Sleeping posture detection device and method based on bending sensor

Abstract

The invention provides a sleeping posture detection method based on bending sensors, which comprises the following steps: S1, respectively acquiring bending angles of each bending sensor when a human body lies on the back, lies on the left side and lies on the right side, and respectively establishing a supine angle sample set, a left-side angle sample set and a right-side angle sample set; s2, according to the angle sample sets of left lying and right lying, calculating a discrimination threshold value of supine lying and side lying; s3, according to the supine angle sample set, calculating an angle difference value set of the bending sensors on the two sides; s4, according to the angle difference value set, calculating a judgment threshold value of left lying and right lying; and S5, judging the sleep posture of the human body according to the judgment threshold. The pressure condition of the mattress is detected by combining a plurality of bending sensors, so that the sleeping posture of a human body is judged, a camera is not needed, and the problems that the privacy of a user is leaked and the detection result is easily interfered by the environment are fundamentally avoided; and the detection precision can be ensured without arranging a relatively large sensor array, so that the hardware laying cost is greatly reduced, and large-scale production and family popularization are facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of sleeping posture detection, and in particular to a sleeping posture detection device and method based on a bending sensor. Background Art

[0002] With the rapid development of the economy, people's living standards are constantly improving, but the pace of life is also accelerating. The "China Sleep Research Report (2024)" shows that the average sleep duration of Chinese people has decreased by approximately 1.5 hours since 2012. Poor sleep quality is the main problem that bothers respondents. Sleeping posture, as one of the important parameters for measuring sleep quality, can be used to improve sleep and prevent diseases. Sleeping posture is closely related to human health. An improper sleeping posture can cause spinal curvature, affect breathing conditions, and even cause long-term health problems. With the aging population and the shortage of nursing staff in today's world, using human resources to monitor sleep posture is challenging.

[0003] With the continuous integration of sensing technology and artificial intelligence technology, more and more researchers have begun to use various sensors and machine learning methods to monitor human sleeping posture. Currently, the methods of sleeping posture recognition can be divided into three categories: 1) Using thermal imaging or infrared imaging technology, the sleeping posture image is input into the classifier for sleeping posture recognition (references [1], [2]); 2) Sleeping posture recognition based on vital sign data recorded by wearable devices such as accelerometers (references [3], [4]); 3) Based on the pressure sensor array, the human body pressure distribution image is obtained for sleeping posture recognition (references [5], [6]).

[0004] However, among the above methods, the first type of image and video-based methods require a camera device, and the recognition results have problems such as privacy leakage and interference from the brightness of the ambient field of view; the second type of methods based on wearable devices have the risk of interfering with natural sleep and affecting the accuracy of data collection; the third type of methods based on pressure sensor arrays often require a larger sensor array to improve recognition accuracy, and the cost of hardware deployment is high, which limits its promotion and application in daily home monitoring.

[0005] Based on the above problems, there is an urgent need to invent a sleeping posture detection method in which the detection results are not affected by the environment, the data collection accuracy is high, and the cost is low.

[0006] References: Literature[1]: Chen Z, Wang Y. Remote recognition of in-bed postures using athermopile array sensor with machine learning[J]. IEEE Sensors Journal, 2021,21(9): 10428-10436. Literature [2]: Mahvash SM, Shirin E, Adrian H, et al. Transfer Learning for Clinical Sleep Pose Detection using a Single 2D IR Camera. [J]. IEEE transactions on neural systems and rehabilitation engineering: a publication of the IEEE Engineering in Medicine and Biology Society, 2020, PP. Literature [3]: Abdulsadig RS, Rodriguez-Villegas E. Sleep posture monitoring using a single neck-situated accelerometer: A proof-of-concept[J]. IEEE Access, 2023, 11: 17693-17706. Literature [4]: ​​Fallmann S, Van Veen R, Chen L, et al. Wearable accelerometerbased extended sleep position recognition[C] / / 2017 IEEE 19th International Conference on e-Health Networking, Applications and Services (Healthcom). IEEE, 2017: 1-6. Reference [5]: Huang Zhen, Yu Suiran. Low-cost and high-performance sleeping posture monitoring system based on pressure sensor array [J]. Electronic Measurement Technology, 2023, 46(19): 14-20. DOI: 10.19651 / j.cnki.emt.2212488. Reference [6]: Zhao Haiwen, Zhao Yuekun, Qi Dexuan, et al. Real-time sleeping posture monitoring system based on unconstrained pressure sensor [J]. Sensors and Microsystems, 2025, 44(03): 120-124. DOI: 10.13873 / J.1000-9787(2025)03-0120-05. Summary of the Invention

[0007] The present invention proposes a sleeping posture detection device and method based on a bending sensor, which solves the problems of the sleeping posture detection method in the prior art, such as the detection results are easily affected by the environment, the data collection accuracy is low, and the cost is high.

[0008] The technical solution of the present invention is achieved as follows: A first aspect of the present invention provides a sleeping posture detection device based on a bending sensor, comprising a microprocessor, a power supply module, and a first detection component and a second detection component arranged on a mattress, wherein the microprocessor determines the sleeping posture of a human body based on data detected by the first detection component and the second detection component; the arrangement area of ​​the first detection component corresponds to the back area of ​​the human body, and the arrangement area of ​​the second detection component corresponds to the buttocks area of ​​the human body; the first detection component comprises a first bending sensor and a second bending sensor symmetrically arranged on the left and right sides of the back of the human body, and a third bending sensor arranged in the center of the back of the human body; the second detection component comprises a fourth bending sensor and a fifth bending sensor symmetrically arranged on the left and right sides of the buttocks of the human body, and a sixth bending sensor arranged in the center of the buttocks of the human body.

[0009] Specifically, the sleeping posture detection device also includes a signal amplifier, a filter and an analog-to-digital converter connected in sequence. The output ends of the first detection component and the second detection component are connected to the signal amplifier, and the output end of the analog-to-digital converter is connected to the microprocessor.

[0010] Specifically, a plurality of transverse airbags are arranged on the mattress along the length direction, and the first detection component and the second detection component are respectively arranged on two different airbags.

[0011] A second aspect of the present invention provides a sleeping posture detection method based on a bending sensor, comprising the following steps: S1, respectively obtain the bending angles of each bending sensor when the human body lies on the back, left side, and right side, and establish the angle sample sets for lying on the back, left side, and right side respectively; S2, based on the angle sample sets of lying on the left side and lying on the right side, calculate the discrimination thresholds of supine and side lying in the back area and buttocks area respectively; S3, based on the supine angle sample set, respectively calculate the angle difference value set of the bending sensors on both sides of the back area and the buttocks area; S4, calculating the discrimination thresholds for left-side lying and right-side lying in the back region and the buttocks region respectively based on the angle difference set of the bending sensors on both sides of the back region and the buttocks region; S5, judging the sleeping posture of the human body according to the discrimination thresholds of supine and side lying and the discrimination thresholds of left and right lying in the back region and the buttocks region.

[0012] Specifically, step S2 includes the following steps: Calculate the discrimination threshold between supine and lateral lying in the back area : ; in, The angle value of the third bending sensor in the left and right side lying angle sample sets The mean of The angle value of the third bending sensor in the left and right side lying angle sample sets The standard deviation of is the coefficient; Calculate the discrimination threshold between supine and lateral lying in the buttocks area : ; in, The angle value of the sixth bending sensor in the left and right side lying angle sample sets The mean of The angle value of the sixth bending sensor in the left and right side lying angle sample sets The standard deviation of is the coefficient.

[0013] Specifically, step S3 includes the following steps: Calculate the angle difference of the bending sensors on both sides of the back area : ; , ; in, n is the total number of samples in the supine angle sample set; and Respectively i angle values ​​of the first bending sensor and the second bending sensor within a supine angle sample; Calculate the angle difference of the bending sensors on both sides of the hip area : ; , ; in, and Respectively i The angle values ​​of the fourth bending sensor and the fifth bending sensor within the supine angle sample.

[0014] Specifically, step S4 includes the following steps: Calculate the discrimination threshold between left and right side lying in the back area : ; in, Represents a set of angle differences The maximum angle difference in is the coefficient; Calculate the discrimination threshold between lying on the left side and lying on the right side in the buttocks area : ; in, Represents a set of angle differences The maximum angle difference in is the coefficient.

[0015] Specifically, in step S5, the conditions for determining the human body's sleeping posture are: When satisfied When , it is judged as supine; When satisfied When the patient lies on the left side, he / she is judged to be lying on the left side; When satisfied When the patient lies on the right side, he / she is judged to be lying on the right side; in, , .

[0016] A third aspect of the present invention provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and the processor implements the steps of the sleeping posture detection method when executing the computer program.

[0017] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the sleeping posture detection method are implemented.

[0018] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention combines multiple bending sensors to detect the pressure of the mattress, thereby determining the sleeping posture of the human body, without the need for a camera. This fundamentally avoids the problems of user privacy leakage and the detection results being susceptible to environmental interference. Moreover, the detection accuracy can be guaranteed without the need to deploy a large sensor array, which greatly reduces the hardware laying cost and is easy to mass-produce and popularize in households. (2) The present invention uses a rule-based threshold discrimination method to determine sleeping posture. Compared with the traditional machine learning-based discrimination method, it does not require the consumption of a large amount of computing resources. The detection process is highly real-time and has a fast response speed. It can be easily ported to a low-computing-power microcontroller, thereby reducing system power consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 This is a schematic diagram of the bending sensor layout of a bending sensor-based sleeping posture detection device of the present invention; Figure 2 This is a schematic diagram of the hardware circuit connection of the sleeping posture detection device in an embodiment of the present invention; Figure 3 Schematic diagram of a flow chart of a sleeping posture detection method based on a bending sensor according to the present invention; In the figure: 1. Microprocessor; 2. Power module; 3. Mattress; 4. First bending sensor; 5. Second bending sensor; 6. Third bending sensor; 7. Fourth bending sensor; 8. Fifth bending sensor; 9. Sixth bending sensor; 10. Signal amplifier; 11. Filter; 12. Analog-to-digital converter; 13. Airbag. DETAILED DESCRIPTION

[0021] The following will clearly and completely describe the technical solutions of the present invention in conjunction with the embodiments of the present invention. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0022] Reference Figure 1 、 2In a first aspect, the present invention provides a sleeping posture detection device based on a bending sensor, comprising a microprocessor 1, a power module 2, and a first detection component and a second detection component arranged on a mattress 3. The microprocessor 1 determines a sleeping posture of a human body based on data detected by the first detection component and the second detection component. The arrangement area of ​​the first detection component corresponds to the back area of ​​the human body, and the arrangement area of ​​the second detection component corresponds to the buttocks area of ​​the human body. The first detection component includes a first bending sensor 4 and a second bending sensor 5 symmetrically arranged on the left and right sides of the back of the human body, and a third bending sensor 6 arranged in the center of the back of the human body. The second detection component includes a fourth bending sensor 7 and a fifth bending sensor 8 symmetrically arranged on the left and right sides of the buttocks of the human body, and a sixth bending sensor 9 arranged in the center of the buttocks of the human body.

[0023] In this embodiment, the left-right spacing between the first bend sensor 4 and the second bend sensor 5 is 10 cm, and the vertical spacing between the third bend sensor 6 and the first or second bend sensor 4 and 5 is 5 cm. The left-right and vertical spacing between the fourth bend sensor 7, the fifth bend sensor 8, and the sixth bend sensor 9 can be referenced to the spacing between the first bend sensor 4, the second bend sensor 5, and the third bend sensor 6. The vertical spacing between the third bend sensor 6 and the sixth bend sensor 9 is 45 cm. During implementation, these spacings can be flexibly adjusted based on actual conditions.

[0024] In this embodiment, the bend sensor is essentially a flexible variable resistor whose resistance increases with its degree of bending (i.e., bending angle). By leveraging the principle that physical bending causes structural changes in the internal conductive material, this sensor converts mechanical deformation into a measurable electrical signal. The bend sensor in this embodiment can directly utilize commercially available products, such as the Adafruit FLX-03, the Spectra Symbol Flex Sensor FS-L-0055-103-ST, or the DFRobot SEN0004 (V2). The specific model used can be flexibly adjusted based on actual needs.

[0025] Specifically, if Figure 2 As shown, the sleeping posture detection device also includes a signal amplifier 10, a filter 11 and an analog-to-digital converter 12 connected in sequence. The output ends of the first detection component and the second detection component are connected to the signal amplifier 10, and the output end of the analog-to-digital converter 12 is connected to the microprocessor 1.

[0026] In this embodiment, six bending sensors each lead out six signal lines and are connected to the input end of the signal amplifier 10. The signal amplifier 10 amplifies the analog signal output by the bending sensor, and then filters the amplified signal through the filter 11. The amplified and filtered analog signal is then converted into a digital signal through the analog-to-digital converter 12 and input into the microprocessor 1 for processing. The microprocessor 1 automatically determines and outputs the current sleeping posture status to the monitoring background terminal based on the six detection signals and the stored sleeping posture detection algorithm.

[0027] Specifically, if Figure 1 As shown, a plurality of transverse airbags 13 are arranged along the length direction of the mattress 3, and the first detection component and the second detection component are respectively arranged on two different airbags 13; by arranging the bending sensor on the airbag 13, the bending sensor is more easily squeezed by the human body to produce bending deformation, thereby amplifying the signal collected by the sensor and improving the sensitivity of data acquisition.

[0028] like Figure 3 As shown, the second aspect of the present invention provides a sleeping posture detection method based on a bending sensor, comprising the following steps: S1, respectively obtain the bending angles of each bending sensor when the human body lies on the back, left side, and right side, and establish the angle sample sets for lying on the back, left side, and right side respectively; S2, based on the angle sample sets of lying on the left side and lying on the right side, calculate the discrimination thresholds of supine and side lying in the back area and buttocks area respectively; S3, based on the supine angle sample set, respectively calculate the angle difference value set of the bending sensors on both sides of the back area and the buttocks area; S4, calculating the discrimination thresholds for left-side lying and right-side lying in the back region and the buttocks region respectively based on the angle difference set of the bending sensors on both sides of the back region and the buttocks region; S5, judging the sleeping posture of the human body according to the discrimination thresholds of supine and side lying and the discrimination thresholds of left and right lying in the back region and the buttocks region.

[0029] Specifically, in step S1, the angle values ​​of each bending sensor in the back and buttocks area of ​​the user are obtained when the human body lies on the back, left side, and right side on the mattress 3 respectively; the first bending sensor 4 to the sixth bending sensor 9 are numbered in advance, and the obtained sensor angle values ​​correspond to the numbers one by one. In this embodiment, the number of repeated sampling is 30 times (i.e., 30 times for lying on the back, 30 times for lying on the left side, and 30 times for lying on the right side). The specific number of sampling times can be flexibly adjusted according to actual conditions. In theory, the more sampling times, the higher the subsequent judgment accuracy; based on the data samples collected in different sleeping positions, a sensor bending angle sample set P for lying on the back, a sensor bending angle sample set L for lying on the left side, and a sensor bending angle sample set R for lying on the right side are respectively established.

[0030] Specifically, step S2 includes the following steps: Calculate the discrimination threshold between supine and lateral lying in the back area : ; in, is the angle value of the third bending sensor 6 in the left and right side lying angle sample sets The mean of is the angle value of the third bending sensor 6 in the left and right side lying angle sample sets The standard deviation of is the coefficient; Calculate the discrimination threshold between supine and lateral lying in the buttocks area : ; in, is the angle value of the sixth bending sensor 9 in the left and right side lying angle sample sets The mean of is the angle value of the sixth bending sensor 9 in the left and right side lying angle sample sets The standard deviation of is the coefficient; coefficient 、 It can be obtained through multiple experiments and is used to prevent misjudgment of sleeping posture.

[0031] Specifically, step S3 includes the following steps: Calculate the angle difference of the bending sensors on both sides of the back area : ; , ; in, n is the total number of samples in the supine angle sample set; and Respectively i Angle values ​​of the first bending sensor 4 and the second bending sensor 5 within a supine angle sample; Calculate the angle difference of the bending sensors on both sides of the hip area : ; , ; in, and Respectively i The angle values ​​of the fourth bending sensor 7 and the fifth bending sensor 8 within the supine angle samples.

[0032] Specifically, step S4 includes the following steps: Calculate the discrimination threshold between left and right side lying in the back area : ; in, Represents a set of angle differences The maximum angle difference in is the coefficient; Calculate the discrimination threshold between lying on the left side and lying on the right side in the buttocks area : ; in, Represents a set of angle differences The maximum angle difference in is the coefficient, the coefficient 、 It can be obtained through multiple experiments and is used to prevent misjudgment of sleeping posture.

[0033] Specifically, in step S5, the conditions for determining the human body's sleeping posture are: When satisfied When , it is judged as supine; When satisfied When the patient lies on the left side, he / she is judged to be lying on the left side; When satisfied When the patient lies on the right side, he / she is judged to be lying on the right side; in, , .

[0034] A third aspect of the present invention provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and the processor implements the steps of the sleeping posture detection method when executing the computer program.

[0035] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the sleeping posture detection method are implemented.

[0036] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A sleeping posture detection device based on a bending sensor, characterized in that: The invention comprises a microprocessor (1), a power module (2), and a first detection component and a second detection component arranged on a mattress (3); the microprocessor (1) determines the sleeping posture of a human body based on data detected by the first detection component and the second detection component; The arrangement area of ​​the first detection component corresponds to the back area of ​​the human body, and the arrangement area of ​​the second detection component corresponds to the buttocks area of ​​the human body; the first detection component includes a first bending sensor (4) and a second bending sensor (5) symmetrically arranged on the left and right sides of the back of the human body, and a third bending sensor (6) arranged in the center of the back of the human body; the second detection component includes a fourth bending sensor (7) and a fifth bending sensor (8) symmetrically arranged on the left and right sides of the buttocks of the human body, and a sixth bending sensor (9) arranged in the center of the buttocks of the human body.

2. The sleeping posture detection device based on a bending sensor according to claim 1, characterized in that: The sleeping posture detection device further comprises a signal amplifier (10), a filter (11) and an analog-to-digital converter (12) connected in sequence, wherein the output ends of the first detection component and the second detection component are connected to the signal amplifier (10), and the output end of the analog-to-digital converter (12) is connected to the microprocessor (1).

3. The sleeping posture detection device based on a bending sensor according to claim 1, characterized in that: A plurality of transverse airbags (13) are arranged on the mattress (3) along the length direction, and the first detection component and the second detection component are respectively arranged on two different airbags (13).

4. A sleeping posture detection method based on a bending sensor, based on the sleeping posture detection device according to any one of claims 1 to 3, characterized in that: The following steps are involved: S1, respectively obtain the bending angles of each bending sensor when the human body lies on the back, left side, and right side, and establish the angle sample sets for lying on the back, left side, and right side respectively; S2, based on the angle sample sets of lying on the left side and lying on the right side, calculate the discrimination thresholds of supine and side lying in the back area and buttocks area respectively; S3, based on the supine angle sample set, respectively calculate the angle difference value set of the bending sensors on both sides of the back area and the buttocks area; S4, calculating the discrimination thresholds for left-side lying and right-side lying in the back region and the buttocks region respectively based on the angle difference set of the bending sensors on both sides of the back region and the buttocks region; S5, judging the sleeping posture of the human body according to the discrimination thresholds of supine and side lying and the discrimination thresholds of left and right lying in the back region and the buttocks region.

5. The sleeping posture detection method based on a bending sensor according to claim 4, characterized in that: Step S2 includes the following steps: Calculate the discrimination threshold between supine and lateral lying in the back area : ； in, The angle value of the third bending sensor (6) in the left side lying and right side lying angle sample sets The mean of The angle value of the third bending sensor (6) in the left side lying and right side lying angle sample sets The standard deviation of is the coefficient; Calculate the discrimination threshold between supine and lateral lying in the buttocks area : ； in, The angle value of the sixth bending sensor (9) in the left-side lying and right-side lying angle sample sets The mean of The angle value of the sixth bending sensor (9) in the left-side lying and right-side lying angle sample sets The standard deviation of is the coefficient.

6. The sleeping posture detection method based on a bending sensor according to claim 5, characterized in that: Step S3 includes the following steps: Calculate the angle difference of the bending sensors on both sides of the back area : ； ， ； in, n is the total number of samples in the supine angle sample set; and Respectively i Angle values ​​of the first bending sensor (4) and the second bending sensor (5) within a supine angle sample; Calculate the angle difference of the bending sensors on both sides of the hip area : ； ， ； in, and Respectively i The angle values ​​of the fourth bending sensor (7) and the fifth bending sensor (8) within the supine angle samples.

7. The sleeping posture detection method based on a bending sensor according to claim 6, characterized in that: Step S4 includes the following steps: Calculate the discrimination threshold between left and right side lying in the back area : ； in, Represents a set of angle differences The maximum angle difference in is the coefficient; Calculate the discrimination threshold between lying on the left side and lying on the right side in the buttocks area : ； in, Represents a set of angle differences The maximum angle difference in is the coefficient.

8. The sleeping posture detection method based on a bending sensor according to claim 7, characterized in that: In step S5, the conditions for determining the sleeping posture of the human body are: When satisfied When , it is judged as supine; When satisfied When the patient lies on the left side, he / she is judged to be lying on the left side; When satisfied When the patient lies on the right side, he / she is judged to be lying on the right side; in, , .

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, wherein: When the processor executes the computer program, the steps of the sleeping posture detection method according to any one of claims 4 to 8 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the sleeping posture detection method according to any one of claims 4 to 8 are implemented.

Images