Body posture recognition method and device and intelligent mattress

By regularly obtaining the airbag air pressure value on the smart mattress and calculating the variance sum, the problem that the existing smart mattress cannot recognize the sleeping posture is solved, and the accurate judgment of the user's posture is achieved, and the accuracy of sleep quality evaluation is improved.

CN118924069APending Publication Date: 2024-11-12DEEP COGNITION (SHENZHEN) TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310540525.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-05-12
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

Existing smart mattresses cannot accurately identify the specific sleeping posture of users during sleep, which affects the evaluation of sleep quality.

Method used

By setting the airbag in the smart mattress, obtaining the air pressure value regularly, calculating the variance sum of the airbags, and comparing the variance sum of the adjacent time intervals to judge the user's body posture.

Benefits of technology

It realizes more accurate recognition of the user's body posture, can recognize multiple posture changes, and improves the accuracy of sleep quality assessment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118924069A_ABST
    Figure CN118924069A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of intelligent mattresses, in particular to a body posture recognition method and device and an intelligent mattress. The body posture recognition method comprises the following steps: acquiring an air pressure value in each air bag in a mattress at a preset time interval; calculating a variance corresponding to each air bag in adjacent preset time intervals according to the air pressure value; calculating the sum of variances of all the air bags in adjacent preset time intervals according to the variances; and judging the body posture of the current user according to calculation and comparison of the variance sum of the adjacent preset time intervals. According to the method, the air pressure value of each air bag in the mattress is obtained every other preset time, the variance corresponding to each air bag is calculated, the variances of the air bags are added, the variances of two adjacent time periods are compared, and then the difference between the variances and the expectation is judged. According to the pressure change of the plurality of air bags, the pressure variance and the pressure difference value are specifically utilized, so that the body posture of the current user can be more accurately judged.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of smart mattresses, and in particular to a body posture recognition method, device and a smart mattress. Background Art

[0002] In recent years, the global healthy sleep market has expanded rapidly, and people are paying more and more attention to healthy sleep. Smart mattresses that can monitor people's sleep quality can better help people solve healthy sleep problems, and the demand in the market has shown a significant increase.

[0003] In the prior art, smart mattresses can only identify the user's getting in and out of bed through the pressure changes of the pressure sensor in the airbag, and cannot accurately detect and identify the user's specific sleeping posture during sleep. Since the change of the user's body posture is usually related to their sleep quality, it is a reference indicator of sleep quality. The better the sleep quality, the lower the number of body movements at night. On the contrary, people with poor sleep quality will turn over more frequently at night than those with good sleep quality. Summary of the invention

[0004] The problem solved by the present invention is that the existing intelligent mattress cannot perform corresponding recognition and detection of the specific sleeping posture of the user.

[0005] In order to solve the above problems, the present invention provides a body posture recognition method, device and smart mattress.

[0006] In a first aspect, the present invention provides a body posture recognition method, comprising:

[0007] Obtain the air pressure value in each air bag in the mattress at each preset time interval;

[0008] Calculate the variance corresponding to each of the airbags within adjacent preset time intervals according to the air pressure value;

[0009] Calculate the sum of variances of all the airbags in adjacent preset time intervals according to the variances;

[0010] The body posture of the current user is determined by calculating and comparing the sum of the variances of adjacent preset time intervals.

[0011] Optionally, judging the body posture of the current user according to calculating and comparing the sum of the variances of two adjacent preset time intervals comprises:

[0012] If the difference between the sum of the variances of the first time interval and the sum of the variances of the second time interval is greater than a first preset value, it is determined that the user's body posture is in bed.

[0013] Optionally, after determining that the user's body posture is in a bed-entering state, the method includes:

[0014] If the absolute value of the difference between the sum of the variances of the second time interval and the sum of the variances of the first time interval is greater than a fourth preset value;

[0015] If the absolute value of the difference between the sum of the variances of the second time interval and the sum of the variances of the first time interval is smaller than the first preset value, it is determined that the user's body posture has changed.

[0016] Optionally, after determining that the user's body posture has changed, the method further includes:

[0017] When the absolute value of the difference between the variance of any one of the airbags in the second time interval and the variance of the airbag in the first time interval is greater than a second preset value, it is determined that the pressure of the airbag has changed to determine the user body area corresponding to the airbag.

[0018] Optionally, after determining the user body area corresponding to the airbag, the method further includes:

[0019] If it is detected that the difference between the air pressure value of the first airbag located on the shoulder / back of the user in the second time interval and the air pressure value of the first airbag in the first time interval is greater than the third preset value, it is determined that the user's body posture has changed from lying flat to lying sideways.

[0020] Optionally, after determining that the user's body posture has changed, the method further includes:

[0021] If the difference between the air pressure value of the first airbag located on the shoulder / back of the user in the first time interval and the air pressure value of the first airbag in the second time interval is greater than the third preset value, it is determined that the user's body posture has changed from lying on its side to lying flat.

[0022] Optionally, after determining that the user's body posture is in a bed-entering state, the method further includes:

[0023] If the absolute value of the difference between the sum of the variance of the second time interval and the sum of the variance of the first time interval is less than a fourth preset value, it is determined that the user's body posture has not changed at this time.

[0024] Optionally, the determining the body posture of the current user based on calculating and comparing the sum of the variances of two adjacent preset time intervals further includes:

[0025] If the difference between the sum of the variances of the second time interval and the sum of the variances of the first time interval is greater than a first preset value, it is determined that the user's body posture is out of bed at this time.

[0026] Optionally, the step of obtaining the air pressure value in each air bag in the mattress at each preset time interval includes:

[0027] If the difference between the pressure value of any one of the airbags in the second time interval and the pressure value of the airbag in the first time interval is less than a fifth preset value, it is determined that the mattress is in an empty state at this time.

[0028] In a second aspect, the present invention provides a detection device, comprising:

[0029] An acquisition module, used for acquiring the air pressure value in each air bag in the mattress at each preset time interval;

[0030] A calculation module, used to calculate the variance corresponding to each of the airbags in adjacent preset time intervals according to the air pressure value, and used to calculate the sum of the variances of all the airbags in adjacent preset time intervals according to the variance;

[0031] The judging module is used to judge the body posture of the current user by calculating and comparing the sum of the variances of adjacent preset time intervals.

[0032] In a third aspect, the present invention provides a smart mattress, comprising a memory and a processor;

[0033] The pressure sensor is used to obtain the air pressure value of each airbag in the airbag mattress and send the air pressure value of each airbag to the processor;

[0034] The processor is used to calculate the variance corresponding to each airbag according to the air pressure value of each airbag within a preset time interval, to calculate the sum of the variances of all the airbags according to the variance corresponding to each airbag, and to judge the body posture of the current user by calculating and comparing the sum of the variances of two adjacent preset time intervals.

[0035] The beneficial effects of the body posture recognition method, device and storage medium of the present invention are:

[0036] By obtaining the air pressure value of each airbag in the mattress at a preset interval, and calculating the variance corresponding to each airbag through the air pressure value of each airbag, and adding up the variance of each airbag, and comparing the sum of the variances of two adjacent time periods, the difference between the sum of the variances and the expectation is determined. In addition, more types of body postures can be identified, and according to the pressure changes of multiple airbags, and specifically using the pressure variance and pressure difference, the current user's body posture can be determined more accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A schematic diagram of a flow chart of a body posture recognition method according to an embodiment of the present invention;

[0038] Figure 2 A schematic diagram of the detailed steps after step S400 of the body posture recognition method according to an embodiment of the present invention;

[0039] Figure 3 It is a schematic diagram of the detailed steps after step S410 of the body posture recognition method according to an embodiment of the present invention;

[0040] Figure 4 A schematic diagram of the detailed steps after step S411 of the body posture recognition method according to an embodiment of the present invention;

[0041] Figure 5 It is a schematic diagram of the detailed steps after step S100 of the body posture recognition method according to an embodiment of the present invention;

[0042] Figure 6 The figure is a schematic diagram of the structure of a detection device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only an embodiment of a part of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work should belong to the scope of protection of the present invention. It should be understood that the various steps recorded in the method implementation of the present invention can be performed in different orders and / or in parallel. In addition, the method implementation may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0044] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0045] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0046] It should be noted that the modifications of "one" and "plurality" mentioned in the present invention are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0047] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only used for illustrative purposes, and are not used to limit the scope of these messages or information.

[0048] like Figure 1 As shown, an embodiment of the present invention provides a body posture recognition method, the body posture recognition method comprising:

[0049] Step S100, obtaining the air pressure value in each air bag in the mattress at each preset time interval;

[0050] Specifically, the preset time can be set as needed, such as 1s, 2s, 3s, 4s, etc. Preferably, the preset time is 3s, and each 3s is divided into a frame, and an air pressure detection is performed once in each frame. The number of airbags can be several, and the corresponding air pressure detection is performed on each airbag in the mattress in each frame, and the pressure values ​​of all airbags in the mattress in each frame are recorded.

[0051] The pressure of each airbag can be recorded by the formula: Pti (i=1~n), i represents the corresponding airbag, for example, the first airbag is Pt1, the second airbag is Pt2, t represents the preset time, for example, in this embodiment, preferably, the preset time is 3s, then the first airbag is recorded as P31; the second airbag is recorded as P32; the third airbag is recorded as P33, and so on.

[0052] like Figure 1 As shown, step S200, calculating the variance corresponding to each of the airbags within the adjacent preset time intervals according to the air pressure value;

[0053] Specifically, the variance of each airbag can be recorded by the formula: Vti = Var (Pti), where i in the formula represents the number of airbags, for example, i for the first airbag is 1; i for the second airbag is 2. The value range of i can be 1 to n, and the final value of n is determined according to the total number of airbags. According to different customized requirements, the number of airbags can be 1, 2, 3, 4, 5, 6, etc. t represents the preset time. In the preferred embodiment of this embodiment, the preset time is 3 seconds, but in other embodiments, the value of t can also be 1 second, 2 seconds, 3 seconds, 4 seconds, 5 seconds, etc.

[0054] Therefore, through the above formula, for example, the variance of the first airbag can be recorded as V31=Var(P31); the variance of the second airbag can be recorded as V32=Var(P32); the variance of the third airbag can be recorded as V33=Var(P33), and so on.

[0055] like Figure 1 As shown, step S300, calculating the sum of variances of all the airbags within adjacent preset time intervals according to the variance;

[0056] Specifically, the sum of the variances of all the airbags within a preset time interval can be recorded as The final value of n is determined according to the total number of airbags. According to different customized requirements, the number of airbags can be 1, 2, 3, 4, 5, 6, etc. The value of t can be preferably 3 seconds in this embodiment.

[0057] Therefore, by the above formula, for example, when there are three airbags, the value of ST can be calculated as V31+V32+V33; and when there are four airbags, the value of ST can be calculated as V31+V32+V33+V34. The value of ST can be calculated by the above formula and recorded.

[0058] like Figure 1 As shown, in step S400, the body posture of the current user is determined by calculating and comparing the sum of the variances of adjacent preset time intervals.

[0059] Specifically, according to the recorded ST values, by comparing the changes in the ST values ​​in adjacent preset time intervals, the body posture can be judged. The value of T can be an integer, the first time interval can be recorded as S1, the second time interval can be recorded as S2, the third time interval can be recorded as S3, and so on. The body posture of the user can be judged by comparing the sum of the variances in two adjacent time intervals, for example, comparing S1 with S2; S2 with S3; S3 with S4, and so on.

[0060] By obtaining the air pressure value of each airbag in the mattress at a preset interval, and calculating the variance corresponding to each airbag through the air pressure value of each airbag, and adding up the variance of each airbag, and comparing the sum of the variances of two adjacent time periods, the difference between the sum of the variances and the expectation is determined. In addition, more types of body postures can be identified, and according to the pressure changes of multiple airbags, and specifically using the pressure variance and pressure difference, the current user's body posture can be determined more accurately.

[0061] like Figure 2 As shown, step S410, judging the current user's body posture according to calculating and comparing the sum of the variances of two adjacent preset time intervals includes:

[0062] If the difference between the sum of the variances of the first time interval and the sum of the variances of the second time interval is greater than a first preset value, it is determined that the user's body posture is in bed.

[0063] Specifically, when the user gets into bed, the pressure on the airbag will suddenly increase after the user gets into bed, and the increase is large. By setting a first preset value ΔP1 and judging whether the difference between S(T-1) and ST is greater than ΔP1, it is judged whether the user has gotten into bed. That is, when S(T-1)-ST>ΔP1, it can be judged that the user has gotten into bed. The value of ΔP1 can be set according to the weight of the user. Different values ​​can be set for ΔP1 according to the different weights of the users. It is also related to the initial pressure of the airbag and the hardness of the comfort layer, and can be changed according to actual testing, so as to ensure accurate recognition of the user's bed-entering status.

[0064] When making a comparison, the difference change of the mattress within two preset time intervals is monitored in real time, so as to make an instant judgment. The difference between S2 and S1, the difference between S3 and S2, and the difference between S4 and S3 are detected respectively. The detection is continuous until it is recognized that the difference between S(T-1) and ST is greater than ΔP1. After it is determined that the user has gone to bed, the recognition detection can be paused.

[0065] like Figure 3 As shown, step S411, after determining that the user's body posture is in bed, includes:

[0066] If the absolute value of the sum of the variances of the second time interval and the difference between the sum of the variances of the first time interval is greater than a fourth preset value; and the absolute value of the sum of the variances of the second time interval and the difference between the sum of the variances of the first time interval is less than the first preset value, it is determined that the user's body posture has changed at this time.

[0067] Generally, when a user has good sleep quality, the number of body movements at night will be less. When the user's sleep quality is poor, the body will subconsciously change different sleeping positions to maintain its own comfort. By recording the number of body movements of the user in bed, it can be used as a parameter indicator to judge the user's sleep quality.

[0068] Specifically, after the user gets into bed, the user's body posture can be further determined by setting a fourth preset value ΔP0 and determining whether the absolute value of the difference between ST and S(T-1) is between the fourth preset value ΔP0 and the first preset value ΔP1. If |ST-S(T-1)| is between ΔP0 and ΔP1, it can be determined that the user has body movement. That is, when ΔP0<|ST-S(T-1)|<ΔP1, it is determined that the user has body movement, such as turning over, moving, etc. The value of ΔP0 can be changed according to the actual situation, and is also related to the initial pressure of the airbag and the hardness of the comfort layer, and can be changed according to actual testing.

[0069] like Figure 3 As shown, step S412, after determining that the user's body posture is in the bed state at this time, also includes:

[0070] If the absolute value of the difference between the sum of the variance of the second time interval and the sum of the variance of the first time interval is less than a fourth preset value, it is determined that the user's body posture has not changed at this time.

[0071] Specifically, after the user gets into bed, the user's body posture can be further determined by setting a fourth preset value ΔP0 and determining whether the absolute value of the difference between ST and S(T-1) is less than the fourth preset value ΔP0. If |ST-S(T-1)| is less than ΔP0, it can be determined that the user has no body movement at this time. That is, when ΔP0<|ST-S(T-1)|<ΔP0, it is determined that the user has body movement at this time.

[0072] like Figure 4 As shown, step S413, after determining that the user's body posture has changed, further includes:

[0073] When the absolute value of the difference between the variance of any one of the airbags in the second time interval and the variance of the airbag in the first time interval is greater than a second preset value, it is determined that the pressure of the airbag has changed to determine the user body area corresponding to the airbag.

[0074] Specifically, by setting a second preset value ΔP2 and judging whether the absolute value of the difference between Vti and V(t-1)i is greater than the second preset value ΔP2, it is judged at which part of the airbag the pressure change occurs. A plurality of long strip airbags can be arranged in parallel along the length direction of the mattress. The number of airbags can be customized according to actual needs. The airbags can correspond to different parts of the user's body. By setting a plurality of airbags, body motion detection can be performed on different parts of the user's body.

[0075] For example: when the first airbag Vt1-V(t-1)1>ΔP2, it means that the pressure of the first airbag has changed, so it can be considered that the posture of the body part corresponding to the first airbag has changed. When the second airbag Vt2-V(t-1)2>ΔP2, it means that the pressure of the second airbag has changed, so it can be considered that the posture of the body part corresponding to the second airbag has changed. The judgment methods for the third and fourth airbags are similar to the above, and will not be repeated here.

[0076] like Figure 4 As shown, step S414, after determining the user body area corresponding to the airbag, also includes:

[0077] If it is detected that the difference between the air pressure value of the first airbag located on the shoulder / back of the user in the second time interval and the air pressure value of the first airbag in the first time interval is greater than the third preset value, it is determined that the user's body posture has changed from lying flat to lying sideways.

[0078] Specifically, by setting the third preset value, the first airbag can be marked as Pt1, and by comparing whether the difference between Pt1 and P(t-1)1 is greater than ΔP3, it can be judged whether the pressure of the first airbag located on the user's shoulder / back has changed. That is, when Pt1-P(t-1)1>ΔP3, the user changes from a lying position to a side-lying position. Because when the user switches from lying flat to lying on the side, his shoulder will produce a greater pressure on the mattress airbag than when lying flat. Therefore, according to relevant factors such as the user's weight, the initial pressure of the airbag, and the hardness of the comfort layer, jointly determining the third preset value can ensure accurate recognition of the user's transition from a lying position to a side-lying position.

[0079] like Figure 4 As shown, step S415, after determining that the user's body posture has changed, further includes:

[0080] If the difference between the air pressure value of the first airbag located on the shoulder / back of the user in the first time interval and the air pressure value of the first airbag in the second time interval is greater than the third preset value, it is determined that the user's body posture has changed from lying on its side to lying flat.

[0081] Specifically, by setting the third preset value, the first airbag can be marked as Pt1, and by comparing whether the difference between P(t-1)1 and Pt1 is greater than ΔP3, it can be judged whether the pressure of the first airbag located on the user's shoulder / back has changed. That is, when P(t-1)1-Pt1>ΔP3, the user changes from a side-lying posture to a flat posture. Because when the user switches from lying on the side to lying flat, the pressure exerted on the mattress airbag when the user is lying flat will be smaller than the pressure when lying on the side. Therefore, according to relevant factors such as the user's weight, the initial pressure of the airbag, and the hardness of the comfort layer, jointly determining the third preset value can ensure accurate recognition of the user's conversion from a side-lying posture to a flat posture.

[0082] like Figure 2 As shown, step S420, judging the body posture of the current user based on calculating and comparing the sum of the variances of two adjacent preset time intervals also includes:

[0083] If the difference between the sum of the variances of the second time interval and the sum of the variances of the first time interval is greater than a first preset value, it is determined that the user's body posture is out of bed at this time.

[0084] Specifically, when the user leaves the bed, the pressure on the airbag will suddenly decrease after the user leaves the bed, and the decrease is large. By setting a first preset value ΔP1 and judging whether the difference between ST and S(T-1) is less than ΔP1, it is judged whether the user has left the bed. That is, when ST-S(T-1)>ΔP1, it can be judged that the user has left the bed.

[0085] like Figure 5 As shown, step S110, the air pressure value in each air bag in the mattress is obtained at each preset time interval, including:

[0086] If the difference between the pressure value of any one of the airbags in the second time interval and the pressure value of the airbag in the first time interval is less than a fifth preset value, it is determined that the mattress is in an empty state at this time.

[0087] Specifically, by setting the fifth preset value ΔP4, and judging whether the absolute value of the difference between Vti and V(t-1)i is less than the fifth preset value ΔP4, that is, Vti-V(t-1)i<ΔP4, it can be judged that the mattress is in an empty state. The value of ΔP4 can be set according to the weight of the user, and different values ​​can be set for ΔP4 according to different weights of the users.

[0088] For example, when placing sundries, quilts, pillows and other items on the mattress, since the weight of these items is much smaller than the weight of the human body, such items can be excluded by setting an appropriate ΔP4 value, thus avoiding misidentification of the mattress due to other heavy items placed on the mattress, thereby ensuring the accuracy of the mattress's recognition of the human body's posture.

[0089] like Figure 6 As shown, an embodiment of the present invention provides a detection device applied to a smart mattress, the detection device comprising:

[0090] An acquisition module is used to acquire the air pressure value of each air bag in the mattress at a preset time interval. The acquisition module can acquire the air pressure value of the air bag once at a preset time interval to ensure the accuracy of real-time detection of the air bag pressure;

[0091] A calculation module, used to calculate the variance corresponding to each of the airbags in adjacent preset time intervals according to the air pressure value, and used to calculate the sum of the variances of all the airbags in adjacent preset time intervals according to the variance;

[0092] The judgment module is used to judge the body posture of the current user according to the calculation and comparison of the sum of the variances of adjacent preset time intervals, and to recognize different body postures of the user through different judgment conditions.

[0093] Another embodiment of the present invention provides a smart mattress, the smart mattress comprising a pressure sensor and a processor;

[0094] The pressure sensor is used to obtain the air pressure value of each airbag in the airbag mattress and send the air pressure value of each airbag to the processor;

[0095] The processor is used to calculate the variance corresponding to each airbag according to the air pressure value of each airbag within a preset time interval, to calculate the sum of the variances of all the airbags according to the variance corresponding to each airbag, and to judge the body posture of the current user by calculating and comparing the sum of the variances of two adjacent preset time intervals.

[0096] Specifically, the pressure sensor can detect the pressure of the airbag within a preset time, and the pressure sensor can detect the air pressure value of each frame of the airbag respectively, preferably, the time of each frame is 3s, and store the detected air pressure value in the storage system. The processor can read the data in the storage system, and process the data to determine the current body posture of the user, and store the processed data in the storage system.

[0097] After the processor determines the user's body posture, the storage system can store the information, and the communication module can send the user's sleep data to the user's mobile phone or the cloud for sleep quality detection and analysis. For example, the overall sleep duration can be determined based on the time of entering and leaving the bed, and deep sleep and light sleep can be determined based on body movements. Users can intuitively view their own sleep quality based on the analysis of sleep data.

[0098] Although the disclosure is disclosed as above, the protection scope of the disclosure is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the disclosure, and these changes and modifications will fall within the protection scope of the present invention.

Claims

1. A body posture recognition method, It is characterized in that include: Obtain the air pressure value in each air bag in the mattress at each preset time interval; Calculate the variance corresponding to each of the airbags within adjacent preset time intervals according to the air pressure value; Calculate the sum of variances of all the airbags in adjacent preset time intervals according to the variances; The body posture of the current user is determined by calculating and comparing the sum of the variances of adjacent preset time intervals.

2. The body posture recognition method according to claim 1, It is characterized in that The step of determining the current user's body posture by calculating and comparing the sum of the variances of two adjacent preset time intervals comprises: If the difference between the sum of the variances of the first time interval and the sum of the variances of the second time interval is greater than a first preset value, it is determined that the user's body posture is in bed.

3. The body posture recognition method according to claim 2, It is characterized in that After determining that the user's body posture is in the bed state, the method includes: If the absolute value of the difference between the sum of the variances of the second time interval and the sum of the variances of the first time interval is greater than a fourth preset value; If the absolute value of the difference between the sum of the variances of the second time interval and the sum of the variances of the first time interval is smaller than the first preset value, it is determined that the user's body posture has changed.

4. The body posture recognition method according to claim 3, It is characterized in that After determining that the user's body posture has changed, the method further includes: When the absolute value of the difference between the variance of any one of the airbags in the second time interval and the variance of the airbag in the first time interval is greater than a second preset value, it is determined that the pressure of the airbag has changed to determine the user body area corresponding to the airbag.

5. The body posture recognition method according to claim 4, It is characterized in that After determining the user body area corresponding to the airbag, the method further includes: If it is detected that the difference between the air pressure value of the first airbag located on the shoulder / back of the user in the second time interval and the air pressure value of the first airbag in the first time interval is greater than the third preset value, it is determined that the user's body posture has changed from lying flat to lying sideways.

6. The body posture recognition method according to claim 4, It is characterized in that After determining that the user's body posture has changed, the following steps are also included: If the difference between the air pressure value of the first airbag located on the shoulder / back of the user in the first time interval and the air pressure value of the first airbag in the second time interval is greater than the third preset value, it is determined that the user's body posture has changed from lying on its side to lying flat.

7. The body posture recognition method according to claim 2, It is characterized in that After determining that the user's body posture is in the bed state, the method further includes: If the absolute value of the difference between the sum of the variance of the second time interval and the sum of the variance of the first time interval is less than a fourth preset value, it is determined that the user's body posture has not changed at this time.

8. The body posture recognition method according to claim 1, It is characterized in that The step of determining the current user's body posture by calculating and comparing the sum of the variances of two adjacent preset time intervals further comprises: If the difference between the sum of the variances of the second time interval and the sum of the variances of the first time interval is greater than a first preset value, it is determined that the user's body posture is out of bed.

9. The body posture recognition method according to claim 1, It is characterized in that The step of obtaining the air pressure value in each air bag in the mattress at each preset time interval includes: If the difference between the pressure value of any one of the airbags in the second time interval and the pressure value of the airbag in the first time interval is less than a fifth preset value, it is determined that the mattress is in an empty state at this time.

10. A detection device, applied to a smart mattress, It is characterized in that include: An acquisition module, used for acquiring the air pressure value in each air bag in the mattress at each preset time interval; A calculation module, used to calculate the variance corresponding to each of the airbags in adjacent preset time intervals according to the air pressure value, and used to calculate the sum of the variances of all the airbags in adjacent preset time intervals according to the variance; The judging module is used to judge the body posture of the current user by calculating and comparing the sum of the variances of adjacent preset time intervals.

11. A smart mattress, It is characterized in that including a pressure sensor and a processor; The pressure sensor is used to obtain the air pressure value of each airbag in the airbag mattress and send the air pressure value of each airbag to the processor; The processor is used to calculate the variance corresponding to each airbag according to the air pressure value of each airbag within a preset time interval, to calculate the sum of the variances of all the airbags according to the variance corresponding to each airbag, and to judge the body posture of the current user by calculating and comparing the sum of the variances of two adjacent preset time intervals.