Pose abnormity level detection method and system and headphone

Abstract

The invention provides a pose abnormity grade detection method and system and a headset, and the method comprises the steps: obtaining a neck muscle physiological signal of a user, and determining a neck muscle abnormity coefficient corresponding to the neck muscle physiological signal; obtaining a pose signal of a user, and determining a pose abnormal coefficient corresponding to the pose signal; weighting the neck muscle abnormal coefficient and the pose abnormal coefficient to obtain a comprehensive abnormal coefficient; and determining the abnormal level of the user pose according to the comprehensive abnormal coefficient. According to the method, the neck muscle physiological signal of the user is obtained, the neck muscle abnormal coefficient corresponding to the neck muscle physiological signal and the posture abnormal coefficient corresponding to the posture signal are weighted, and then the posture abnormal level of the user is determined through the comprehensive abnormal coefficient obtained after the posture signal and the neck muscle physiological signal are combined. The abnormal condition of the pose of the user can be accurately detected.

Description

Technical Field

[0001] This invention relates to the field of smart headphone technology, and in particular to a method, system, and headphones for detecting abnormal posture levels. Background Technology

[0002] Currently, posture detection in headphones largely relies on inertial measurement units (IMUs). The core functions implemented using these IMUs are still concentrated on audio control, such as audio playback and head-nodding to switch tracks. While acquiring user pose signals through an IMU can detect whether the user's pose is abnormal and the specific abnormal situation, the detection method only considers the pose signal and cannot accurately detect abnormal user poses.

[0003] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0004] The main objective of this invention is to provide a method, system, and headset for detecting abnormal posture levels, aiming to solve the technical problem that existing technologies cannot accurately detect abnormal user posture.

[0005] To achieve the above objectives, this invention proposes a pose anomaly level detection method, applied to a pose anomaly level detection system, the pose anomaly level detection method comprising: The method includes: Acquire the user's neck muscle physiological signals and determine the neck muscle abnormality coefficient corresponding to the neck muscle physiological signals; Acquire the user's pose signal and determine the pose anomaly coefficient corresponding to the pose signal; The combined abnormality coefficient is obtained by weighting the abnormality coefficient of the neck muscles and the abnormality coefficient of the posture. The anomaly level of the user's pose is determined based on the comprehensive anomaly coefficient.

[0006] Optionally, the step of weighting the neck muscle abnormality coefficient and the posture abnormality coefficient to obtain a comprehensive abnormality coefficient includes: Get the scene type of the user's current scene; The neck muscle abnormality weight and the pose abnormality weight of the pose abnormality coefficient are assigned according to the scenario type. The comprehensive abnormality coefficient is obtained by weighting the neck muscle abnormality coefficient and the posture abnormality coefficient according to the neck muscle abnormality weight and the posture abnormality weight.

[0007] Optionally, the pose signal includes: head pose signal and / or trunk pose signal; the pose abnormality coefficient includes trunk abnormality coefficient and / or head abnormality coefficient. Determining the pose anomaly coefficient corresponding to the pose signal includes: Obtain the torso tilt angle within the torso pose signal; The torso anomaly coefficient of the torso pose signal is determined based on the preset tilt angle range in which the torso tilt angle is located. Obtain the head offset of the head position pose signal; The head anomaly coefficient of the head pose signal is determined based on the preset offset range in which the offset is located.

[0008] Optionally, the step of weighting the neck muscle abnormality coefficient and the posture abnormality coefficient to obtain a comprehensive abnormality coefficient includes: Get the scene type of the user's current scene; The neck muscle abnormality weight, the trunk abnormality weight, and the head abnormality weight are assigned according to the scenario type. The abnormal coefficients of the neck muscles, trunk, and head are weighted according to the abnormal weights of the neck muscles, trunk, and head to obtain a comprehensive abnormal coefficient.

[0009] Optionally, determining the neck muscle abnormality coefficient corresponding to the neck muscle physiological signal includes: Extract the feature values ​​reflecting muscle activity from the physiological signals of the neck muscles; The root mean square of the neck muscle physiological signal is calculated based on the eigenvalues. The neck muscle abnormality coefficient corresponding to the neck muscle physiological signal is determined based on the preset root mean square range in which the root mean square is located.

[0010] Optionally, after calculating the root mean square of the neck muscle physiological signal based on the feature values, the method further includes: If the root mean square of the neck muscle physiological signal is greater than a preset root mean square, a first adjustment prompt signal for neck muscle adjustment is output.

[0011] Optionally, after determining the anomaly level of the user pose based on the comprehensive anomaly coefficient, the method further includes: The priority of the abnormality alert is determined based on the neck muscle abnormality coefficient and the posture abnormality coefficient. Based on the aforementioned priority, a first adjustment prompt signal for neck muscle adjustment and a second adjustment prompt signal for posture signal adjustment are output.

[0012] Optionally, after outputting the first adjustment prompt signal for neck muscle adjustment and the second adjustment prompt signal for posture signal adjustment according to the priority, the method further includes: Acquire the neck muscle physiological signal after the first adjustment prompt signal is output and the posture signal after the second adjustment prompt signal is output; The user's posture adjustment result is verified based on the neck muscle physiological signal after the first adjustment prompt signal is output and the posture signal after the second adjustment prompt signal is output.

[0013] Optionally, after determining the anomaly level of the user pose based on the comprehensive anomaly coefficient, the method further includes: Obtain the duration of the abnormal state in which the user's pose is in an abnormal state; Acquire the first abnormal feature of the neck muscle physiological signal and the second abnormal feature of the pose signal within the abnormal duration; A health report on the user's posture is generated based on the abnormal duration, the neck muscle physiological signals, and the posture signals, and the health report is output to an external terminal.

[0014] In addition, to achieve the above objectives, the present invention also provides a posture abnormality level detection system for performing the posture abnormality level detection method described in any of the above claims; the posture abnormality level detection system includes: a data processing module, an electromyography sensing module, an image acquisition module, an inertial detection module, a communication module, and a prompting module; The data processing module is connected to the electromyography sensing module, the image acquisition module, the inertial detection module, the communication module, and the prompting module.

[0015] In addition, to achieve the above objectives, the present invention also provides a headset, including: a headband; The first and second earcups are connected to the headband; The first earmuff and / or the second earmuff are equipped with the aforementioned posture abnormality level detection system.

[0016] This invention provides a method, system, and headset for detecting posture abnormality levels. The method includes: acquiring a user's neck muscle physiological signals and determining a neck muscle abnormality coefficient corresponding to the neck muscle physiological signals; acquiring a user's posture signal and determining a posture abnormality coefficient corresponding to the posture signal; weighting the neck muscle abnormality coefficient and the posture abnormality coefficient to obtain a comprehensive abnormality coefficient; and determining the abnormality level of the user's posture based on the comprehensive abnormality coefficient. In this invention, by acquiring the user's neck muscle physiological signals and weighting the neck muscle abnormality coefficient corresponding to the neck muscle physiological signals with the posture abnormality coefficient corresponding to the posture signal, the comprehensive abnormality coefficient obtained by combining the posture signal and the neck muscle physiological signals is used to determine the abnormality level of the user's posture, thus accurately detecting abnormalities in the user's posture. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the first embodiment of the pose anomaly level detection method proposed in this invention; Figure 2 This is a schematic diagram of the first process of the second embodiment of the pose anomaly level detection method proposed in this invention; Figure 3 This is a schematic diagram of the second process of the second embodiment of the pose anomaly level detection method proposed in this invention; Figure 4 This is a schematic diagram of the first process of the third embodiment of the pose anomaly level detection method proposed in this invention; Figure 5 This is a schematic diagram of the second process of the third embodiment of the pose anomaly level detection method proposed in this invention; Figure 6 This is a schematic diagram of the first process of the fourth embodiment of the pose anomaly level detection method proposed in this invention; Figure 7 This is a schematic diagram of the second process of the fourth embodiment of the pose anomaly level detection method proposed in this invention; Figure 8 This is a schematic diagram of the posture anomaly level detection system proposed in this invention; Figure 9 This is a schematic diagram of the structure of the headphones proposed in this invention.

[0019] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0023] Furthermore, the use of terms such as "first" and "second" in this invention is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature. Additionally, the technical solutions of the various embodiments can be combined with each other, but only on the basis of being achievable by those skilled in the art. When the combination of technical solutions is contradictory or impossible to implement, such a combination of technical solutions should be considered non-existent and not within the scope of protection claimed by this invention.

[0024] Reference Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the pose anomaly level detection method proposed in this invention. Based on Figure 1 The first embodiment of a method for detecting abnormal pose levels is proposed.

[0025] In this embodiment, the pose anomaly level detection method includes: Step S10: Obtain the user's neck muscle physiological signal and determine the neck muscle abnormality coefficient corresponding to the neck muscle physiological signal.

[0026] It should be understood that, in the embodiments, the executing entity can be a pose anomaly level detection system, a data processing module within the pose anomaly level detection system, etc. In this embodiment and the following embodiments, the pose anomaly level detection method is described using a pose anomaly level detection system.

[0027] It should be noted that the neck muscle physiological signal is a signal used to represent the physiological state of the user's neck muscles. This neck muscle physiological signal can include signals indicating that the user's neck muscles are in a tense state, signals indicating that the user's neck muscles are in a relaxed state, and signals indicating the degree of tension or relaxation of the user's neck muscles. The neck muscle abnormality coefficient is the coefficient corresponding to the abnormal state exhibited by the user's current neck muscles. This neck muscle abnormality coefficient can have different values ​​depending on the state of the user's neck muscles. For example, if the user's neck is continuously highly tense, the coefficient value of the neck muscle abnormality coefficient will be large; similarly, if the user's neck muscles are in a relaxed state, the coefficient value of the neck muscle abnormality coefficient will be small or even zero.

[0028] In practical implementation, acquiring neck muscle physiological signals can be achieved by placing pressure sensors, current sensors, or other devices on the user's neck, or by placing these devices inside the earcups of headphones. The earcups then bring the pressure sensors and current sensors into contact with the user's neck, allowing for the collection of pressure readings or current values ​​passing through the neck, thereby determining the user's neck muscle physiological signals. For example, different states of tension or relaxation in the user's neck muscles will exert different forces on the pressure sensors placed on the neck, and the collected forces can then be used to determine the user's neck muscle physiological signals.

[0029] The abnormality coefficient of neck muscles can be determined based on the degree of tension in the neck muscles as indicated by physiological signals. For example, when the neck muscles are completely relaxed, the abnormality coefficient can be determined to be 0; when the neck muscles are completely tense, the abnormality coefficient can be determined to be 3.

[0030] Step S20: Obtain the user's pose signal and determine the pose anomaly coefficient corresponding to the pose signal.

[0031] It should be understood that pose signals are used to represent the position and posture of a user's head, shoulders, and other body parts. Examples include the position of the user's shoulders and the angle of head movement. The pose anomaly coefficient is a coefficient used to represent the degree of deviation between the user's pose and the standard pose. The greater the deviation between the user's pose and the standard pose, the larger the value of the pose anomaly coefficient; similarly, the smaller the deviation, the smaller the value of the pose anomaly coefficient. For example, when the user's head is tilted to the side, the larger the angle of deviation, the larger the corresponding value of the pose anomaly coefficient; the smaller the angle of deviation, the smaller the corresponding value of the pose anomaly coefficient.

[0032] In practice, user pose signals can be determined by acquiring user images and then identifying the acquired images, or by directly acquiring user pose signals using devices with pose measurement capabilities such as IMUs. Then, the pose anomaly coefficient corresponding to the pose signal is determined based on the offset between the acquired pose signal and the standard pose.

[0033] Step S30: Weight the neck muscle abnormality coefficient and the posture abnormality coefficient to obtain a comprehensive abnormality coefficient.

[0034] It should be understood that the comprehensive anomaly coefficient is a coefficient used to reflect the overall abnormal state of the user's posture. In this embodiment, the comprehensive anomaly coefficient includes the posture anomaly coefficient corresponding to the user's posture anomaly and the neck muscle anomaly coefficient corresponding to the user's neck muscle anomaly. For example, if the user's posture as presented by the posture signal is slightly abnormal, and the user's neck muscles are also slightly abnormal, combining the two can determine that the user's overall posture is moderately abnormal.

[0035] In practice, the comprehensive abnormality coefficient of user posture can be determined by weighting the neck muscle abnormality coefficient (which represents the user's neck muscle state) and the posture abnormality coefficient (which represents the user's posture abnormality).

[0036] Step S40: Determine the anomaly level of the user pose based on the comprehensive anomaly coefficient.

[0037] It should be understood that the anomaly level reflects the degree of anomaly in the user's overall pose, and this level can include mild, moderate, and severe anomalies. The specific classification of the anomaly level can be determined according to specific requirements; for example, if a more detailed understanding of the degree of anomaly is needed, the anomaly level can be divided into more levels. There is a certain mapping relationship between this anomaly level and the comprehensive anomaly coefficient. A higher comprehensive anomaly coefficient indicates a higher level of anomaly in the user's pose, and vice versa.

[0038] In practice, a mapping relationship between the comprehensive anomaly coefficient and the anomaly level can be established in advance, and then the anomaly level of the user's pose in the current state can be determined based on the calculated comprehensive anomaly coefficient and the mapping relationship.

[0039] This embodiment provides a method for detecting posture abnormality levels. The method includes: acquiring a user's neck muscle physiological signals and determining a neck muscle abnormality coefficient corresponding to the neck muscle physiological signals; acquiring a user's posture signal and determining a posture abnormality coefficient corresponding to the posture signal; weighting the neck muscle abnormality coefficient and the posture abnormality coefficient to obtain a comprehensive abnormality coefficient; and determining the abnormality level of the user's posture based on the comprehensive abnormality coefficient. In this embodiment, by acquiring the user's neck muscle physiological signals and weighting the neck muscle abnormality coefficient corresponding to the neck muscle physiological signals with the posture abnormality coefficient corresponding to the posture signals, and then using the comprehensive abnormality coefficient obtained by combining the posture signals and neck muscle physiological signals to determine the abnormality level of the user's posture, abnormalities in the user's posture can be accurately detected.

[0040] Based on the first embodiment of the pose anomaly level detection method described above, a second embodiment of the pose anomaly level detection method of the present invention is proposed. (Refer to...) Figure 2 , Figure 2 This is a schematic diagram of the first process of the second embodiment of the pose anomaly level detection method proposed in this invention.

[0041] In this embodiment, the step of determining the pose anomaly coefficient corresponding to the pose signal includes: It should be understood that the pose signal may include at least one of a head pose signal and a trunk pose signal; correspondingly, when the pose signal includes a head pose signal, the pose anomaly coefficient includes a trunk anomaly coefficient, and when the pose signal includes a trunk pose signal, the pose anomaly coefficient includes a head anomaly coefficient. The head pose signal may include information such as the head's pitch angle, yaw angle, and roll angle. The trunk pose signal includes information such as the shoulder horizontal angle and the trunk forward tilt angle.

[0042] The trunk anomaly coefficient is used to represent the abnormal deviation of the head posture. The head anomaly coefficient is used to represent the abnormal deviation of the trunk. When the deviation between the head posture and the standard head posture is large, the value of the trunk anomaly coefficient is large; when the deviation is small, the value of the trunk anomaly coefficient is small. Similarly, when the deviation between the trunk posture and the standard trunk posture is large, the value of the head anomaly coefficient is large; when the deviation is small, the value of the head anomaly coefficient is small.

[0043] During the acquisition of trunk pose signals, head pose signals, and neck muscle physiological signals, a sensor module (200Hz), a camera (30fps), and an IMU (100Hz) can be used to simultaneously acquire neck muscle physiological signals, trunk pose signals, and head pose signals, and the timestamps of neck muscle physiological signals, trunk pose signals, and head pose signals can be aligned through a synchronous precise time protocol.

[0044] In acquiring torso pose signals, images of the torso and face can be captured using a camera. A lightweight neural network model is then used to extract key points within the images, which are used to determine the torso pose signal. For example, the horizontal shoulder angle and forward tilt angle can be determined based on these key points. Furthermore, during the acquisition of head pose data using an IMU, Kalman filtering can be used to eliminate IMU drift, resulting in a more accurate head pose signal.

[0045] Step S201: Obtain the torso tilt angle within the torso pose signal.

[0046] It should be understood that the user's torso tilt angle is usually less than 15 degrees. If a torso tilt angle greater than 15 degrees is detected, it can be considered that there is an abnormality in the torso pose signal. In specific implementations, the torso pose signal includes the torso tilt angle. When acquiring the torso pose signal, the torso tilt angle can be directly extracted from the torso pose signal.

[0047] Step S202: Determine the trunk anomaly coefficient of the trunk pose signal based on the preset tilt angle range in which the trunk tilt angle is located.

[0048] It should be noted that the preset tilt angle range is a pre-defined range used to determine the degree of abnormality corresponding to the torso tilt angle. Multiple preset tilt angle ranges can be set; for example, 0 to 15 degrees can be set as the first preset tilt angle range, 15 to 20 degrees as the second, 20 to 30 degrees as the third, and greater than 30 degrees as the fourth. Different preset angle ranges can correspond to different torso abnormality coefficients. For example, if the torso tilt angle in the first preset tilt angle range is not abnormal, the torso abnormality coefficient is 0; if the torso tilt angle in the second preset tilt angle range is a low-degree abnormal angle, the torso abnormality coefficient can be 1; if the torso tilt angle in the third preset tilt angle range is a high-degree abnormal angle, the torso abnormality coefficient can be 2; and if the torso tilt angle in the fourth preset tilt angle range is a very high-degree abnormal angle, the torso abnormality coefficient can be 3.

[0049] Therefore, given a determined torso tilt angle, the torso anomaly coefficient corresponding to the torso pose signal can be directly determined based on the preset tilt angle range within which the torso tilt angle falls. Alternatively, in this embodiment, the torso anomaly coefficient of the torso pose signal can also be determined based on the shoulder horizontal angle, similarly using the preset horizontal angle range within which the shoulder horizontal angle falls. Furthermore, the torso tilt angle and shoulder horizontal angle can be combined to determine the torso anomaly coefficient. For example, one anomaly coefficient can be determined using the torso tilt angle, and another anomaly coefficient can be determined using the shoulder horizontal angle. Finally, the two coefficients are weighted to obtain the torso anomaly coefficient.

[0050] Step S203: Obtain the head offset of the head pose signal.

[0051] It should be noted that the nose offset includes the offset of the pitch angle, yaw angle, and roll angle. The pitch, yaw, and roll angles all have suitable ranges; for example, the pitch angle is normally between -10 and 10 degrees, and the yaw angle is normally between -30 and 30 degrees. For instance, with a pitch angle of 20 degrees, the pitch angle offset is 10 degrees; with a yaw angle of 50 degrees, the yaw angle offset is 20 degrees.

[0052] In the specific determination process, the corresponding offset can be determined based on the appropriate range of pitch angle, yaw angle and roll angle of normal users, as well as the pitch angle, yaw angle and roll angle of the collected head posture signal. Then, the offset of the head can be obtained by weighting each offset.

[0053] Step S204: Determine the head anomaly coefficient of the head pose signal based on the preset offset range in which the offset is located.

[0054] Understandably, the preset offset range is a pre-defined range used to determine the degree of abnormality corresponding to the head offset. This preset offset range can also be set to multiple ranges. For example, the first preset offset range could be a difference in angle between the offset and the standard offset of less than 5 degrees; the second preset offset range could be a difference in angle between the offset and the standard offset of 5 to 10 degrees; the third preset offset range could be a difference in angle between the offset and the standard offset of 10 to 20 degrees; and the fourth preset offset range could be a difference in angle between the offset and the standard offset of greater than 20 degrees.

[0055] Different preset offset ranges correspond to different head anomaly coefficients. For example, in the first preset offset range, the various pose angles of the user's head are not abnormal angles, so the head anomaly coefficient is 0; in the second preset offset range, the various pose angles of the user's head are low-degree abnormal angles, so the head anomaly coefficient is 1; in the third preset offset range, the various pose angles of the user's head are high-degree abnormal angles, so the head anomaly coefficient is 2; and in the fourth preset offset range, the various pose angles of the user's head are very high-degree abnormal angles, so the head anomaly coefficient is 3.

[0056] Therefore, given the head offset in the head pose signal, the head anomaly coefficient corresponding to the head pose signal can be determined directly based on the preset tilt offset range in which the head offset falls.

[0057] Understandably, when the pose signal includes head pose signal, using neck muscle physiological signals and head pose signal can improve the system's adaptability when the image acquisition module cannot acquire accurate images. When the pose signal includes body pose signal, the dual determination of body shape and neck muscle physiological signals improves the accuracy of pose anomaly recognition. Of course, when the pose signal includes both head pose signal and trunk pose signal, it can not only improve the system's adaptability but also further improve the accuracy of pose anomaly recognition.

[0058] In addition, refer to Figure 3 , Figure 3 This is a schematic diagram of the second process of the second embodiment of the pose anomaly level detection method proposed in this invention.

[0059] In this embodiment, step S30 includes: Step S301: Obtain the scene type of the user's current scene.

[0060] It should be understood that the parameters acquired in determining the comprehensive anomaly coefficient include neck muscle physiological signals, head pose signals, and trunk pose signals. Neck muscle physiological signals can be acquired using sensors, head pose signals can be acquired using an IMU, and trunk pose signals can be acquired using a camera. However, the data acquired by sensors, IMUs, and cameras vary depending on the scene. For example, in low-light conditions, the image captured by the camera to determine the trunk pose signal is not clear, resulting in an inaccurate trunk pose signal.

[0061] Therefore, before determining the comprehensive anomaly coefficient, it is necessary to determine the scene type of the user's current environment. This scene type can be categorized into static, dynamic, and low-light scenes. Specifically, the scene type can be determined by collecting the light intensity in the user's environment. For example, a standard light intensity that can accurately capture images can be set, and then the light intensity of the user's current environment can be collected. If the light intensity is less than the standard light intensity, the scene type can be identified as a low-light scene. For acquiring dynamic and static scene types, a standard change frequency of the pose signal can be set. By collecting the change frequency of the pose signal, if the change frequency is lower than the standard change frequency, the scene type can be identified as static; if the change frequency is not lower than the standard change frequency, the scene type can be identified as dynamic.

[0062] Step S302: Assign neck muscle abnormality weights to the neck muscle abnormality coefficients and pose abnormality weights to the pose abnormality coefficients according to the scene type.

[0063] It should be noted that the neck muscle abnormality weight refers to the proportion of neck muscle physiological signals in the calculation of the comprehensive abnormality coefficient under the current scene type; the pose abnormality weight refers to the proportion of pose signals in the calculation of the comprehensive abnormality coefficient under the current scene type. When the pose signals include head pose signals and trunk pose signals, the pose abnormality weights include trunk abnormality weights and head abnormality weights. The trunk abnormality weight refers to the proportion of trunk pose signals in the calculation of the comprehensive abnormality coefficient under the current scene type; the head abnormality weight refers to the proportion of head pose signals in the calculation of the comprehensive abnormality coefficient under the current scene type.

[0064] The specific values ​​of the weights for abnormal neck muscles and abnormal poses vary depending on the type of scenario. For example, in dynamic scenarios, the pose signal is constantly changing, so the specific value of the weight for abnormal poses is relatively large, while the accuracy of neck muscle physiological signal acquisition is low, so the specific value of the weight for abnormal neck muscles is relatively low.

[0065] In the specific allocation process, under static scene type, the weight of abnormal neck muscles can be set to 40%, which can effectively reflect the state of muscle fatigue; the weight of abnormal posture can be set to 60%, which can accurately measure the user's posture signal.

[0066] In dynamic scene types, the weight of pose abnormality is set to 80% to accurately capture the dynamic changes in user pose; the weight of neck muscle abnormality is set to 20% to exclude interference from normal muscle activity.

[0067] In low-light scenarios, the weight for abnormal neck muscles and abnormal pose is set to 50% to compensate for the decrease in the accuracy of pose signal acquisition under low-light conditions.

[0068] When the pose signal includes the trunk pose signal and the head pose signal, step S302 should be: assigning the neck muscle abnormality weight of the neck muscle abnormality coefficient, the trunk abnormality weight of the trunk abnormality coefficient, and the head abnormality weight of the head abnormality coefficient according to the scene type.

[0069] Similarly, the specific values ​​of the abnormal weights for neck muscles, trunk, and head vary depending on the type of scene. For example, in low-light conditions, the acquisition conditions for trunk pose signals are poor, so the specific value of the trunk abnormal weight is low; in dynamic scenes, the head pose signals are constantly changing, so the specific value of the head abnormal weight is large, while the accuracy of neck muscle physiological signal acquisition is low, so the specific value of the neck muscle abnormal weight is relatively low.

[0070] In the specific allocation process, under static scene type, the weight of neck muscle abnormality can be set to 40%, which can effectively reflect the state of muscle fatigue; the weight of trunk abnormality can be set to 40%, which can accurately measure trunk posture; and the weight of head abnormality can be set to 20%, which is used for auxiliary calibration.

[0071] In dynamic scene types, the weight of head abnormalities is set to 50% to accurately capture dynamic changes in head pose; the weight of torso abnormalities is set to 30%; and the weight of neck muscle abnormalities is set to 20% to exclude interference from normal muscle activity.

[0072] In low-light scenes, the weight of neck muscle abnormalities is set to 50%; the weight of head abnormalities is set to 30%; and the weight of torso abnormalities is set to 20% to compensate for the decrease in visual accuracy in low light.

[0073] Step S303: The abnormal coefficients of the neck muscles and the abnormal posture are weighted according to the abnormal weights of the neck muscles and the abnormal posture to obtain a comprehensive abnormal coefficient.

[0074] Understandably, given the neck muscle abnormality coefficient and its corresponding weight, as well as the pose abnormality coefficient and its corresponding weight, the comprehensive abnormality coefficient can be obtained by directly summing the products of each abnormality coefficient and its corresponding weight.

[0075] Furthermore, when the pose signal includes the trunk pose signal and the head pose signal, step S303 should be: weighting the neck muscle abnormality coefficient, the trunk abnormality coefficient, and the head abnormality coefficient according to the neck muscle abnormality weight, the trunk abnormality weight, and the head abnormality weight to obtain a comprehensive abnormality coefficient.

[0076] It should be understood that, given the abnormal coefficients of the neck muscles and their corresponding weights, the abnormal coefficients of the pose and their corresponding weights, and the abnormal coefficients of the torso and their corresponding weights of the head, the overall abnormal coefficient can be obtained by summing the products of each abnormal coefficient and its corresponding weight. For example, if the abnormal coefficient of the torso is 1, the abnormal coefficient of the neck muscles is 2, and the abnormal coefficient of the pose is 3, then in a static scene, the overall abnormal coefficient is 1. 0.4+2 0.4+3 0.2 = 1.8; In dynamic scene types, the comprehensive anomaly coefficient is: 1 0.3+2 0.2+3 0.5 = 2.2; In low-light scenes, the overall anomaly coefficient is 1. 0.2+2 0.5+3 0.3 = 2.1.

[0077] Furthermore, once the comprehensive anomaly coefficient is determined, the anomaly level of the user's pose can be directly determined based on the comprehensive anomaly coefficient. For example, if the comprehensive anomaly coefficient is less than 1, the anomaly level is identified as mild; if the comprehensive anomaly coefficient is between 1 and 2, the anomaly level is identified as moderate; and if the comprehensive anomaly coefficient is greater than 2, the anomaly level is identified as moderate.

[0078] Based on the first or second embodiment of the above-described pose anomaly level detection method, a third embodiment of the pose anomaly level detection method of the present invention is proposed. (Refer to...) Figure 4 , Figure 4 This is a schematic diagram of the first process of the third embodiment of the pose anomaly level detection method proposed in this invention.

[0079] In this embodiment, the step of determining the neck muscle abnormality coefficient corresponding to the neck muscle physiological signal includes: Step S101: Extract the feature values ​​reflecting muscle activity from the physiological signals of the neck muscles.

[0080] It should be noted that the feature values ​​are parameter values ​​used to reflect the characteristics of neck muscle activity. These feature values ​​can be integrals of neck muscle physiological signals, peak values ​​of neck muscle physiological signals, etc. In the specific acquisition process, considering that the root mean square (RMS) is a relatively accurate parameter reflecting neck muscle physiological signals, feature values ​​related to the RMS calculation can be selected within the neck muscle physiological signals based on the calculation of the RMS.

[0081] Furthermore, considering that neck muscle physiological signals can be acquired through sensors, the detected signals may contain interference signals, such as power frequency interference and noise interference. Therefore, noise reduction of the neck muscle physiological signals is necessary before extracting their feature values.

[0082] In practice, the neck muscle physiological signal can be filtered by a 50Hz notch filter to remove power frequency interference. Then, the effective part of the neck muscle physiological signal can be extracted using a bandpass filter from 20 to 500Hz. Finally, the effective part of the neck muscle physiological signal can be decomposed into 5 frequency ranges by wavelet transform to further reduce noise.

[0083] Step S102: Calculate the root mean square of the neck muscle physiological signal based on the feature value.

[0084] It should be noted that the root mean square (RMS) of the neck muscle physiological signal is a parameter that directly reflects the tension of the neck muscles. Under normal user posture, the RMS of the neck muscle physiological signal is typically less than or equal to 0.1 mV. If the RMS of the neck muscle physiological signal is detected to be greater than 0.1 mV, it can be concluded that the user's posture is abnormal. In practice, the RMS of the neck muscle physiological signal can be directly calculated using the RMS calculation formula based on the characteristic values ​​of the signal.

[0085] Step S103: Determine the neck muscle abnormality coefficient corresponding to the neck muscle physiological signal based on the preset root mean square range where the root mean square is located.

[0086] Understandably, the root mean square (RMS) can directly and quantitatively reflect the tension of neck muscles. A larger RMS indicates tighter neck muscles, resulting in a higher abnormality coefficient. The preset RMS range is a pre-defined range used to determine the abnormality coefficient of neck muscles based on their physiological signals. There are multiple preset RMS ranges, each corresponding to a specific abnormality coefficient. For example, when the root mean square (RMS) of the neck muscle physiological signal is less than 0.1 mV, the neck muscles are in a relaxed state, there is no abnormality, and the abnormality coefficient is zero. When the RMS of the neck muscle physiological signal is between 0.1 and 0.2 mV, the neck muscles are in a slightly tense state, the physiological signal corresponds to a mild abnormality, and the abnormality coefficient is 1. When the RMS of the neck muscle physiological signal is between 0.2 and 0.3 mV, the neck muscles are in a moderately tense state, the physiological signal corresponds to a moderate abnormality, and the abnormality coefficient is 2. When the RMS of the neck muscle physiological signal is greater than 0.3 mV, the neck muscles are in a tense state, the physiological signal corresponds to a severe abnormality, and the abnormality coefficient is 3.

[0087] Therefore, given the root mean square (RMS) of the physiological signals of the neck muscles, the preset RMS range within which the RMS is located can be directly determined, thereby obtaining the abnormal coefficient of the neck muscles corresponding to the preset RMS range.

[0088] In addition, refer to Figure 5 , Figure 5 This is a schematic diagram of the second process of the third embodiment of the pose anomaly level detection method proposed in this invention.

[0089] In this embodiment, after step S102, the method further includes: Step S104: When the root mean square of the neck muscle physiological signal is greater than the preset root mean square, output the first adjustment prompt signal for neck muscle adjustment.

[0090] It should be noted that the first adjustment prompt signal is used to remind the user that their neck muscles are too tense. This first adjustment prompt signal can be at least one of the following: high-frequency vibration + red indicator light + voice message "Muscles are severely tense, please adjust and relax immediately." The preset root mean square is a pre-set value used to determine if the user's neck muscles are too tense and prone to muscle strain. This preset root mean square can be set to 0.3mV.

[0091] It should be understood that in cases of severe abnormality with excessive neck muscle tension, this tension, if maintained for a period of time, can easily lead to muscle strain. Therefore, after obtaining the root mean square (RMS) of the neck muscle physiological signal, it is necessary to compare the RMS of the neck muscle physiological signal with a preset RMS. If the RMS of the neck muscle physiological signal is greater than the preset RMS, a first adjustment prompt signal is immediately output to prompt the user to relax the neck muscles. The audio volume is automatically reduced, and a white noise relaxation mode is activated. This method can promptly prompt or force the user to relax their neck muscles 5-10 seconds before the user perceives muscle strain.

[0092] Of course, if the root mean square of the neck muscle physiological signal is not greater than the preset root mean square, the step of determining the neck muscle abnormality coefficient corresponding to the neck muscle physiological signal based on the preset root mean square range in which the root mean square is located can continue.

[0093] A fourth embodiment of the pose anomaly level detection method of the present invention is proposed based on any one of the first to third embodiments of the above-described pose anomaly level detection method. (Refer to...) Figure 6 , Figure 6 This is a schematic diagram of the first process of the fourth embodiment of the pose anomaly level detection method proposed in this invention.

[0094] In this embodiment, after step S40, the method further includes: Step S50: Determine the priority of the abnormality prompt based on the abnormality coefficient of the neck muscles and the abnormality coefficient of the posture.

[0095] It should be understood that, once the level of abnormality in the user's pose is determined, it is also necessary to prompt the user so that the user can adjust their pose in a timely manner.

[0096] It should be noted that the priority of the abnormality prompts is between those prompting for neck muscle relaxation and those prompting for posture adjustment. When the degree of abnormality in the neck muscles is greater than the degree of abnormality in the posture signal, the abnormality prompt for neck muscle relaxation has a higher priority than that for posture adjustment; conversely, when the degree of abnormality in the neck muscles is less than the degree of abnormality in the posture signal, the abnormality prompt for neck muscle relaxation has a lower priority than that for posture adjustment.

[0097] In determining priorities, the priority of abnormal cues can be determined based on the coefficients of the neck muscle abnormality coefficient and the posture abnormality coefficient. For example, if the coefficient of the neck muscle abnormality coefficient is 2 and the coefficient of the posture abnormality coefficient is 1, then the priority of the abnormal cues for neck relaxation is higher than the priority of posture adjustment. Furthermore, the posture signal can be further subdivided into trunk posture signal and head posture signal. Then, the priority among the three cues—neck muscle relaxation, head posture adjustment, and trunk posture adjustment—can be determined based on the coefficients of the neck muscle abnormality coefficient, trunk abnormality coefficient, and head abnormality coefficient.

[0098] Step S60: Output a first adjustment prompt signal for neck muscle adjustment and a second adjustment prompt signal for posture signal adjustment according to the priority.

[0099] It should be noted that the first adjustment prompt signal is used to prompt the user to adjust neck muscles; the second adjustment prompt signal is used to prompt the user to adjust posture. The second adjustment prompt signal can be further divided into a third adjustment prompt signal for trunk posture adjustment and a fourth adjustment prompt signal for head posture adjustment. The prompt signal's corresponding prompt method differs depending on the level of abnormality. For example, in the case of mild abnormality, the adjustment prompt signal may be unilateral low-frequency vibration and a flashing green indicator light. In the case of moderate abnormality, the adjustment prompt signal may be bilateral vibration + a yellow indicator light + a voice prompt such as "Forward tilt, mild muscle tension, etc." In the case of severe abnormality, the adjustment prompt signal may be high-frequency vibration + a red indicator light + a voice prompt such as "Severe muscle tension, please adjust and relax immediately."

[0100] It should be understood that, given a defined priority, the first or second adjustment prompt signal can be output first, based on the priority order of the prompt signals. Of course, if the second adjustment prompt signal is further divided into a third and a fourth adjustment signal, the priority among the first, third, and fourth adjustment prompt signals can be determined first, and then the corresponding adjustment prompt signals can be output sequentially according to their priority order.

[0101] Step S70: Obtain the neck muscle physiological signal after the first adjustment prompt signal is output and the pose signal after the second adjustment prompt signal is output.

[0102] It should be understood that after the adjustment prompt signal is sent to the user, the user may not directly adjust the neck muscles or posture, or the adjustment time may be insufficient. Therefore, in this embodiment, a verification mechanism can also be set.

[0103] In practice, after outputting the first adjustment prompt signal and / or the second adjustment prompt signal, the neck muscle physiological signals after the first adjustment prompt signal and the pose signals after the second adjustment prompt signal can be re-acquired. The neck muscle physiological signals can be acquired using sensors, and the pose signals can be acquired using an image acquisition module.

[0104] Step S80: Verify the user's posture adjustment result based on the neck muscle physiological signal after the first adjustment prompt signal output and the posture signal after the second adjustment prompt signal output.

[0105] Understandably, after reacquiring the neck muscle physiological signal after the first adjustment prompt signal and the pose signal after the second adjustment prompt signal, the neck muscle abnormality coefficient of the neck muscle physiological signal can be verified based on the neck muscle physiological signal after the first adjustment prompt signal. If the neck muscle abnormality coefficient decreases, or if the root mean square of the neck muscle physiological signal decreases to below 0.1mV, the neck muscle adjustment process can be considered effective. Similarly, with the pose signal after the second adjustment prompt signal, the pose abnormality coefficient corresponding to the pose signal can be verified. If the pose abnormality coefficient decreases, the pose adjustment process can be considered effective. Of course, the image acquisition module can also directly capture the user's adjustment process. If the user adjusts according to the prompts, the pose and neck muscle adjustment processes can also be considered effective. For example, if the first adjustment prompt signal includes: "Please slowly rotate your neck 30 degrees to the left and right to relieve muscle tension," and the image acquisition module captures the user slowly rotating their neck 30 degrees to the left and right, the neck muscle adjustment process can be considered effective. For example, the second adjustment prompt signal output includes: "Please look up and make your facial key points parallel to the screen." If the image acquisition module captures the user looking up and the facial key points being parallel to the screen, then the pose adjustment process can be considered effective.

[0106] Further reference Figure 7 , Figure 7 This is a schematic diagram of the second process of the fourth embodiment of the pose anomaly level detection method proposed in this invention.

[0107] In this embodiment, after step S40, the method further includes: Step S90: Obtain the duration of the abnormal state in which the user's pose is abnormal.

[0108] It should be noted that abnormal states include: abnormal neck muscle states and abnormal posture states. The abnormal duration is the duration during which the neck muscle state and posture are in an abnormal state during the user's posture detection process.

[0109] During the acquisition process, the abnormal duration of the user's neck muscle physiological signals and the abnormal duration of the user's pose signals corresponding to the abnormal state can be recorded. The abnormal duration of the two states can be recorded separately or in combination.

[0110] Step S100: Obtain the first abnormal feature of the neck muscle physiological signal and the second abnormal feature of the pose signal within the abnormal duration.

[0111] It should be noted that the first abnormal feature is a feature reflecting an abnormal state of the neck muscles, which can be the root mean square of the physiological signals of the neck muscles. The second abnormal feature is a feature reflecting an abnormal state of posture, which can include abnormal trunk features and abnormal head features, such as trunk forward tilt angle and head pitch angle.

[0112] In practice, the neck muscle physiological signals and pose signals recorded throughout the entire detection process can be used to extract the neck muscle physiological signals and pose signals within the abnormal duration. Then, the first abnormal feature can be extracted from the neck muscle physiological signals within the abnormal duration, and the second abnormal feature can be extracted from the pose signals within the abnormal duration.

[0113] Step S110: Generate a health report of the user's pose based on the abnormal duration, the first abnormal feature, and the second abnormal feature, and output the health report to an external terminal.

[0114] Understandably, upon completion of the detection process, a health report can be generated based on the abnormal features of the user's posture throughout the entire detection process and the corresponding abnormalities. This health report displays the changes in the user's first and / or second abnormal features over time. The health report includes: peak muscle tension periods and analysis of the causes of abnormal posture, such as high neck muscle tension between 4:00 and 16:00, which may be related to looking down at a mobile phone, and pushes a first adjustment prompt signal, such as unilateral neck stretching for 30 seconds, repeated 3 times.

[0115] In the specific generation process, the changes of the first abnormal feature within the abnormal duration can be used to generate the first feature change situation, and the changes of the second abnormal feature within the abnormal duration can be used to generate the second feature change situation. The first feature change situation and the second feature change situation can be used to generate a health report of the user's pose.

[0116] The external terminal is a device that establishes a connection with the posture abnormality level detection system through a communication module. This external terminal can be a mobile phone, computer, or other device, and can intuitively display the user's posture health report.

[0117] Furthermore, to achieve the above objectives, the present invention also provides a pose anomaly level detection system, referring to... Figure 8 , Figure 8 This is a schematic diagram of the posture anomaly level detection system proposed in this invention.

[0118] The pose anomaly level detection system includes: The system includes a data processing module 10, an electromyography sensing module 20, an image acquisition module 30, an inertial detection module 40, a communication module 50, and a prompting module 60. The data processing module 10 is connected to the electromyography sensing module 20, the image acquisition module 30, the inertial detection module 40, the communication module 50, and the prompting module 60, respectively. The electromyography (EMG) sensing module 20 makes contact with the user's neck.

[0119] It should be noted that the data processing module 10 can adopt a main control chip with an integrated embedded neural network model. The computing power of the main control chip is ≥2 TOPS. The main control chip is equipped with multiple data receiving interfaces, supporting synchronous data reception across multiple interfaces. The electromyography sensing module 20 can be connected to the central control chip via an SPI interface, the image acquisition module 30 can be connected to the central control chip via a MIPI-CSI interface, and the inertial detection module 40 can be connected to the central control chip via I2C. The embedded neural network model is used for feature extraction of neck muscle physiological signals and the operation of multiple signal processing algorithms.

[0120] It should be understood that the electromyography (EMG) sensing module 20 may include: a miniature EMG sensing unit located inside the earcup, conforming to the upper edge of the trapezius muscle in the neck; the EMG sensing module 20 may be set to a sampling rate of 200Hz, a range of ±1mV, a resolution of 0.1μV, and a power consumption of ≤2mA. The EMG sensing module 20 may be encapsulated in medical-grade silicone to improve wearing comfort. The image acquisition module 30 may include: one or more miniature cameras located at the front of the headband or the upper part of the earcup, with the lenses tilted downwards at a certain angle to cover the user's face, shoulders, and upper body contours, for example, at 15 degrees. In low-light mode, it can be adapted to low-light office scenarios for acquiring body posture signals. The inertial detection module 40 may include: an IMU module integrating a three-axis accelerometer / gyroscope / magnetometer located inside the earcup, with a sampling rate of 100Hz, used to acquire head posture signals.

[0121] In practical implementation, the electromyography (EMG) sensing module 20, image acquisition module 30, and inertial detection module 40 can respectively acquire neck muscle physiological signals, body posture signals, and head posture signals. The data processing module 10 can timestamp-align the neck muscle physiological signals, body posture signals, and head posture signals, and then determine the abnormality level of the user's posture based on these signals. The prompting module 60 controls the built-in vibration motor, RGB indicator light, and speaker in the earcups to provide adjustment prompts. The data processing module 10 can also generate a health report and transmit it to an external terminal via the communication module 50. The posture abnormality level detection system also includes an audio device connected to the data processing module 10 for outputting voice prompts.

[0122] In this embodiment, the pose anomaly level detection system is used to execute the pose anomaly level detection method in any of the above embodiments. The specific implementation steps can be referred to the above embodiments of the pose anomaly level detection method, which will not be repeated here.

[0123] Furthermore, to achieve the above objectives, the present invention also provides a headset, referring to... Figure 9 , Figure 9 This is a schematic diagram of the structure of the headphones proposed in this invention.

[0124] The headset includes: a headband 70; The first earcup 80 and the second earcup 90 are connected to the headband 70; The posture abnormality level detection system is provided inside the first earmuff 80 and / or the second earmuff 90; The electromyography (EMG) sensor module in the posture abnormality level detection system fits against the user's neck through the first earmuff 80 and / or the second earmuff 90.

[0125] In this embodiment, the headphone or the posture abnormality level detection system within the headphone is used to execute the posture abnormality level detection method in any of the above embodiments. The specific implementation steps can be referred to the above embodiments of the posture abnormality level detection method, which will not be repeated here.

[0126] The above are merely preferred embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for detecting the level of pose abnormality, characterized in that, Applied to posture anomaly level detection systems; The method includes: Acquire the user's neck muscle physiological signals and determine the neck muscle abnormality coefficient corresponding to the neck muscle physiological signals; Acquire the user's pose signal and determine the pose anomaly coefficient corresponding to the pose signal; The combined abnormality coefficient is obtained by weighting the neck muscle abnormality coefficient and the posture abnormality coefficient. The anomaly level of the user's pose is determined based on the comprehensive anomaly coefficient.

2. The method for detecting pose abnormality level as described in claim 1, characterized in that, The step of weighting the neck muscle abnormality coefficient and the posture abnormality coefficient to obtain a comprehensive abnormality coefficient includes: Get the scene type of the user's current scene; The neck muscle abnormality weight and the pose abnormality weight of the pose abnormality coefficient are assigned according to the scenario type. The comprehensive abnormality coefficient is obtained by weighting the neck muscle abnormality coefficient and the posture abnormality coefficient according to the neck muscle abnormality weight and the posture abnormality weight.

3. The method for detecting pose abnormality level as described in claim 2, characterized in that, The pose signal includes: head pose signal and / or trunk pose signal; the pose abnormality coefficient includes trunk abnormality coefficient and / or head abnormality coefficient. Determining the pose anomaly coefficient corresponding to the pose signal includes: Obtain the torso tilt angle within the torso pose signal; The torso anomaly coefficient of the torso pose signal is determined based on the preset tilt angle range in which the torso tilt angle is located. Obtain the head offset of the head position pose signal; The head anomaly coefficient of the head pose signal is determined based on the preset offset range in which the offset is located.

4. The method for detecting pose abnormality level as described in claim 1, characterized in that, Determining the neck muscle abnormality coefficient corresponding to the neck muscle physiological signal includes: Extract the feature values ​​reflecting muscle activity from the physiological signals of the neck muscles; The root mean square of the neck muscle physiological signal is calculated based on the eigenvalues. The neck muscle abnormality coefficient corresponding to the neck muscle physiological signal is determined based on the preset root mean square range in which the root mean square is located.

5. The method for detecting pose abnormality level as described in claim 4, characterized in that, After calculating the root mean square of the neck muscle physiological signal based on the feature value, the method further includes: If the root mean square of the neck muscle physiological signal is greater than a preset root mean square, a first adjustment prompt signal for neck muscle adjustment is output.

6. The method for detecting pose abnormality level as described in claim 1, characterized in that, After determining the anomaly level of the user pose based on the comprehensive anomaly coefficient, the method further includes: The priority of the abnormality alert is determined based on the neck muscle abnormality coefficient and the posture abnormality coefficient. Based on the aforementioned priority, a first adjustment prompt signal for neck muscle adjustment and a second adjustment prompt signal for posture signal adjustment are output.

7. The method for detecting pose abnormality level as described in claim 6, characterized in that, After outputting the first adjustment prompt signal for neck muscle adjustment and the second adjustment prompt signal for posture signal adjustment according to the priority, the method further includes: Acquire the neck muscle physiological signal after the first adjustment prompt signal is output and the posture signal after the second adjustment prompt signal is output; The user's posture adjustment result is verified based on the neck muscle physiological signal after the first adjustment prompt signal is output and the posture signal after the second adjustment prompt signal is output.

8. The method for detecting pose abnormality level as described in claim 1, characterized in that, After determining the anomaly level of the user pose based on the comprehensive anomaly coefficient, the method further includes: Obtain the duration of the abnormal state in which the user's pose is in an abnormal state; Acquire the first abnormal feature of the neck muscle physiological signal and the second abnormal feature of the pose signal within the abnormal duration; A health report on the user's posture is generated based on the abnormal duration, the neck muscle physiological signals, and the posture signals, and the health report is output to an external terminal.

9. A posture anomaly level detection system, characterized in that, Used to perform the pose anomaly level detection method according to any one of claims 1 to 8; The posture abnormality level detection system includes: a data processing module, an electromyography sensing module, an image acquisition module, an inertial detection module, a communication module, and a prompting module; The data processing module is connected to the electromyography sensing module, the image acquisition module, the inertial detection module, the communication module, and the prompting module, respectively. The electromyography (EMG) sensor module makes contact with the user's neck.

10. A type of headset, characterized in that, include: Head beam; The first and second earcups are connected to the headband; The first earmuff and / or the second earmuff are provided with the posture abnormality level detection system as described in claim 9; The electromyography (EMG) sensor module in the posture abnormality level detection system fits against the user's neck through the first earmuff and / or the second earmuff.

Images