Sitting posture recognition method

Abstract

The present application relates to the field of intelligent seats, and more particularly to a sitting posture recognition method, which adopts multiple pressure sensors distributed on a seating surface and performs the following steps: Step 1, seat detection; if the seat is detected, Step 2 is performed to obtain an initial pressure value when the user sits; if the seat is not detected, Step 1 is repeatedly performed; Step 2, judging whether to trigger the sitting posture detection; Step 3 is performed after the sitting posture detection is triggered; Step 3, sitting posture detection judgment: Step 3.1, preliminary detection judgment: first, compare the difference between the integral value of each sensor and the correct sitting posture state setting integral value with a second threshold value within a T2 time period; if the difference is greater than the second threshold value, it is judged as an incorrect sitting posture; if the difference is not greater than the second threshold value, Step 3.2 is performed; re-detection judgment: the pressure values of each two sensors of all sensors are subjected to integral processing to obtain a two-sensor integral difference value, and the integral difference value is compared with a third threshold value set for the correct sitting posture state.

Description

Technical Field

[0001] This invention relates to the field of smart seats, and more particularly to a sitting posture recognition method. Background Technology

[0002] Office chairs are chairs provided for convenience during daily work and social activities. Based on ergonomic principles, existing office chairs offer supportive designs to help users maintain correct posture. However, a significant number of users still lack awareness of maintaining correct posture and frequently work in incorrect positions, such as hunching over or crossing their legs. Furthermore, with increased office hours, the muscles in the lower back and neck are prone to static fatigue. In a state of muscle fatigue, users find it even more difficult to maintain correct posture, unconsciously adopting a slouching or crooked posture, which is extremely detrimental to health in the long run. Therefore, chair design needs to consider how to accurately identify users' poor posture and provide reminders.

[0003] Based on this, Chinese invention patent publication number "CN115153230A" describes an intelligent health chair and its posture recognition method. It includes a backrest, a seat, armrests on both sides of the seat, and a chassis below the seat. Pressure sensors are installed in the seat, with multiple pressure sensing points arrayed on the sensors. A motherboard is connected to the lower end of the seat and electrically connected to the pressure sensors. The motherboard acquires pressure data from the multiple pressure sensing points and compares it with pressure data from a preset correct sitting posture to determine if the user is in an incorrect sitting posture. In this technical solution, when a user uses the intelligent health chair, the arrayed pressure sensing points can accurately collect the physical data applied to each sensing point by the human body in the current sitting posture. The motherboard receives this data and compares it one-to-one with the data from each pressure point in the correct sitting posture to determine if the user is in an incorrect sitting posture.

[0004] The prior patent primarily aimed to detect changes in pressure sensor points distributed in a matrix. It compared the reading of each pressure sensor point with data from various pressure points under correct sitting posture, giving higher weight to pressure sensor points that are considered strong correlation points. However, this detection method requires a large number of pressure sensor points, resulting in high detection costs and a large computational load; therefore, improvement is urgently needed. Summary of the Invention

[0005] To address the aforementioned problems, the present invention aims to provide a sitting posture recognition method that uses fewer pressure sensors to accurately identify incorrect sitting postures.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] A sitting posture recognition method, characterized by employing multiple pressure sensors distributed on the sitting surface and performing the following steps:

[0008] Step 1, Seating Detection: Detect the pressure value of each pressure sensor and compare it with its set calibration value to determine if the user is seated. If seated, proceed to Step 2 to obtain the initial pressure value when the user sits down. If not seated, repeat Step 1. Step 2: Determine if Posture Detection is Triggered: Based on the initial pressure value when the user sits down, periodically obtain the average pressure change of each sensor within the time period T1. If the average pressure change of all sensors does not exceed the first threshold, it is determined to be a normal sitting posture, and posture detection is not performed. The current pressure value is updated and replaced with the initial pressure value, and Step 2 is repeated. If the average pressure change of any sensor exceeds the first threshold, proceed to Step 3.

[0009] Step 3: Posture detection and judgment:

[0010] Step 3.1, Preliminary detection and judgment: During the T2 time period, the pressure value of each sensor is processed by definite integration to obtain the integral value. The difference between the integral value and the integral value set for the correct sitting posture is compared with the second threshold. If it is greater than the second threshold, it is judged as an incorrect sitting posture, and an incorrect sitting posture reminder is issued. If it is not greater than the second threshold, proceed to step 3.2.

[0011] Step 3.2, re-detection and judgment: Perform definite integral processing on the pressure values ​​of every two sensors for all sensors, obtain the definite integral difference between the two sensors, and compare the definite integral difference with the third threshold set for the correct sitting posture; if it is greater than the third threshold, it is judged as an incorrect sitting posture and an incorrect sitting posture reminder is issued; if it is not greater than the third threshold, it is judged as a correct sitting posture.

[0012] This invention adopts the above-mentioned technical solution, which relates to a sitting posture recognition method. In step 1, the method first performs a sitting detection. When a sitting state is detected, in step 2, the average pressure change of each sensor within a time period T1 is obtained periodically (e.g., every 50ms). The time period T1 can be 20ms. The average pressure sensor value is calculated for 20ms. The average value is compared with a first threshold to determine whether to start step 3. In step 3, when judging the sitting posture, the method compares the difference between the integral value of each sensor within a time period T2 and the integral value set for the correct sitting posture with a second threshold. If the difference is greater than the second threshold, it is considered that the adjustment is too large compared to the correct sitting posture and is judged as an incorrect sitting posture.

[0013] However, if the difference between the integral value of each sensor and the integral value set for the correct sitting posture is not greater than the second threshold, it is also necessary to obtain the definite integral difference between every two sensors. Based on the definite integral difference, it is compared with the set third threshold to further determine whether it is an incorrect sitting posture. Only when the definite integral difference of each group is less than the third threshold is the current sitting posture considered to have been slightly adjusted and is still considered a correct sitting posture.

[0014] The aforementioned posture recognition method considers that existing methods only compare the readings of each sensor with those of a normal sitting posture, thus requiring a large number of sensors arranged in a matrix to comprehensively calculate and judge incorrect postures. This scheme, however, considers not only the definite integral value of a single sensor during the detection and judgment process, but also the difference in definite integrals between any two sensors. Based on this scheme, only a few sensors can be used in key locations. When the reading of any sensor fluctuates significantly compared to the standard reading, it can be directly judged as an incorrect posture, thereby reducing the computational load. When the fluctuation range is small, the correlation fluctuation between every two sensors needs to be considered. The principle is that when adjusting posture, the center of gravity of the human body changes, so most sensors on the seating surface will change their readings due to posture adjustment. This judgment of the correlation fluctuation between every two sensors can comprehensively consider the small changes between each pair of sensors, thus it can be used to identify posture adjustments with small differences in individual sensor readings, thereby improving the accuracy of posture recognition.

[0015] Furthermore, the correlation fluctuations based on multiple sensors can also be used to determine the direction of the center of gravity's tendency to shift, and thus to determine the type of incorrect sitting posture. For example, when adjusting from a correct sitting posture to a slightly forward-leaning posture, the readings of the left and right front sensors increase, but the increase is not significant enough to directly determine an incorrect sitting posture. At this point, it is necessary to combine the readings of the left and right rear sensors. During the re-detection and judgment process, both the left and right front sensors increase, and the difference in the definite integrals of the two sensors is small, insufficient to determine an incorrect sitting posture; however, comparing the definite integral differences between the left and left rear sensors, and the definite integral differences between the right and right front sensors, the differences are relatively large, sufficient to determine an incorrect sitting posture.

[0016] In a specific implementation, the first threshold of each sensor represents the maximum pressure change of that sensor; the second threshold of each sensor represents the maximum integral change of that sensor; and the third threshold of every two sensors represents the maximum change of the integral difference between the two sensors.

[0017] In a specific implementation, this sitting posture recognition method employs at least a left front sensor for acquiring pressure on the left thigh, a right front sensor for acquiring pressure on the right thigh, and left and right rear sensors for acquiring pressure on both sides of the buttocks. This solution uses only four sensors—left front, right front, left rear, and right rear—to achieve sitting posture judgment and ensure accuracy.

[0018] In a further embodiment, the left front sensor and right front sensor are arranged side-by-side on the left and right sides of the front of the seat surface, and the left rear sensor and right rear sensor are arranged side-by-side on the left and right sides of the middle of the seat surface. The distance d1 between the left front sensor and right front sensor is greater than the distance d2 between the left rear sensor and right rear sensor. In this embodiment, to adapt to the force distribution on the seat surface when a person is sitting, the distance d1 between the left front sensor and right front sensor is relatively large, fitting the area around the upper thighs, while the distance d2 between the left rear sensor and right rear sensor is relatively small, fitting the area around the center of the buttocks.

[0019] Preferably, the T2 time period is longer than the T1 time period, and the larger T2 time period ensures the accuracy of the integral value and the difference between the definite integrals. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the sensor layout on the seat.

[0021] Figure 2 This is a schematic diagram of sensor readings when the seat is unloaded.

[0022] Figure 3 This is a schematic diagram of sensor readings when the person is seated.

[0023] Figure 4 This is a schematic diagram of sensor readings when tilted to the left.

[0024] Figure 5 This is a schematic diagram of sensor readings when tilted to the right.

[0025] Figure 6 This is a schematic diagram of sensor readings in a forward-leaning position.

[0026] Figure 7 This is a schematic diagram of sensor readings in a tilted-back position.

[0027] Figure 8 This is a schematic diagram of sensor readings when the left leg is crossed.

[0028] Figure 9 This is a schematic diagram of sensor readings when the right leg is crossed.

[0029] Figure 10This is a schematic diagram of sensor readings under continuously changing conditions (correct, leaning left, leaning right, leaning forward, leaning backward, crossing left leg, crossing right leg, correct). Detailed Implementation

[0030] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0031] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0032] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more, unless explicitly defined otherwise.

[0033] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0034] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can include direct contact between the first and second features, or contact between the first and second features through another feature between them. Furthermore, "above," "over," and "on top" of the second feature includes the first feature directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature includes the first feature directly below or diagonally below the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.

[0035] like Figure 1 As shown, this embodiment relates to a sitting posture recognition method applicable to smart chairs using multiple pressure sensors distributed on the seating surface. As shown in the figure, this sitting posture recognition method employs at least a left front sensor for acquiring pressure on the left thigh, a right front sensor for acquiring pressure on the right thigh, and left and right rear sensors for acquiring pressure on both sides of the buttocks. This solution uses only four sensors—left front, right front, left rear, and right rear—to achieve sitting posture judgment and ensure accuracy. In a further embodiment, the left and right front sensors are arranged side-by-side on the left and right sides of the front of the seating surface, and the left and right rear sensors are arranged side-by-side on the left and right sides of the middle of the seating surface; the distance d1 between the left and right front sensors is greater than the distance d2 between the left and right rear sensors. In this embodiment, to adapt to the force distribution on the seating surface under human sitting conditions, the distance d1 between the left and right front sensors is relatively large, fitting the position of the upper thighs, while the distance d2 between the left and right rear sensors is relatively small, fitting the center position of the buttocks.

[0036] In the diagram, the width d1 between the left and right front sensors is 30cm, the distance d3 from the front edge of the seating surface is 5cm, the distance d4 from the rear edge of the seating surface is 50cm, the spacing d2 between the left and right rear sensors is 13cm, and the length d5 ​​from the line containing the left and right front sensors is 27cm. These dimensions are only those mentioned in this embodiment and can be adjusted adaptively according to actual usage, such as proportionally adjusting to the dimensions of the seating surface.

[0037] The posture recognition method performs the following steps:

[0038] Step 1, Seating Detection: Detect the pressure value of each pressure sensor and compare it with its set calibration value to determine if the user is seated. If seated, proceed to Step 2 to obtain the initial pressure value when the user sits down. If not seated, repeat Step 1. Step 2: Determine if Posture Detection is Triggered: Based on the initial pressure value when the user sits down, periodically obtain the average pressure change of each sensor within the time period T1. If the average pressure change of all sensors does not exceed the first threshold, it is determined to be a normal sitting posture, and posture detection is not performed. The current pressure value is updated and replaced with the initial pressure value, and Step 2 is repeated. If the average pressure change of any sensor exceeds the first threshold, proceed to Step 3.

[0039] Step 3: Posture detection and judgment:

[0040] Step 3.1, Preliminary detection and judgment: During the T2 time period, the pressure value of each sensor is processed by definite integration to obtain the integral value. The difference between the integral value and the integral value set for the correct sitting posture is compared with the second threshold. If it is greater than the second threshold, it is judged as an incorrect sitting posture, and an incorrect sitting posture reminder is issued. If it is not greater than the second threshold, proceed to step 3.2.

[0041] Step 3.2, re-detection and judgment: Perform definite integral processing on the pressure values ​​of every two sensors for all sensors, obtain the definite integral difference between the two sensors, and compare the definite integral difference with the third threshold set for the correct sitting posture; if it is greater than the third threshold, it is judged as an incorrect sitting posture and an incorrect sitting posture reminder is issued; if it is not greater than the third threshold, it is judged as a correct sitting posture.

[0042] In a specific implementation, the first threshold of each sensor represents the maximum pressure change of that sensor; the second threshold of each sensor represents the maximum integral change of that sensor; and the third threshold of every two sensors represents the maximum change of the integral difference between the two sensors.

[0043] In the preferred embodiment, the T2 time period is longer than the T1 time period, and the larger T2 time period ensures the accuracy of the integral value and the difference between the definite integral and the integral.

[0044] This embodiment adopts the above-mentioned technical solution, which relates to a sitting posture recognition method. In step 1, the method first performs a sitting detection. When a sitting state is detected, in step 2, the average pressure change of each sensor within a time period T1 is obtained periodically (e.g., every 50ms). The T1 time period can be 20ms. The average pressure sensor value is calculated for 20ms. The average value is compared with a first threshold to determine whether to start step 3. In step 3, when judging the sitting posture, the method compares the difference between the integral value of each sensor within a time period T2 and the integral value set for the correct sitting posture with a set second threshold. If the difference is greater than the second threshold, it is considered that the adjustment is too large compared to the correct sitting posture and is judged as an incorrect sitting posture.

[0045] However, if the difference between the integral value of each sensor and the integral value set for the correct sitting posture is not greater than the second threshold, it is also necessary to obtain the definite integral difference between every two sensors. Based on the definite integral difference, it is compared with the set third threshold to further determine whether it is an incorrect sitting posture. Only when the definite integral difference of each group is less than the third threshold is the current sitting posture considered to have been slightly adjusted and is still considered a correct sitting posture.

[0046] The aforementioned posture recognition method considers that existing methods only compare the readings of each sensor with those of a normal sitting posture, thus requiring a large number of sensors arranged in a matrix to comprehensively calculate and judge incorrect postures. This scheme, however, considers not only the definite integral value of a single sensor during the detection and judgment process, but also the difference in definite integrals between any two sensors. Based on this scheme, only a few sensors can be used in key locations. When the reading of any sensor fluctuates significantly compared to the standard reading, it can be directly judged as an incorrect posture, thereby reducing the computational load. When the fluctuation range is small, the correlation fluctuation between every two sensors needs to be considered. The principle is that when adjusting posture, the center of gravity of the human body changes, so most sensors on the seating surface will change their readings due to posture adjustment. This judgment of the correlation fluctuation between every two sensors can comprehensively consider the small changes between each pair of sensors, thus it can be used to identify posture adjustments with small differences in individual sensor readings, thereby improving the accuracy of posture recognition.

[0047] Furthermore, the correlation fluctuations based on multiple sensors can also be used to determine the direction of the center of gravity's tendency to shift, and thus to determine the type of incorrect sitting posture. For example, when adjusting from a correct sitting posture to a slightly forward-leaning posture, the readings of the left and right front sensors increase, but the increase is not significant enough to directly determine an incorrect sitting posture. At this point, it is necessary to combine the readings of the left and right rear sensors. During the re-detection and judgment process, both the left and right front sensors increase, and the difference in the definite integrals of the two sensors is small, insufficient to determine an incorrect sitting posture; however, comparing the definite integral differences between the left and left rear sensors, and the definite integral differences between the right and right front sensors, the differences are relatively large, sufficient to determine an incorrect sitting posture.

[0048] Figure 2 The diagram shows the readings of the left front sensor, right front sensor, left rear sensor, and right rear sensor used in this solution under different sitting postures. The following is a related illustration. Figures 2-10 The following example illustrates the implementation steps of this solution:

[0049] Step 1, Seating Detection: Calibrate each pressure sensor so that the pressure value is at its maximum when the user is not seated; this indicates the user is not seated. (See attached document) Figure 2 As shown, the maximum value of the left front sensor is between 6700 and 6800, the maximum value of the left rear sensor is between 7100 and 7200, the maximum value of the right front sensor is between 6800 and 6900, and the maximum value of the right rear sensor is between 7200 and 7300. Further reference is provided in the appendix. Figure 3 As shown, when the four pressure sensors are not at their maximum values, it is initially determined that the user has taken a seat, the initial pressure value when the user takes a seat is obtained, and the following step 2 is executed.

[0050] Step 2: Determine if posture detection is triggered:

[0051] Record whether the pressure sensor values ​​are stable (for each pressure sensor, record whether the change in value within 20ms is greater than 500 after every 50ms interval), record the average value of the pressure sensor values ​​for the next 20ms, the pressure value p, and the left anterior pressure value p. a The right anterior pressure value is p b The right rear pressure value is p c The left rear pressure value is p d The average value is taken as F0. (Time interval T1, n = 20ms)

[0052]

[0053] If the mean change F0 within 20ms is greater than 500, then proceed to step 3;

[0054] If the average change value F0 within 20ms is less than or equal to 500, it is judged as a normal sitting posture, and no sitting posture detection is performed. The current pressure value is then updated and replaced with the initial pressure value.

[0055] Step 3: Posture detection and judgment:

[0056] Step 3.1, Preliminary Detection and Judgment: Within the time period T2 (30ms), perform definite integral processing on the pressure value of each sensor to obtain the integral value. The pressure value is p, and the definite integral value is P. l0 (n=30ms), with the integral value of the left front sensor being P l01 For example

[0057]

[0058] When the error is determined to be error 1

[0059] a) Current integral value P l01 The difference P between the integral and the initial value l0 If the value is less than 15000 (the second threshold), proceed to step 3.2:

[0060] |P l01 -P l0 |<15000.

[0061] b) Current integral value P l01 and initial value integral P l0 A difference greater than 15000 (the second threshold) is considered an incorrect sitting posture.

[0062] |P l01 -P l0 15000.

[0063] Step 3.2, re-detection and judgment: Perform definite integral processing on the pressure values ​​of every two sensors for all sensors, obtain the definite integral difference between the two sensors, and compare this definite integral difference with the third threshold set for the correct sitting posture; specifically: the integral difference is P c0 (n = 30 ms), with the integral values ​​of the left anterior and right anterior sides being P. c01 For example.

[0064]

[0065] a) The integral values ​​for the left anterior and right anterior sides are P. c01 The integral difference P between the correct sitting posture setting and the correct sitting posture setting. c0 Less than 30,000 (third threshold):

[0066] |P c01 -P c0 |<30000.

[0067] b) The integral values ​​for the left anterior and right anterior sides are P. c01 The integral difference P between the correct sitting posture setting and the correct sitting posture setting. c0 A score greater than 30,000 (the third threshold) is considered an incorrect sitting posture.

[0068] |P c1 -P c0 30000.

[0069] In the above scheme, the threshold of the pressure sensor can be defined according to specific needs and usage scenarios. Generally, the threshold should be set above the pressure range generated by a normal sitting posture in order to detect pressure changes caused by poor posture. At the same time, the threshold should not be set too high, otherwise it may result in false alarms.

[0070] A normal sitting posture is judged as a bad sitting posture.

[0071] To determine a suitable threshold, the following steps can be taken:

[0072] 1. Have multiple people sit on the seat cushion in the correct sitting posture, and record the average and standard deviation of the pressure values ​​they generate.

[0073] 2. Determine a reasonable threshold based on this data. Typically, the threshold can be set as the mean plus one standard deviation or twice the standard deviation.

[0074] 3. In practical applications, if the false positive rate or false negative rate is found to be too high, the threshold can be adjusted according to the actual situation.

[0075] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0076] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention without departing from the principles and spirit of the present invention.

Claims

1. A sitting posture recognition method, characterized in that: Multiple pressure sensors distributed on the seating surface are used, and the following steps are performed: Step 1, Seating Detection: Detect the pressure value of each pressure sensor and compare it with its set calibration value to determine if the user has sat down; if seated, proceed to Step 2 to obtain the initial pressure value when the user sits down; if not seated, repeat Step 1. Step 2: Determine if posture detection is triggered: Based on the initial pressure value when the user sits down, periodically obtain the average pressure change of each sensor within the time period T1. If the average pressure change of all sensors does not exceed the first threshold, it is judged as a normal sitting posture, and posture detection is not performed. The current pressure value is updated and replaced with the initial pressure value, and Step 2 is repeated. If the average pressure change of any sensor exceeds the first threshold, Step 3 is executed. Step 3: Posture detection and judgment: Step 3.1, Preliminary detection and judgment: During the T2 time period, the pressure value of each sensor is processed by definite integration to obtain the integral value. The difference between the integral value and the integral value set for the correct sitting posture is compared with the second threshold. If it is greater than the second threshold, it is judged as an incorrect sitting posture, and an incorrect sitting posture reminder is issued. If the value is not greater than the second threshold, proceed to step 3.2; Step 3.2, re-detection and judgment: Perform definite integral processing on the pressure values ​​of every two sensors for all sensors, obtain the definite integral difference between the two sensors, and compare the definite integral difference with the third threshold set for the correct sitting posture; if it is greater than the third threshold, it is judged as an incorrect sitting posture, and an incorrect sitting posture reminder is issued; If the value is not greater than the third threshold, it is judged as a correct sitting posture; The sitting posture recognition method employs at least a left front sensor for acquiring pressure on the left thigh, a right front sensor for acquiring pressure on the right thigh, and a left rear sensor and a right rear sensor for acquiring pressure on both sides of the buttocks.

2. The sitting posture recognition method according to claim 1, characterized in that: The first threshold for each sensor represents the maximum pressure change of that sensor; the second threshold for each sensor represents the maximum integral change of that sensor; and the third threshold for every two sensors represents the maximum change in the integral difference between the two sensors.

3. The sitting posture recognition method according to claim 1, characterized in that: The left front sensor and the right front sensor are arranged side by side on the left and right sides of the front of the seat surface, and the left rear sensor and the right rear sensor are arranged side by side on the left and right sides of the middle of the seat surface; the distance d1 between the left front sensor and the right front sensor is greater than the distance d2 between the left rear sensor and the right rear sensor.

4. The sitting posture recognition method according to claim 1, characterized in that: The T2 time period is longer than the T1 time period.

Images