A method and device for detecting head movement

Abstract

The present invention relates to the technical field of motion detection, and specifically discloses a head motion detection method and device. The method uses a three-axis acceleration sensor to measure the pitch angle, roll angle, and heading angle during the head motion process to calculate the current attitude value, and further calculates the trend value according to the change of the attitude value, so as to combine the attitude value and the trend value to detect the head motion situation, which is easy to intuitively obtain the number of head motions and the amplitude of each motion. A large number of experiments have verified the accuracy of this method. The device is designed with an outer sponge, an inner sponge adapted to the size of the head, and a circuit module embedded between the outer sponge and the inner sponge. The circuit module is used to implement the head motion detection method to detect the head motion. The head motion detection device has a simple structure and a compact installation. Both the outer sponge and the inner sponge are flexibly designed, bringing a comfortable wearing experience to the user and having almost no interference to the test subject.

Description

Technical Field

[0001] The present invention relates to the technical field of motion detection, and particularly to a head motion detection method and device. Background Art

[0002] Sleep is crucial for people. It is not only an important link for the human body to recover, integrate and consolidate memories, but also directly related to aspects such as human growth and development, energy accumulation, and immune function maintenance. Surveys by the World Health Organization (WHO) show that 27% of people globally have varying degrees of sleep problems. This data reveals that sleep-related diseases have seriously threatened the health of people worldwide. Effectively monitoring people's sleep in daily life, enabling them to understand their own sleep problems and thus timely improve their sleep conditions, is of great importance.

[0003] Sleep monitoring systems can generally be divided into two categories. One is non-contact, such as a monitoring system based on microwave radar, and the other is contact, such as a polysomnography (PSG) monitoring system. For the system based on microwave radar, non-intrusive monitoring is achieved. However, due to interference from items such as clothes and quilts to the collected signals and the fact that the collected signals are not comprehensive enough, the accuracy is not high. The PSG method is the "gold standard" for diagnosing sleep disorder diseases. Generally, it requires the subject to go to a hospital with relevant qualifications for monitoring of multiple whole-night sleep activities. Moreover, there are many electrodes and the connections are complex, which may bring psychological burden and physical discomfort to the subject, making it difficult for the subject to fall asleep instead.

[0004] Among many sleep monitoring systems, systems with good experience and not too low accuracy, such as those based on heart rate variability systems, are often favored by users. Such systems usually use sensors to detect body movements to assist in monitoring, making the definition of sleep and wakefulness more accurate. According to the different sensors used, the related technical status is introduced from three aspects: monitoring based on microwave technology, flexible force-sensitive sensors, and accelerometers.

[0005] Monitoring based on microwave technology utilizes the microwave Doppler principle. A radio frequency bio-radar emits a continuous radio frequency signal, which is reflected by the target and then received by the receiver. The signal with a larger amplitude separated from the echo signal is the body movement. In specific applications, the separated body movement signal is preprocessed through wavelet transform, and the energy value and extreme value of the body movement signal are calculated, and the number of body movements is extracted accordingly. This type of technology achieves non-intrusive monitoring of the subject and does not affect their normal sleep, but the collected signals are greatly interfered with and the accuracy is not high.

[0006] Based on the monitoring of flexible force sensors, the main carrier is the mattress, which is divided into capacitive and resistive types. It converts the change in pressure signal into the change in capacitance or resistance, and then converts it into electrical quantity (voltage, current, etc.) through a circuit for output. Before and after each body movement, the pressure dot matrix image will change accordingly. After image denoising processing, the frame difference method or the image gray center of gravity method is used to analyze the amplitude and frequency characteristics of body movement, and it is classified into large body movement, medium body movement, small body movement and stillness according to these characteristics, and then quantified. The advantages of this type of technology are similar to those of microwave technology-based monitoring, with almost no interference to the subjects. The disadvantage is that when the subject turns over frequently, it may be out of the monitoring range of the pressure dot matrix.

[0007] Based on the monitoring of accelerometers, generally the three-axis acceleration in the MPU6050 sensor is used, and the square sum is taken and then square-rooted to combine into a one-dimensional acceleration value. There are three commonly used body movement measurement methods. One is the time threshold method, which records the time length when the combined acceleration is greater than the reference value within a unit time; the second is the digital integration method, which records the integral value of the combined acceleration with respect to time within a unit time, that is, the area under the combined acceleration curve; the third is the zero-crossing method, which records the number of times the combined acceleration is greater than the reference value (a value slightly larger than zero) within a unit time. The advantage of this type of technology is that it is suitable for wearable applications and has a higher accuracy than the previous two. However, it only directly reflects the number and amplitude of body movements with acceleration, and the detection result is still not precise enough. Summary of the Invention

[0008] The present invention provides a head movement detection method and device, and the technical problem to be solved is that: in the existing body movement detection method based on accelerometers, only the number and amplitude of body movements are directly reflected by acceleration, and the detection result is still not precise enough.

[0009] To solve the above technical problems, the present invention provides a head movement detection method, including the steps of:

[0010] S1. Obtain the pitch angle, roll angle and heading angle measured by a three-axis accelerometer fixed to the head at a fixed frequency;

[0011] S2. Calculate the current attitude value according to the pitch angle, roll angle and heading angle at the current sampling moment;

[0012] S3. Determine the current trend value according to the difference between the current attitude value and the previous attitude value;

[0013] S4. If the corresponding movement trend of the current trend value remains unchanged, increment the number of consecutive unchanged trends by 1, otherwise reset the number of consecutive unchanged trends to zero;

[0014] S5. Determine whether the number of consecutive unchanged trends exceeds the consecutive unchanged trend threshold. If so, go to step S9, otherwise proceed to the next step;

[0015] S6. Calculate the absolute difference between the current trend value and the previous trend value. If the absolute difference is greater than 0, record the absolute difference between the previous attitude value and the last attitude value of the previous movement as the amplitude of this movement and go to step S9; otherwise, proceed to the next step.

[0016] S7. Calculate the absolute difference between the current attitude value and the last attitude value of the previous movement. If the absolute difference exceeds the attitude absolute difference threshold, proceed to the next step; otherwise, go to step S10.

[0017] S8. Increment the number of head movements by 1 and go to step S10.

[0018] S9. Assign the previous attitude value to the last attitude value of the previous movement, assign 0 to the number of consecutive unchanged trends, assign 0 to the absolute difference between the current attitude value and the last attitude value of the previous movement, and proceed to the next step.

[0019] S10. Assign the current attitude value to the previous attitude value, assign the current trend value to the previous trend value, and return to step S2.

[0020] Further, in step S2, the current attitude value is equal to the integer part of the square root of the sum of the squares of the pitch angle, roll angle, and yaw angle.

[0021] Further, in step S3, if the difference between the current attitude value and the previous attitude value is greater than 0, the current trend value trend_cur is the first preset value; if the difference between the current attitude value and the previous attitude value is less than 0, the current trend value trend_cur is the second preset value; if the difference between the current attitude value and the previous attitude value is equal to 0, the current trend value trend_cur is the third preset value; the first preset value, the second preset value, and the third preset value are not equal to each other.

[0022] Further, the first preset value is 1, the second preset value is 2, and the third preset value is 0.

[0023] A head movement detection method provided by the present invention uses a three-axis acceleration sensor to measure the pitch angle, roll angle, and yaw angle during the head movement to calculate the current attitude value, and further calculates the trend value based on the change of the attitude value, so as to combine the attitude value and the trend value to detect the head movement situation, which is easy to intuitively obtain the number of head movements and the amplitude of each movement. A large number of experiments have verified the accuracy of this method. In addition, this method has low requirements for computing power and is easy to apply in wearable devices.

[0024] The present invention also provides a head movement detection device, which is characterized in that it includes an outer sponge, an inner sponge adapted to the size and shape of the head, and a circuit module embedded between the outer sponge and the inner sponge. The circuit module is provided with a main board, and the main board is used to implement the head movement detection method described above.

[0025] Specifically, the circuit module further includes a battery module, a display and button module, and a magnetic charging port electrically connected to the main board, wherein the display and button module and the magnetic charging port are openably embedded on the surface of the outer sponge.

[0026] Specifically, the outer sponge and the inner sponge are set as flexible structures and can be bent inwardly into an annular structure adapted to the size of the head.

[0027] Specifically, the outer sponge and the inner sponge are provided with mutually cooperating magic tapes at their tails.

[0028] Specifically, the battery module includes a battery, a top frame shell, and a bottom frame shell, and the battery is fixed in the top frame shell and the bottom frame shell.

[0029] Specifically, the head movement detection device further includes an upper frame shell, a middle frame shell, and a lower frame shell, and the main board and the display and button module are fixed in the upper frame shell, the middle frame shell, and the lower frame shell.

[0030] The head movement detection device provided by the present invention is designed with an outer sponge, an inner sponge adapted to the size of the head, and a circuit module embedded between the outer sponge and the inner sponge. The circuit module is used to implement the above head movement detection method to detect the head movement, and display the detected number of movements, the amplitude of each movement, and the remaining power on the display screen. The outer sponge and the inner sponge are used to adapt to the size and shape of the head, facilitating wearing on the user's head to complete the head movement detection. The head movement detection device has a simple structure and a compact installation. The outer sponge and the inner sponge are both flexibly designed, which not only brings a comfortable wearing experience to the user but also has a light weight, and can be stably worn on the user's head. No matter how the user moves, the head movement of the user can be detected, and there is almost no interference to the subject. Moreover, the detection method implemented by the circuit module ensures a high detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 is a flowchart of a head movement detection method provided by an embodiment of the present invention;

[0032] Figure 2 is a diagram showing 9 change trends of attitude values provided by an embodiment of the present invention;

[0033] Figure 3It is a schematic diagram of missed detection of motion provided by an embodiment of the present invention;

[0034] Figure 4 It is a graph of experimental results provided by an embodiment of the present invention;

[0035] Figure 5 It is a structural diagram of a head motion detection device provided by an embodiment of the present invention.

[0036] Reference numerals: 1 - outer sponge, 2 - inner sponge, 3 - main board, 4 - battery module, 41 - battery, 42 - top layer frame shell, 43 - bottom layer frame shell, 5 - display and button module, 6 - magnetic charging port, 71 - upper layer frame shell, 72 - middle layer frame shell, 73 - lower layer frame shell. Detailed implementation manners

[0037] The following specifically clarifies the implementation manners of the present invention in conjunction with the accompanying drawings. The given embodiments are only for illustrative purposes and should not be construed as limiting the present invention. The accompanying drawings are only for reference and illustration, and do not constitute a limitation on the protection scope of the present invention's patent, because many changes can be made to the present invention without departing from its spirit and scope.

[0038] Example 1

[0039] A head motion detection method provided by an embodiment of the present invention, as Figure 1 shown in the flowchart, includes the steps:

[0040] S1. Obtain the pitch angle (Pitch), roll angle (Roll), and yaw angle (Yaw) measured by a triaxial accelerometer fixed to the head at a fixed frequency;

[0041] S2. Calculate the current pose value (pose_cur) based on the pitch angle (Pitch), roll angle (Roll), and yaw angle (Yaw) at the current sampling moment;

[0042] S3. Determine the current trend value (trend_cur) based on the difference between the current pose value (pose_cur) and the previous pose value (pose_pre);

[0043] S4. If the corresponding motion trend of the current trend value (trend_cur) remains unchanged, increment the number of consecutive unchanged trends (trend_unchange) by 1; otherwise, reset the number of consecutive unchanged trends (trend_unchange) to zero;

[0044] S5. Determine whether the number of consecutive unchanged trends (trend_unchange) exceeds the consecutive unchanged trend threshold (TREND_THRESHOLD). If so, go to step S9; otherwise, proceed to the next step;

[0045] S6. Calculate the absolute difference (diff_trend) between the current trend value (trend_cur) and the previous trend value (trend_pre). If the absolute difference (diff_trend) is greater than 0, then record the absolute difference (|pose_pre - pose_last|) between the previous pose value (pose_pre) and the last pose value of the previous movement as the amplitude of this movement and go to step S9; otherwise, go to the next step.

[0046] S7. Calculate the absolute difference (diff_pose) between the current pose value (pose_cur) and the last pose value of the previous movement (pose_last). If the absolute difference (diff_pose) exceeds the pose absolute difference threshold (ACTION_THRESHOLD), then go to the next step; otherwise, go to step S10.

[0047] S8. Increment the number of head movements (total_action) by 1 and go to step S10.

[0048] S9. Assign the previous pose value (pose_pre) to the last pose value of the previous movement (pose_last), assign 0 to the number of consecutive unchanged trends (trend_unchange), assign 0 to the absolute difference (diff_pose) between the current pose value and the last pose value of the previous movement, and go to the next step.

[0049] S10. Assign the current pose value (pose_cur) to the previous pose value (pose_pre), assign the current trend value (trend_cur) to the previous trend value (trend_pre), and return to step S2.

[0050] The following provides a more detailed description of each step.

[0051] Before step S1, it also includes the steps:

[0052] S0. Initialization: Assign the pose value of the first calculation (pose[0]) to the last pose value of the previous movement (pose_last) and the previous pose value (pose_last), assign the previous trend value to 0, assign the number of consecutive unchanged trends (trend_unchange) to 0, assign the number of head movements (total_action) to 0, assign the consecutive unchanged trend threshold (TREND_THRESHOLD) to a, and assign the pose absolute difference threshold (ACTION_THRESHOLD) to b. In this example, a = 3 and b = 30 sampling periods are taken, which are determined based on the experimental results of multiple people and multiple times to maximize the detection accuracy.

[0053] In step S1, the three-axis accelerometer used in this example is the sensor MPU6050. In other embodiments, other three-axis accelerometers can be used, such as LIS3DH.

[0054] In step S2, the current pose value (pose_cur) is equal to the integer part of the square root of the sum of the squares of the pitch angle (Pitch), roll angle (Roll), and yaw angle (Yaw), that is:

[0055]

[0056] where, [] represents taking the integer part. The advantages of taking the integer part are, firstly, to filter out the burrs to make subsequent detection easier, and secondly, the zero drift problem of the sensor can be ignored. It is reasonable to take the integer part in this way because the head movement detection considers a process. The decimal part ignored currently will be compensated to a certain extent at the next moment, so it will not affect the detection.

[0057] Each moment can be used as the current moment, and the pose value corresponding to each moment is calculated using formula (1). The pose value of the previous moment corresponding to the current pose value (pose_cur) is the previous pose value (pose_pre).

[0058] For the following steps, in order to obtain an effective detection method, it is necessary to analyze the change trend of Pose. As Figure 2 , before and after time T, the change of Pose can be abstracted into 9 cases, divided into 3 categories. The first category corresponds to Figure 2 (a1), (a2), and (a3) in, and the trend before time T is upward; the second category corresponds to Figure 2 (b1), (b2), and (b3) in, and the trend before time T is downward; the third category corresponds to Figure 2Among (c1), (c2), and (c3), the trend is unchanged before time T. Through analysis, the following ideas can be easily obtained: First, when the trend does not change before and after time T, no new movement occurs, as shown in (a1), (b1), and (c1); Second, when there is an upward or downward trend after time T, new movement may occur, but the subsequent change amplitude needs to be further observed, as shown in (a2), (b2), (c2), and (c3); Third, when the trend remains unchanged after time T, the previous possible movement ends, as shown in (a3) and (b3).

[0059] In step S3, based on the change trend of the aforementioned Pose, in this example, the current trend value (trend_cur) is obtained according to the difference (diff) between the current pose value (pose_cur) and the previous pose value (pose_pre).

[0060] First, calculate the difference (diff) between the current pose value (pose_cur) and the previous pose value (pose_pre):

[0061] diff = pose_cur - pose_pre (2)

[0062] If diff > 0, the current trend value trend_cur = 1 (the first preset value); if diff < 0, the current trend value trend_cur = 2 (the second preset value); if diff = 0, then the current trend value trend_cur = 0 (the third preset value). The values 1, 2, and 0 of the current trend value trend_cur represent upward, downward, and unchanged trends respectively. In other embodiments, the first preset value, the second preset value, and the third preset value can take other combinations of unequal values.

[0063] In step S4, when the trend corresponding to the current trend value trend_cur is upward or downward (i.e., trend_cur is not equal to 0), the number of consecutive unchanged trends trend_unchange is cleared; when the trend is unchanged (trend_cur is equal to 0), the number of consecutive unchanged trends trend_unchange is incremented by 1.

[0064] In step S5, it is judged whether the number of consecutive unchanged trends trend_unchange exceeds the threshold continuous unchanged trend threshold TREND_THRESHOLD. If it exceeds, go to step S9; otherwise, proceed to the next step. The purpose of this step is to solve the problem of missed movement detection as Figure 3 shown. As Figure 3In (a) of [the figure], the trend is upward before time T1, remains unchanged between T1 and T2, and continues to be upward after T2. Since the trend remains unchanged between T1 and T2, that is, it is still considered an upward trend, the new movement that may occur after T2 is considered a continuation of the movement before T1; when the interval between T1 and T2 is long enough, the movement after T2 can be considered new and not just a continuation of the previous movement, and then missed detection will occur. Figure 3 Similar problems may occur in (b) of [the figure].

[0065] In step S6, first calculate the absolute difference (diff_trend) between the current trend value (trend_cur) and the previous trend value (trend_pre):

[0066] diff_trend = |trend_cur - trend_pre| (3)

[0067] If diff_trend > 0, it indicates that the trend has changed (if the number of head movements has been counted after the previous trend change, then record |pose_pre - pose_last| as the amplitude of this movement), and go to step S9; otherwise, go to the next step.

[0068] In step S7, first calculate the absolute difference (diff_pose) between the current pose value (pose_cur) and the last pose value (pose_last) of the previous movement:

[0069] diff_pose = |pose_cur - pose_last| (4)

[0070] If diff_pose exceeds the pose absolute difference threshold ACTION_THRESHOLD, go to the next step; otherwise, go to step S10.

[0071] In step S9, reset some data and parameters, assign the previous pose value (pose_pre) to the last pose value (pose_last) of the previous movement, assign the number of consecutive unchanged trends (trend_unchange) to 0, and assign the absolute difference (diff_pose) between the current pose value and the last pose value of the previous movement to 0, that is:

[0072] pose_last = pose_pre, trend_unchange = 0, diff_pose = 0 (5)

[0073] In step S10, update the previous value and the previous trend. Assign the current pose value (pose_cur) to the previous pose value (pose_pre), and assign the current trend value (trend_cur) to the previous trend value (trend_pre), that is:

[0074] pose_pre = pose_cur, trend_pre = trend_cur (6)

[0075] Then go to step S2 to enter the next round of detection.

[0076] Taking 30s as a detection period, a large number of experiments were carried out using the above method. Figure 4 Listed are partial experimental results, including the detection results of 9 detection periods from (a) to (i). It can be seen that according to the Pose value, this method can detect the number of head movements in various head movement situations, which is consistent with the actual visual observation, and the amplitude size recorded for each movement coincides with the actual situation. Figure 4

[0077] A head movement detection method provided by an embodiment of the present invention uses a three-axis acceleration sensor to measure the pitch angle, roll angle, and heading angle during the head movement process to calculate the current pose value, and further calculates the trend value based on the change of the pose value, so as to combine the pose value and the trend value to detect the head movement situation, which is easy to intuitively obtain the number of head movements and the amplitude of each movement. A large number of experiments have verified the accuracy of this method. In addition, this method has low requirements for computing power and is easy to be applied in wearable devices.

[0078] Example 2

[0079] To apply the above head movement detection method, the present invention provides a head movement detection device, the structure of which is as Figure 5 shown, including an outer sponge 1, an inner sponge 2 adapted to the size and shape of the head, and a circuit module embedded between the outer sponge 1 and the inner sponge 2. The circuit module is provided with a main board 3, and the main board 3 is used to implement the head movement detection method in Embodiment 1.

[0080] Specifically, the outer sponge 1 and the inner sponge 2 are set as flexible structures, which can be bent inward to form an annular structure adapted to the size of the head, facilitating wearing on the head. The ends of the outer sponge 1 and the inner sponge 2 are provided with mutually cooperating magic tapes.

[0081] Specifically, as Figure 5 shown, the circuit module further includes a battery module, a display and button module 5, and a magnetic charging port 6 electrically connected to the main board 3, wherein the display and button module 5 and the magnetic charging port 6 are openably embedded on the surface of the outer sponge 1, facilitating the user to view the display screen, operate the buttons, and charge.​

[0082] Specifically, the battery module includes a battery 41, a top frame case 42 and a bottom frame case 43 for fixing and protecting the battery 41.

[0083] Specifically, in order to fix and protect the main board 3 and the display and button module 5, the head motion detection device further includes an upper frame case 71, a middle frame case 72 and a lower frame case 73.

[0084] Specifically, the main board 3 is located inside the right side of the inner sponge 2. The display and button module 5 is electrically connected to the main board 3 and is located outside the outer sponge 1. The battery 41 is located inside the left side of the inner sponge 2. The magnetic charging port 6 is located on the right side of the outer sponge 1. The inner sponge 2 and the outer sponge 1 are edge-sealed by a wrapping cloth.

[0085] The specific installation steps of a head motion detection device provided in this embodiment are as follows:

[0086] 1) Fix the main board 3 in the lower frame case 73 with screws and apply a little environment-friendly glue to fix it to the inner sponge 2;

[0087] 2) Snap the display and button module 5 into the middle frame case 72 and reinforce it on the lower frame case 73 through positioning posts;

[0088] 3) Install the upper frame case 71 on the middle frame case 72;

[0089] 4) Fix the battery 41 in the top frame case 42 and the bottom frame case 43 and apply a little environment-friendly glue to fix it to the sponge;

[0090] 5) Apply a little environment-friendly glue to fix the magnetic charging port 6 on the outer sponge 1;

[0091] 6) Sew the inner sponge 2 and the outer sponge 1 together and edge-seal them with a wrapping cloth strip.

[0092] The operation process of this device is as follows:

[0093] 1) Align the concave area of the device with the bridge of the nose and eyes, and fix it on the user's head relying on the magic tapes at both ends.

[0094] 2) Press and hold the button for 2 seconds to turn on the machine. After the main board 3 obtains the detection signal, obtain the pitch angle, roll angle and heading angle data of the sensor in the most recent recording period (30 seconds) at a certain frequency, detect the head motion by the method of Embodiment 1, and display the detected number of motions and the amplitude of each motion on the OLED screen. At the same time, the main board 3 also displays the power data obtained from the coulomb meter on the OLED screen.

[0095] 3) After the test is over, press and hold the button for 2 seconds to turn off the machine.

[0096] 4) During the test, if the power display area is flashing, it needs to be charged in time. When charging, use the attached magnetic charger to connect. The charging progress will be displayed on the screen. When charging, the magnetic charger charges the battery 41 and sends a charging signal to the main board 3 at the same time. The main board 3 obtains the power data of the coulomb meter and displays it on the OLED screen.

[0097] It should also be noted that the installation method and installation position of the circuit module on the inner sponge 2 and the outer sponge 1 of the device are not limited, the structure of the circuit module is not limited, and whether the inner and outer sponges are fixed with magic tape is not limited. It can be designed according to actual needs and limitations to design a head motion detection device suitable for wearing on the head.

[0098] A head motion detection device provided in this embodiment is designed with an outer sponge 1, an inner sponge 2 adapted to the size of the head, and a circuit module embedded between the outer sponge 1 and the inner sponge 2. The circuit module is used to detect the head motion by implementing the head motion detection method shown in Embodiment 1, and display the detected number of motions, the amplitude of each motion, and the remaining power on the display screen. The outer sponge 1 and the inner sponge 2 are used to adapt to the size and shape of the head, facilitate wearing on the user's head, and complete the head motion detection. The head motion detection device has a simple structure and a compact installation. Both the outer sponge 1 and the inner sponge 2 are flexibly designed, which not only brings a comfortable wearing experience to the user, but also has a light weight and can be stably worn on the user's head. No matter how the user moves, the head motion of the user can be detected, and there is almost no interference to the subject. Moreover, the detection method implemented by the circuit module ensures a high detection accuracy.

[0099] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited by the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications made without departing from the spirit and principle of the present invention shall be equivalent replacement methods and are all included in the protection scope of the present invention.

Claims

1. In a head movement detection method, characterized in that, Includes steps: S1, obtaining the pitch angle, roll angle and heading angle measured by the three-axis accelerometer fixed on the head at a fixed frequency; S2. Calculate the current attitude value according to the pitch angle, roll angle and heading angle at the current sampling time; the current attitude value is equal to the square root of the sum of the squares of the pitch angle, roll angle and heading angle and round it off; S3, determining the current trend value according to the difference between the current posture value and the previous posture value; S4. If the current trend value corresponds to a constant movement trend, the number of times the continuous trend remains unchanged is increased by 1; otherwise, the number of times the continuous trend remains unchanged is reset to zero; S5, determine whether the number of times the continuous trend remains unchanged exceeds the continuous trend remains unchanged threshold, if so, go to step S9, otherwise go to the next step; S6, calculate the absolute difference between the current trend value and the previous trend value, if the absolute difference is greater than 0, then the absolute difference between the previous posture value and the last posture value of the previous movement is recorded as the amplitude of this movement and go to step S9, otherwise go to the next step; S7, calculating the absolute difference between the current posture value and the last posture value of the previous movement, if the absolute difference exceeds the posture absolute difference threshold, proceed to the next step, otherwise go to step S10; S8, the number of head movements is accumulated once and the process goes to step S10; S9, assign the last posture value to the last posture value of the last movement, assign the number of times the continuous trend remains unchanged to 0, assign the absolute difference between the current posture value and the last posture value of the last movement to 0, and proceed to the next step; S10, assigning the current posture value to the previous posture value, assigning the current trend value to the previous trend value, and returning to step S2.

2. A head movement detection method according to claim 1, characterized in that: In step S3, if the difference between the current posture value and the previous posture value is greater than 0, the current trend value trend_cur is the first preset value; if the difference between the current posture value and the previous posture value is less than 0, the current trend value trend_cur is the second preset value; if the difference between the current posture value and the previous posture value is equal to 0, the current trend value trend_cur is the third preset value; the first preset value, the second preset value, and the third preset value are not equal to each other.

3. A head movement detection method according to claim 2, characterized in that: The first preset value is 1, the second preset value is 2, and the third preset value is 0.

4. A head movement detection device, characterized in that: The invention comprises an outer sponge (1) adapted to the size and shape of a head, an inner sponge (2), and a circuit module embedded between the outer sponge (1) and the inner sponge (2), wherein the circuit module is provided with a main board (3), and the main board (3) is used to implement a head movement detection method according to any one of claims 1 to 3.

5. A head movement detection device according to claim 4, characterized in that: The circuit module further comprises a battery module (4) electrically connected to the mainboard (3), a display and button module (5) and a magnetic charging port (6), wherein the display and button module (5) and the magnetic charging port (6) are open-type and embedded on the surface of the outer sponge (1).

6. A head movement detection device according to claim 4, characterized in that: The outer sponge (1) and the inner sponge (2) are configured as flexible structures, and can be bent inwards into an annular structure that fits the size of the head.

7. A head movement detection device according to claim 6, characterized in that: The tail ends of the outer sponge (1) and the inner sponge (2) are provided with Velcro strips that fit in with each other.

8. The head movement detection device according to claim 5, characterized in that: The battery module (4) comprises a battery (41), a top frame shell (42) and a bottom frame shell (43), and the battery (41) is fixed in the top frame shell (42) and the bottom frame shell (43).

9. The head movement detection device according to claim 4, characterized in that: The head movement detection device further comprises an upper frame shell (71), a middle frame shell (72) and a lower frame shell (73), and the main board (3) and the display and key module (5) are fixed in the upper frame shell (71), the middle frame shell (72) and the lower frame shell (73).

Images