state of the user belongs to a circular motion state; and if the motion state of the user belongs to the non-circular motion state, the motion state judgment submodule judges that the motion state of the user belongs to sitting, squatting, standing, jumping or falling according to the acceleration change information and the angular speed change information.

With reference to the second, third, fourth or fifth embodiment of the second aspect, in a sixth embodiment of the second aspect, the user motion state detection apparatus further includes: and the health state determining module is used for determining the health state of the user according to the actual motion data of the user.

With reference to the sixth implementation manner of the second aspect, in a seventh implementation manner of the second aspect, the health state determination module includes: the health state range determining submodule is used for determining a health state range to be detected according to the personal information and/or the environmental information; the health state judgment sub-module is used for determining the degree of the gait of the user deviating from the normal sample crowd according to the health state range, the acceleration data and a pre-trained gait deviation model; if the degree of the gait of the user deviating from the normal sample crowd is larger than a preset threshold value, judging that the motion state of the user belongs to a normal state or a sick state according to the degree of the gait of the user deviating from the normal sample crowd and the actual motion data, otherwise, judging that the motion state of the user belongs to the normal state.

With reference to the seventh implementation manner of the second aspect, in an eighth implementation manner of the second aspect, the health status determination module further includes: the gait deviation degree determining submodule is used for determining the degree of deviation of the gait of the user from a normal sample crowd according to the acceleration data, the angular velocity data and a pre-trained gait deviation model; and the health state determining sub-module is used for judging that the motion state of the user belongs to a normal state or a sick state according to the degree of the gait of the user deviating from the normal sample crowd and the actual motion data if the degree of the gait of the user deviating from the normal sample crowd is larger than a preset threshold value, or else, judging that the motion state of the user belongs to the normal state.

With reference to the seventh or eighth embodiment of the second aspect, in a ninth embodiment of the second aspect, the user motion state detection apparatus further includes: the verification module is used for verifying a judgment result for judging whether the motion state of the user belongs to a normal state or a sick state; and the optimization module is used for optimizing the weight set and the ill-condition threshold value which correspond to each motion state determined by the SVM algorithm in advance under each ill condition according to the verification result.

With reference to the ninth implementation manner of the second aspect, in a tenth implementation manner of the second aspect, the verification module is specifically configured to: re-determining the degree of the gait of the user deviating from the normal sample crowd according to the actual motion data and the updated gait deviation model; the updated gait deviation model is obtained according to the data of the re-screened sample population in the normal state; if the redetermined degree of the user gait deviating from the normal sample crowd is larger than the preset threshold, the motion state of the user is redetermined to be normal state or sick state according to the redetermined degree of the user gait deviating from the normal sample crowd and the actual motion data; and verifying the last judgment result according to the re-judged judgment result.

According to a third aspect, embodiments of the present invention provide a wearable device, including: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing therein computer instructions, and the processor executing the computer instructions to perform the method for detecting a motion state of a user as set forth in the first aspect or any one of the embodiments of the first aspect.

According to a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, which stores computer instructions for causing a computer to execute the method for detecting a motion state of a user described in the first aspect or any one of the implementation manners of the first aspect.

Drawings

The features and advantages of the present invention will be more clearly understood by reference to the accompanying drawings, which are illustrative and not to be construed as limiting the invention in any way, and in which:

FIG. 1 shows a flow chart of a user motion state detection method according to an embodiment of the present invention;

FIG. 2 shows a schematic view of a sole sensing area of an intelligent footwear of an embodiment of the present invention;

FIG. 3A shows a flow chart of a user motion state detection method of an embodiment of the invention;

FIG. 3B shows a flow chart of a user motion state detection method of an embodiment of the invention (III);

fig. 4A shows a flowchart (four) of a user motion state detection method of an embodiment of the present invention;

FIG. 4B shows a flow chart of a user motion state detection method of an embodiment of the present invention (V);

FIG. 4C shows a flow chart of a user motion state detection method of an embodiment of the present invention (VI);

FIG. 5 is a schematic structural diagram (I) of a user motion state detection apparatus according to an embodiment of the present invention;

fig. 6 shows a schematic structural diagram (two) of a user movement state detection apparatus according to an embodiment of the present invention;

fig. 7 shows a schematic structural diagram (three) of a user movement state detection apparatus according to an embodiment of the present invention;

fig. 8 shows a schematic structural diagram of a wearable device according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

An embodiment of the present invention provides a user motion state detection method, as shown in fig. 1, the user motion state detection method mainly includes:

step S11: acquiring personal information and/or environmental information of a user, and determining the type of the motion data required to be acquired according to the personal information and/or the environmental information.

Optionally, in some embodiments of the present invention, the personal information of the user refers to: at least one of gender information, height information, weight information, and age information of the user. The environmental information is at least one of temperature information and season information.

In practical application, the basis for detecting the motion state of the user is real-time motion data of the user, and the motion state of the user, which can be acquired by analyzing the real-time motion data, is closely related to the personal physical characteristics of the user. For example, the gender of the user determines the difference of the user in walking, running, etc. states; similarly, different physical signs of the user may be different according to different heights, weights and ages, and therefore, corresponding motion data types can be obtained according to different personal information of the user. Alternatively, for some specific physical signs, the physical signs may be concentrated in some specific population, for example, for the signs of the circus legs, the splayfoot, etc., the physical signs are generally concentrated in the population with the smaller age, so that when the subject needing to be studied is the population with the larger age (for example, an adult), the type of the motion data for representing the signs of the circus legs, the splayfoot, etc. may not be considered. Or, for some possible accidents in sports, it may be concentrated on a particular season. For example, the fall during sports is frequently caused in winter, and the main reasons may be influenced by weather, the ground is slippery, and the probability that the user falls carelessly during sports is high. Therefore, when data collection is carried out in summer, the type of obtaining the motion data representing the falling state can be disregarded.

It should be noted that the above is only an example, and in practical applications, the type of the motion data to be acquired may be adjusted as needed. For example, the health history information (medical history information, etc.) of the user can be extracted from the personal information of the user, and the data to be collected can be determined to be acceleration data or both acceleration data and angular velocity data according to the health history information; or extracting health history information of the user from the personal information, and determining data needing to be acquired as acceleration data or both the acceleration data and the angular velocity data according to the health history information and the environmental information; or determining the data needing to be collected as acceleration data or both the acceleration data and the angular velocity data according to the environment information.

Step S12: and acquiring actual motion data of the user according to the motion data type. After the type of the motion data to be acquired is determined according to the personal information and/or the environment information of the user, the actual motion data of the user can be acquired according to the type.

Optionally, in some embodiments of the invention, the actual athletic data may be obtained through a sole of the intelligent footwear worn by the user. As shown in fig. 2, the 5 regions (MFF, LFF, MMF, LMF, HEEL) of the sole are uniformly distributed with a pressure monitoring and power generation integrated soft sensing technology, and the soft sensing technology is composed of an elastic sensing element, a displacement sensing element and a power generation module, wherein the elastic sensing element is used for enabling a measured pressure to act on a certain area and converting the pressure into a displacement or strain wheatstone bridge piezoresistive strain electric signal.

The actual movement data collected in step S12 may include pressure data of the four regions (without MMF) of the left and right soles of the user MFF, LFF, LMF, HEEL in a plurality of movement directions, a proportion of each pressure data to the total pressure, acceleration data, angular velocity data, and corresponding time data.

Optionally, in some embodiments of the invention, the actual motion data comprises acceleration data. As shown in fig. 3A, step S13, analyzing the actual motion data to determine the motion state of the user may specifically include:

step S131: acquiring acceleration change information in a plurality of motion directions according to the acceleration data;

step S132: judging whether the user is in a static state or not according to the acceleration change information;

step S133: and when the user is not in a static state, judging that the motion state of the user belongs to a circular motion state or a non-circular motion state according to the acceleration change information.

The static state and the motion state of human behavior can be well distinguished by using the acceleration data, and the cyclic motion state and the non-cyclic motion state can be well distinguished.

Optionally, the plurality of directions include two directions along a first dimension x of the length direction of the sole, two directions along a second dimension y of the width direction of the sole, and two directions perpendicular to a third dimension z of the plane of the sole.

The step S12 of collecting acceleration data may be to collect acceleration data in two directions of each of the first dimension x, the second dimension y, and the third dimension z by using a three-axis acceleration sensor on the sole according to a sampling frequency of, for example, 76Hz, 88Hz, 100Hz, 105Hz, 120Hz, or 150Hz (the frequency of walking is generally 110 steps/minute (1.8Hz), the frequency during running does not exceed 5Hz, and any one of the sampling frequencies can be selected to achieve a good balance effect in accurately reflecting acceleration changes, system efficiency, energy consumption, and the like).

It is generally determined that the user is not in a stationary state when the acceleration is not zero. In addition, the frequency of occurrence of the peak of the trajectory is counted based on the acceleration change information. In horizontal movement of a typical user, the vertical and forward accelerations will exhibit periodic variations. In the walking and foot-receiving action, the gravity center is upward, and only one foot touches the ground, the vertical acceleration tends to increase in a positive direction, then the gravity center is moved downwards, and the two feet touch the bottom, and the acceleration is opposite. The horizontal acceleration decreases when the foot is retracted and increases when the stride is taken. It can be seen that in a walking exercise, the acceleration generated by the vertical and forward motion is approximately sinusoidal with time, and has a peak at some point where the acceleration change in the vertical direction is greatest.

The above-mentioned statistics on the plurality of cyclic motions can determine that the acceleration of the cyclic motion state in a certain dimension direction changes periodically, so that the above-mentioned step S133 can distinguish the cyclic motion state from the non-cyclic motion state according to this rule.

In practical application, the cyclic motion state and the non-cyclic motion state of human body behaviors can be well distinguished by using the acceleration data. But are more difficult to distinguish for similar athletic performance. Furthermore, the acceleration data is more suitable for motion discrimination with definite direction, and for fall detection, motion cycle links, splayfoot and the like, which cannot be directly discriminated by the acceleration data, angular velocity is required for discrimination. Thus, in some embodiments of the present invention, analytical determinations may be made in conjunction with angular velocity data. In particular, angular velocity data may be collected with a gyroscope on the sole. As shown in fig. 3B, the step S13 further includes:

step S134: if the motion state of the user belongs to the circular motion state, judging that the motion state of the user belongs to going upstairs, going downstairs, walking on flat ground or running according to the acceleration change information and the angular velocity change information;

step S135: and if the motion state of the user belongs to the non-circular motion state, judging that the motion state of the user belongs to sitting, squatting, standing, jumping or falling according to the acceleration change information and the angular velocity change information.

At this time, similar motions can be finely distinguished by combining the acceleration data and the angular velocity data. In addition, different motion links in the non-cyclic motion state can be well distinguished by using an information threshold value method. And through discerning the motion of tumbleing, can fall the early warning.

The following is further explained taking fall detection as an example:

falls are characterized by large acceleration and angular velocity peaks, which are produced during falls by collisions with low-lying objects at a faster rate and therefore greater than most common procedures of walking, going upstairs, etc. during daily activities. Due to the complexity and randomness of the motion behavior process of the human body, the occurrence of the human body falling behavior can be judged by only using the acceleration information, so that great misjudgment can be brought. So embodiments of the present invention use SVM_AAnd SVM_WThe combined information threshold method can accurately distinguish falling and low-intensity motion with small SVM peak value.

In particular, the fall-corresponding acceleration signal vector modulo threshold retrievable SVM_AT＝20m/s²SVM with optional vector norm threshold for angular velocity signal_WT＝4rad/s。

After the motion state of the user is determined, whether the motion state of the user belongs to a normal state or not can be further judged. As described above, the actual exercise data may include pressure data, which can be used to distinguish between a sick state and a normal state, but is more effective in determining in conjunction with an exercise state.

For example, medical documentation and experiments demonstrate that: compared with the patients with rheumatoid arthritis and metatarsalgia, the normal people and the patients with rheumatoid arthritis have the advantages that when the normal people stand statically, the maximum pressure distribution of the front feet of the two groups of people is not obviously different, but when the people walk, the maximum pressure of the diseased feet before the phalanges leave the ground is mostly concentrated on the outer sides of the front feet, and the maximum pressure of the normal feet before the phalanges leave the ground is mostly concentrated in the middle parts of the front feet; the time of the foot sole touchdown and pressurization period of the diabetic patient is obviously longer than that of the normal person, the time of the forefoot touchdown is shorter than that of the normal person, and the touchdown process is a rapid transition process.

According to the user motion state detection method provided by the embodiment of the invention, the motion data type required to be collected can be screened according to the personal information and the environment information of the user, so that the collection amount of the motion data of the user is reduced, and the calculation amount for analyzing the motion data can be further reduced. The user motion state detection method can effectively reduce the power consumption of the wearable device and reduce the power consumption, so that the sustainable use time of the wearable device is prolonged.

Optionally, in some embodiments of the present invention, as shown in fig. 4A and 4B, after the step S13, the method for detecting a user motion state may further include: in step S14, the health status of the user is determined according to the actual exercise data of the user. Specifically, the specific process of step S14 includes:

and step S141, determining the degree of the gait of the user deviating from the normal sample crowd according to the actual motion data and a gait deviation model trained in advance.

And step S142, if the degree of the gait of the user deviating from the normal sample crowd is larger than a preset threshold value, judging that the motion state of the user belongs to a normal state or a sick state according to the degree of the gait of the user deviating from the normal sample crowd and actual motion data, and otherwise, judging that the motion state of the user belongs to the normal state.

In the embodiment of the invention, the continuous big data training subdivision calculation amount is very large, taking sole pressure data as an example, the sampling rate is 100Hz, each sole acquires 6 directional pressure values of 3000 points, the amount of each person's original data per second is 360 ten thousand, data mining is more than 360 ten thousand dimensions, and accordingly huge calculation amount is brought to the following data processing.

In order to reduce the calculation amount of data processing, in the embodiment of the present invention, the dimension reduction is first performed on the data through step S141. Step S142 only distinguishes the ill-conditioned data of the user whose gait deviates from the normal sample population by more than a preset threshold, thereby reducing the amount of calculation and improving the calculation efficiency.

In addition, in the embodiment of the invention, the original problem can be converted into the dual problem to be processed, so that the complexity is further reduced.

Optionally, the actual movement data includes N types of parameter values of N regions of the user sole, where N and N are integers greater than or equal to 1;

in the step S141, the process of determining the degree of the gait of the user deviating from the normal sample population according to the actual motion data and the gait deviation model trained in advance specifically includes:

determining a degree x of deviation of the user's gait from a normal sample population using the formula:

wherein q is_jiThe parameter value of the ith type in the jth area of the sole of the user,

i is more than or equal to 1 and less than or equal to N, and j is more than or equal to 1 and less than or equal to N, which are the average values of N types of parameters in the jth area of the soles of the normal sample crowd obtained in advance.

Specifically, the N regions may include four regions of MFF, LFF, LMF, and HEEL as shown in fig. 2, and the N-type parameter values may include acceleration, angular velocity, pressure values, and the like.

Wherein, the sample population is theoretically normal population. At the moment, normal users can be screened out by determining the degree of the gait of the users deviating from the normal population, and only users with possible pathological conditions are further distinguished.

Optionally, the specific process of step S142 includes:

determining a weight set and a morbid threshold value respectively corresponding to the motion state of the user in each morbid state according to a weight set and a morbid threshold value respectively corresponding to each motion state in each morbid state, which are determined in advance through an SVM algorithm; the weight set comprises a weight value corresponding to each type of parameter in the N types of parameters and a weight value corresponding to x;

performing weighting calculation according to the actual motion data, x and the corresponding weight set of the motion state of the user under each ill condition to obtain a weighted value under each ill condition;

and comparing the weighted value in each ill state with the corresponding ill state threshold, if at least one weighted value is greater than the ill state threshold, determining that the motion state of the user belongs to the ill state, otherwise, determining that the motion state of the user belongs to the normal state.

Here, by analyzing the expressions of different motion states in different pathological conditions, the weight values of the parameters of different motion states in different pathological conditions can be determined. For example, when a normal person stands and walks, the pressure distribution of the pressure peaks of the left sole and the right sole is basically the same; in diabetic patients and patients with borderline symptoms, the pressure of the forefoot/hindfoot is obviously increased due to the reduction of the joint mobility, and the pressure distribution is unbalanced. Thus, the weight of the pressure values corresponding to diabetes is large in the standing and walking states.

At this time, by determining the state of motion of the user, it is possible to perform early warning of various diseases such as diabetic foot, stroke, splayfoot in children, parkinson, and the like, and to realize the auxiliary rehabilitation therapy and the like.

Optionally, in some embodiments of the present invention, the weight set and the ill-conditioned threshold corresponding to the analysis of the normal condition and the ill-conditioned condition of the user in step S142 may be determined through the following steps:

acquiring actual motion data of the sole of the sample crowd in each dimension direction in a plurality of preset dimensions;

extracting gait time domain characteristics and gait frequency domain characteristics according to actual motion data of sample crowds;

performing fusion processing on the gait time domain characteristics and the gait frequency domain characteristics to obtain a gait characteristic set of the sample population after fusion;

and classifying the gait feature set of the sample population according to normal states and pathological states under different motion states by adopting an SVM algorithm, and determining a weight set and a pathological threshold value which respectively correspond to each motion state under each pathological state.

Specifically, the gait sample (gait feature set) can be classified using an SVM classifier. Assuming that M types (M is an integer greater than or equal to 1) of gait samples are registered in the database, inputting the new gait samples into an SVM classifier for training, judging which type of the M types the new gait samples belong to according to input values, if the new gait samples exceed the range of the M types, taking the new gait samples as a new type M +1, and then updating the classifier.

The gait samples of the sick crowd can be subjected to key calculation, for example, 1000 groups of pressure values of four areas of the same point and the same foot of the left foot and the right foot are respectively extracted, and then gait time domain characteristics and gait frequency domain characteristics are extracted so as to accurately determine a weight set according to the gait characteristics of the sick crowd.

Specifically, the step of performing fusion processing on the gait time domain feature and the gait frequency domain feature to obtain the fused gait feature set of the sample population includes:

acquiring a variation curve of each type of parameters in each dimension direction in a plurality of preset dimensions according to actual motion data of a sample crowd; obtaining key points of a change curve of each type of parameter by adopting a difference algorithm; extracting parameter values, driving impulse and braking impulse at key points, and obtaining gait time domain characteristics according to the parameter values, the driving impulse and the braking impulse at the key points; according to the key points, performing waveform alignment on the change curve of each type of parameter by adopting a linear interpolation method; and extracting gait frequency domain characteristics from the change curve after waveform alignment by adopting a wavelet packet decomposition algorithm.

In practical application, acting force of each area of the sole is related to movement gait, time frequency can represent overall characteristics such as gait periodicity, change rate and acceleration, and frequency domain can represent detailed characteristics such as spectral characteristics. The wavelet packet decomposition and difference algorithm can be adopted to respectively extract frequency domain and time domain characteristics from the pressure data of three dimensions of four areas of the sole, so that the SVM algorithm is utilized to identify the motion state, the normal state and the ill state.

The extraction process of the time domain features mainly comprises the following steps: the peak point and the valley point of curves in the front-back direction (x axis) and the vertical direction (z axis) can be detected by adopting a first-order difference algorithm to serve as key points of the acting force curve, and the valley point of the curve in the vertical direction is taken as a reference point of the acting force curve; and then representing the gait time domain characteristics of the whole course by using the pressure value of the key point of the vertical direction curve, the time phase of the occurrence of the pressure value, the acting force change rate and impulse (including driving impulse and braking impulse) of the adjacent key point, the pressure value of the corresponding key point on the front and back direction curve, the driving impulse (the integral of the force above the 0 point on the force-time curve and the time) and the braking impulse (the integral of the force below the 0 point on the force-time curve and the time).

The extraction process of the frequency domain features mainly comprises the following steps: the acting force can be aligned with the waveform of the acting force according to key points on a curve in the vertical direction, so that the frequency domain feature contrast and the classification capability are improved. Specifically, the dimension of the acting force is normalized to the same value by a linear interpolation algorithm, valley points on a force curve in the vertical direction of the normalized acting force are searched by a first-order difference algorithm, the valley points are used as key points for reference, and then the waveforms of the curves in the left-right direction (y axis), the front-back direction and the vertical direction in the acting force are aligned by the linear interpolation method to obtain the aligned acting force. And extracting the gait frequency domain characteristics of the whole course from the acting force by using an L-layer wavelet packet decomposition algorithm.

Optionally, in the above process of classifying the gait feature set of the sample population according to normal states and pathological states in different motion states by using the SVM algorithm, and determining the weight set and the pathological threshold value respectively corresponding to each motion state in each pathological state, the minimum optimal wavelet packet set may be selected from a plurality of wavelet packets of the extracted gait frequency domain features by using a fuzzy C mean method, and then the minimum optimal wavelet packet decomposition coefficient may be selected from the selected set by using the fuzzy C mean method based on fuzzy membership ranking to obtain the minimum optimal gait frequency domain feature subset, which is then combined with the gait time domain features to obtain the fused gait feature set.

Optionally, in some embodiments of the present invention, as shown in fig. 4C, after the analyzing of the normal state and the ill state of the user is performed in step S142, an optimizing step of the weight set and the ill state threshold may further be included, specifically, the optimizing step includes:

and step S143, verifying the judgment result of judging whether the motion state of the user belongs to a normal state or a sick state.

Specifically, the degree of the gait of the user deviating from the normal sample crowd is determined again according to the actual motion data and the updated gait deviation model; the updated gait deviation model is obtained according to the data of the re-screened sample population in the normal state;

if the redetermined degree of the gait of the user deviating from the normal sample crowd is larger than the preset threshold value, the motion state of the user is redetermined to be normal or ill according to the redetermined degree of the gait of the user deviating from the normal sample crowd and the actual motion data;

and verifying the last judgment result according to the re-judged judgment result.

After the verification is performed in step S143, step S144 is performed to optimize, according to the verification result, the weight set and the ill-conditioned threshold value respectively corresponding to each motion state determined in advance by the SVM algorithm in each ill condition. With the increase of the sample size, the SVM classifier can be self-adaptively optimized and perfected continuously, and the execution efficiency of the algorithm is improved.

The SVM classifier can perform sampling calculation aiming at sample crowds without abnormality. When searching for an abnormal sample, the random sampling verification is performed again because the abnormal sample is not found due to the fact that the standard deviation of one area is larger, the standard deviation of the other area is smaller and exactly balanced, and the like. And (4) inputting a new sample every time, and calculating the recognition rate of the SVM classifier according to the cross verification method principle.

And for the characteristic values of the samples which are not found to be abnormal, using an SVM classifier fitness function to divide the accuracy of the samples for the SVM classifier. The parallel execution process is simulated by maintaining a plurality of groups and appropriately controlling the interaction between the groups, thereby improving the execution efficiency of the algorithm even without using a parallel computer.

Further, to ensure accuracy of the collected actual motion data, the method for detecting a motion state of a user according to the embodiment of the present invention may further include: and denoising the actual motion data by adopting a wavelet transform threshold method.

In practical application, electromagnetic interference in a circuit is a main interference source in the acquisition process, and the electromagnetic interference is high-frequency noise; the human motion is mainly low-frequency signals within 50Hz, and the discrete wavelet transform threshold value method is adopted in the embodiment of the invention, so that the method has the advantages of band-pass filtering function and high calculation speed. Specifically, a threshold value and step frequency judgment can be added for detection to filter, namely, the time interval of two adjacent steps is at least more than 0.11, 0.14, 0.17, 0.2, 0.23 and 0.27 seconds, high-frequency noise is filtered, and the best balance effect can be achieved in accurately reflecting acceleration change, system efficiency, energy consumption and the like.

In addition, the wavelet transformation operation of three steps of wavelet decomposition, high-frequency wavelet coefficient processing and wavelet reconstruction can be carried out on the pressure data collected into the four regions, the pressure time domain signals of the four regions are discretized, the mixed signals of various frequency components are decomposed into different frequency bands, and then the mixed signals are processed according to different characteristics of various seed signals on the frequency domain according to frequency bands; then, a matrix-based unsupervised algorithm is used to remove noise and preserve the most representative information. And finally, further improving the resolution capability by using a supervision algorithm. And acquiring gait data with high signal to noise ratio.

In the embodiment of the invention, the data can be filtered by referring to the personal information and/or the environmental information of the user when determining the type of the motion data to be collected. When analyzing the actual movement data, the personal information and/or the environmental information of the user may be referred to as well. As shown in fig. 4C, the process of determining the health status based on the personal information and/or the environmental novelty mainly includes: step S145: determining a health state range to be detected according to personal information and/or environmental information; step S146: determining the degree of the gait of the user deviating from the normal sample crowd according to the health state range, the acceleration data and a gait deviation model trained in advance; if the degree of the gait of the user deviating from the normal sample crowd is larger than the preset threshold value, judging that the motion state of the user belongs to a normal state or a sick state according to the degree of the gait of the user deviating from the normal sample crowd and the actual motion data, otherwise, judging that the motion state of the user belongs to the normal state.

In some embodiments of the present invention, for the judgment of the health state, a neural network deep learning algorithm is required to be used in combination with a training model to perform posture judgment, and further perform the health state judgment. However, when the neural network deep learning algorithm is used, the continuous large data training subdivision calculation amount is very large, and accordingly, huge calculation amount is brought to the subsequent data processing. And aiming at the judgment of some health states, a possible health state range can be determined according to personal information and/or environmental information of a user, and then whether acceleration data of the user can be only collected to analyze the health state of the user is determined based on the personal information and/or the environmental information and the health state range, so that the health state of the user is obtained by combining a gait deviation model trained in advance, the angular velocity data does not need to be collected for further subdivision, the calculation amount of motion data processing can be greatly reduced, the electric quantity consumption of corresponding wearable equipment is effectively reduced, the continuous service time of the wearable equipment is prolonged, and the cruising ability is improved.

The following description is given with reference to specific application examples.

In practical applications, many sports injuries or diseases may be concentrated in a particular environment or climate. For example, in rheumatoid arthritis, the onset symptoms are greatly affected by climate change, and joint pain and other symptoms often appear before the weather turns cold or rains. Therefore, when the user's personal information is informed that he has a history of rheumatoid arthritis, environmental factors can be taken into account when performing analysis based on the user's actual exercise data. When the current environment is judged to be cold (for example, lower than 10 ℃), or the current season belongs to the autumn and winter season, or the current climate belongs to the cloudy days, the rainy days and other conditions, the acceleration change information of the user is obtained according to the acceleration data of the user movement, so that the user is judged to belong to the circular movement state, and the user can be judged to be in the walking, running or abnormal walking state according to the acceleration change information. The abnormal walking state includes sole mopping walking (i.e. the sole has large friction with the ground and does not belong to a normal walking posture). At this time, when the user is judged to be in an abnormal walking state by combining personal information (history of rheumatoid arthritis) of the user and current environmental information (weather becomes cold or before raining, etc.) and combining a gait deviation model trained in advance, it is judged that the user is in a state of rheumatoid arthritis attack.

In some embodiments, the abnormal walking state of the user can be combined with only personal information of the user, or the abnormal walking state of the user can be combined with only environmental information, and the judgment can be made as well.

Therefore, the personal information and/or the environmental information of the user are combined with the motion state obtained by analyzing according to the acceleration data, and the health state of the user can be judged by combining the acceleration data with the neural network algorithm (for example, in the processes of the step S141 and the step S142, the actual motion data is only the acceleration data), without further refining the training model of the neural network algorithm by combining the angular velocity data, so that the computation amount of analysis based on the motion data can be greatly reduced, the power consumption of the wearable device is reduced, and the endurance time of the wearable device is prolonged.

If the possible health status detection range of the user cannot be determined according to the personal information and/or environmental information of the user (for example, the physical health of the user does not provide reference for past medical history), a neural network algorithm subdivided by combining angular velocity data is further adopted to perform more detailed motion status analysis and health status analysis (for example, in the processes of the above-mentioned steps S141 and S142, the actual motion data is acceleration data and angular velocity data).

The method of the embodiment of the invention is used for analyzing the motion state and the health state of the user, and comprises the steps of collecting primary motion data of the user according to the motion data type, and establishing a personal portrait model by combining personal information and/or environmental information; and determining whether to acquire secondary motion data according to the model to establish a secondary artificial intelligence model. The motion data types required to be collected and analyzed can be finely screened according to the personal information of the user and the environment information, and the collected amount and the calculated amount are reduced through a multi-stage model contained in the model, so that the consumption of electric quantity is effectively reduced.

Optionally, in some embodiments of the present invention, the user motion state detection device may be disposed in a wearable intelligent footwear worn by the user, for example, in a sole of the footwear. As shown in fig. 5, the user movement state detection apparatus mainly includes: an information acquisition module 51, an actual motion data acquisition module 52, a motion state determination module 53, and the like.

The information obtaining module 51 is configured to obtain personal information and/or environmental information of a user, and determine a type of motion data to be collected according to the personal information and/or the environmental information; for details, reference may be made to the description relating to step S11 of the above method embodiment.

The actual movement data acquisition module 52 is used for acquiring actual movement data of the user according to the movement data type; for details, reference may be made to the description relating to step S12 of the above method embodiment.

A motion state determination module 53, configured to analyze actual motion data and determine a motion state of the user; for details, reference may be made to the description relating to step S13 of the above method embodiment.

Optionally, in some embodiments of the invention, the actual motion data comprises acceleration data. As shown in fig. 6, the motion state determination module 53 includes:

the acceleration change information obtaining submodule 531 is configured to obtain acceleration change information in a plurality of motion directions according to the acceleration data; for details, reference may be made to the description relating to step S131 of the above-described method embodiment.

The cyclic motion judgment submodule 532 is used for judging whether the user is in a static state or not according to the acceleration change information; and when the user is not in a static state, judging that the motion state of the user belongs to a circular motion state or a non-circular motion state according to the acceleration change information. For details, reference may be made to the related descriptions of step S132 and step S133 of the above method embodiments.

In practical application, the cyclic motion state and the non-cyclic motion state of human body behaviors can be well distinguished by using the acceleration data. But are more difficult to distinguish for similar athletic performance. Furthermore, the acceleration data is more suitable for motion discrimination with definite direction, and for fall detection, motion cycle links, splayfoot and the like, which cannot be directly discriminated by the acceleration data, angular velocity is required for discrimination. Thus, in some embodiments of the present invention, analytical determinations may be made in conjunction with angular velocity data. In particular, angular velocity data may be collected with a gyroscope on the sole. As shown in fig. 6, the motion state determination module further includes: the motion state determination submodule 533 is configured to perform the following steps: if the motion state of the user belongs to the circular motion state, the motion state judgment sub-module 533 judges that the motion state of the user belongs to ascending stairs, descending stairs, walking on flat ground or running according to the acceleration change information and the angular velocity change information; if the motion state of the user belongs to the non-cyclic motion state, the motion state determination sub-module 533 determines, according to the acceleration change information and the angular velocity change information, that the motion state of the user belongs to sitting, squatting, standing, jumping, or falling. For details, reference may be made to the related descriptions of step S134 and step S135 of the above method embodiments.

The user motion state detection device provided by the embodiment of the invention can screen the motion data type required to be acquired according to the personal information and the environment information of the user, thereby reducing the acquisition amount of the motion data of the user and further reducing the calculation amount for analyzing the motion data. This user motion state detection device can effectively reduce wearable equipment's consumption, reduces the consumption of electric quantity to the sustainable use time of this wearable equipment is prolonged.

Optionally, in some embodiments of the present invention, as shown in fig. 7, the user motion state detection apparatus may further include a health state determination module 54 for determining a health state of the user according to the actual motion data of the user, so as to provide early warning information for the health state of the user. For details, reference may be made to the description relating to step S14 of the above method embodiment.

Specifically, the health status determination module 54 includes: the gait deviation degree determining submodule is used for determining the degree of deviation of the gait of the user from the normal sample crowd according to the actual motion data and a pre-trained gait deviation model; for details, reference may be made to the description relating to step S141 of the above-described method embodiment.

The health state determining sub-module is used for judging whether the motion state of the user belongs to a normal state or a sick state according to the degree of the gait of the user deviating from the normal sample crowd and actual motion data if the degree of the gait of the user deviating from the normal sample crowd is larger than a preset threshold value, and otherwise, judging whether the motion state of the user belongs to the normal state; for details, reference may be made to the description related to step S142 of the above method embodiment.

Optionally, in some embodiments of the present invention, after the analyzing the normal state and the ill state of the user by the health state determining module 54, the user motion state detecting apparatus may further include:

a verification module 55, configured to verify a determination result that the motion state of the user belongs to a normal state or a pathological state; for details, reference may be made to the description related to step S143 of the above method embodiment.

An optimization module 56, configured to optimize, according to the verification result, a weight set and a pathological threshold value respectively corresponding to each motion state determined in advance through an SVM algorithm in each pathological state; for details, reference may be made to the description related to step S144 of the above method embodiment.

The reason why the health status determining module 54, the verifying module 55 and the optimizing module 56 shown in fig. 7 are represented by dashed boxes is that, in practical applications, also in consideration of reducing power consumption and battery consumption, the health status determining module 54, the verifying module 55 and the optimizing module 56 may be computing platforms arranged in a cloud, and the obtained actual exercise data of the user is transmitted to the computing platforms in the cloud through a communication module arranged in the intelligent footwear worn by the user to perform analysis and calculation of the health status of the user, so that the consumption of the battery by the intelligent footwear in the use process is further reduced, and the endurance time of the intelligent footwear is prolonged.

In this embodiment, the motion status determination submodule 533 is configured to perform the following steps: if the motion state of the user belongs to the circular motion state, the motion state judgment sub-module 533 judges that the motion state of the user belongs to the walking, running or abnormal walking state according to the acceleration change information.

Correspondingly, the health status determination module 54 further includes: the health state range determining submodule is used for determining a health state range to be detected according to the personal information and/or the environmental information; at this time, the health status determination submodule of the health status determination module 54 is configured to determine, according to the health status range, the acceleration data, and a gait deviation model trained in advance, a degree of deviation of the gait of the user from the normal sample population; if the degree of the gait of the user deviating from the normal sample crowd is larger than the preset threshold value, judging that the motion state of the user belongs to a normal state or a sick state according to the degree of the gait of the user deviating from the normal sample crowd and the actual motion data, otherwise, judging that the motion state of the user belongs to the normal state.

Optionally, in the embodiment of the present invention, the data may be filtered with reference to personal information and/or environmental information of the user not only when determining the type of the exercise data to be collected. When analyzing the actual movement data, the personal information and/or the environmental information of the user may be referred to as well. And, carry out motion state analysis based on this personal information and/or environmental information to and the health condition who goes on the user analyzes, can significantly reduce the calculated amount to motion data processing to effectively reduce the electric quantity consumption of corresponding wearable equipment, prolong its continuous use time, improve duration.

If the possible health status detection range of the user cannot be determined according to the personal information and/or environmental information of the user (for example, the physical health of the user, no reference is provided for past medical history), the health status determining module 54 further needs to use a neural network algorithm subdivided by combining the angular velocity data and the angular velocity data to perform more detailed motion status analysis and health status analysis (for example, in the processes described in the above steps S141 and S142, the actual motion data is the acceleration data and the angular velocity data).

The device of the embodiment of the invention is used for analyzing the motion state and the health state of the user, and is used for collecting the primary motion data of the user according to the motion data type and establishing a personal portrait model by combining personal information and/or environmental information; and determining whether to acquire secondary motion data according to the model to establish a secondary artificial intelligence model. The motion data types required to be collected and analyzed can be finely screened according to the personal information of the user and the environment information, and the collected amount and the calculated amount are reduced through a multi-stage model contained in the model, so that the consumption of electric quantity is effectively reduced.

An embodiment of the present invention further provides a wearable device, as shown in fig. 8, the wearable device may include a processor 81 and a memory 82, where the processor 81 and the memory 82 may be connected by a bus or in another manner, and fig. 8 takes the example of connection by a bus as an example. In a preferred embodiment, the wearable device may be an intelligent footwear, but the invention is not limited thereto.

Processor 81 may be a Central Processing Unit (CPU). The Processor 81 may also be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, or combinations thereof.

The memory 82, which is a non-transitory computer-readable storage medium, may be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as program instructions/modules corresponding to the key shielding method of the vehicle-mounted display device in the embodiment of the present invention (for example, the information acquiring module 51, the actual motion data acquiring module 52, and the motion state determining module 53 shown in fig. 5). The processor 81 executes various functional applications and data processing of the processor by running non-transitory software programs, instructions and modules stored in the memory 82, namely, implements the user motion state detection method in the above method embodiment.

The memory 82 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created by the processor 81, and the like. Further, the memory 82 may include high speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory 82 may optionally include memory located remotely from the processor 81, which may be connected to the processor 81 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The one or more modules are stored in the memory 82 and, when executed by the processor 81, perform a user motion state detection method as in the embodiment shown in fig. 1-4C.

The specific details of the wearable device may be understood by referring to the corresponding related descriptions and effects in the embodiments shown in fig. 1 to 4C, which are not described herein again.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a Flash Memory (Flash Memory), a Hard Disk (Hard Disk Drive, abbreviated as HDD), a Solid State Drive (SSD), or the like; the storage medium may also comprise a combination of memories of the kind described above.

Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations fall within the scope defined by the appended claims.

Claims

1. A method for detecting a motion state of a user, comprising:

acquiring personal information and environmental information of a user, and determining the type of motion data to be acquired according to the personal information and the environmental information;

acquiring actual motion data of a user according to the motion data type;

analyzing the actual motion data and determining the motion state of the user;

determining the type of the motion data required to be collected according to the personal information and the environmental information, wherein the type of the motion data comprises the following steps:

extracting health history information of the user from the personal information, and determining data to be acquired as acceleration data and angular velocity data according to the health history information and the environmental information;

further comprising:

determining the health status of the user from the actual movement data of the user,

the actual motion data comprises N types of parameter values of N areas of the sole of the user, N and N are integers which are larger than or equal to 1, and the health state range to be detected is determined according to the personal information and the environmental information;

determining the degree x of the gait of the user deviating from the normal sample crowd according to the health state range, the acceleration data and a gait deviation model trained in advance:

i is more than or equal to 1 and less than or equal to N, j is more than or equal to 1 and less than or equal to N, which are the average values of N types of parameters in the jth area of the soles of the normal sample crowd obtained in advance;

if the degree of the gait of the user deviating from the normal sample crowd is larger than a preset threshold value, judging that the motion state of the user belongs to a sick state according to the degree of the gait of the user deviating from the normal sample crowd and the actual motion data, and otherwise, judging that the motion state of the user belongs to a normal state.

2. The user motion state detection method according to claim 1, wherein the personal information includes: at least one of sex information, height information, weight information and age information of the user; the environment information includes: at least one of temperature information and season information.

3. The user motion state detection method according to claim 2, wherein the actual motion data includes acceleration data;

the analyzing the actual motion data to determine the motion state of the user includes:

acquiring acceleration change information in a plurality of motion directions according to the acceleration data;

judging whether the user is in a static state or not according to the acceleration change information;

and when the user is not in a static state, judging that the motion state of the user belongs to a circular motion state or a non-circular motion state according to the acceleration change information.

4. The user motion state detection method according to claim 3, wherein if the motion state of the user belongs to a cyclic motion state, it is determined that the motion state of the user belongs to a walking, running or abnormal walking state according to the acceleration change information.

5. The user motion state detection method according to claim 3, wherein the actual motion data further includes angular velocity data;

the analyzing the actual motion data to determine the motion state of the user further comprises:

if the motion state of the user belongs to the circular motion state, judging that the motion state of the user belongs to going upstairs, going downstairs, walking on flat ground or running according to the acceleration change information and the angular velocity change information;

and if the motion state of the user belongs to the non-circular motion state, judging that the motion state of the user belongs to sitting, squatting, standing, jumping or falling according to the acceleration change information and the angular velocity change information.

6. The method according to claim 5, wherein when the health status range to be detected cannot be determined based on the personal information and the environmental information, the determining the health status of the user based on the actual exercise data of the user further comprises:

determining the degree of the gait of the user deviating from the normal sample crowd according to the acceleration data, the angular velocity data and a pre-trained gait deviation model;

7. The user motion state detection method according to claim 5 or 6, further comprising:

verifying a judgment result for judging whether the motion state of the user belongs to a normal state or a sick state;

and optimizing the weight set and the ill-conditioned threshold value respectively corresponding to each motion state determined by the SVM algorithm in advance under each ill condition according to the verification result.

8. The method according to claim 7, wherein the verifying the determination result for determining whether the motion state of the user belongs to a normal state or a pathological state includes:

re-determining the degree of the gait of the user deviating from the normal sample crowd according to the actual motion data and the updated gait deviation model; the updated gait deviation model is obtained according to the data of the re-screened sample population in the normal state;

if the redetermined degree of the user gait deviating from the normal sample crowd is larger than the preset threshold, the motion state of the user is redetermined to be normal state or sick state according to the redetermined degree of the user gait deviating from the normal sample crowd and the actual motion data;

9. A user motion state detection apparatus, comprising:

the information acquisition module is used for acquiring personal information and environmental information of a user and determining the type of the motion data required to be acquired according to the personal information and the environmental information;

the actual motion data acquisition module is used for acquiring actual motion data of the user according to the motion data type;

the motion state determining module is used for analyzing the actual motion data and determining the motion state of the user;

the information acquisition module is specifically configured to:

extracting health history information of the user from the personal information, and determining data to be acquired as acceleration data and angular velocity data according to the health history information and the environmental information; further comprising:

the health state determining module is used for determining the health state of the user according to the actual motion data of the user; the actual movement data comprises N types of parameter values of N areas of the sole of the user, wherein N and N are integers which are more than or equal to 1,

the health status determination module comprises:

the health state range determining submodule is used for determining a health state range to be detected according to the personal information and the environmental information;

the health state judgment sub-module is used for determining the degree x of the gait of the user deviating from the normal sample crowd according to the health state range, the acceleration data and a pre-trained gait deviating model;

10. The user motion state detection apparatus according to claim 9, wherein the personal information includes: at least one of sex information, height information, weight information and age information of the user; the environment information includes: at least one of temperature information and season information.

11. The user motion state detection device according to claim 10, wherein the actual motion data includes acceleration data;

the motion state determination module comprises:

the acceleration change information acquisition submodule is used for acquiring acceleration change information in a plurality of motion directions according to the acceleration data;

the cyclic motion judgment submodule is used for judging whether the user is in a static state or not according to the acceleration change information; and when the user is not in a static state, judging that the motion state of the user belongs to a circular motion state or a non-circular motion state according to the acceleration change information.

12. The user motion state detection apparatus of claim 11, wherein the motion state determination module further comprises:

and the motion state judgment sub-module judges that the motion state of the user belongs to a walking, running or abnormal walking state according to the acceleration change information if the motion state of the user belongs to a circular motion state.

13. The user motion state detection device according to claim 11, wherein the actual motion data further includes angular velocity data;

the motion state determination module further comprises:

the motion state judgment sub-module judges that the motion state of the user belongs to ascending stairs, descending stairs, walking on flat ground or running according to the acceleration change information and the angular velocity change information if the motion state of the user belongs to a circular motion state;

and if the motion state of the user belongs to the non-circular motion state, the motion state judgment submodule judges that the motion state of the user belongs to sitting, squatting, standing, jumping or falling according to the acceleration change information and the angular speed change information.

14. The user motion state detection apparatus of claim 13, wherein the health state determination module further comprises:

the gait deviation degree determining submodule is used for determining the degree of deviation of the gait of the user from a normal sample crowd according to the acceleration data, the angular velocity data and a pre-trained gait deviation model;

and the health state determining sub-module is used for judging that the motion state of the user belongs to a sick state according to the degree of the gait of the user deviating from the normal sample crowd and the actual motion data if the degree of the gait of the user deviating from the normal sample crowd is larger than a preset threshold value, and otherwise, judging that the motion state of the user belongs to a normal state.

15. The user motion state detection device according to claim 13 or 14, further comprising:

the verification module is used for verifying a judgment result for judging whether the motion state of the user belongs to a normal state or a sick state;

and the optimization module is used for optimizing the weight set and the ill-condition threshold value which correspond to each motion state determined by the SVM algorithm in advance under each ill condition according to the verification result.

16. The apparatus according to claim 15, wherein the verification module is specifically configured to:

17. A wearable device, comprising:

a memory and a processor, the memory and the processor being communicatively coupled to each other, the memory having stored therein computer instructions, the processor executing the computer instructions to perform the user motion state detection method of claim 1.

18. A computer-readable storage medium storing computer instructions for causing a computer to perform the user motion state detection method according to claim 1.