A gesture recognition method, gesture recognition controller and medium

Abstract

This application relates to the field of gesture recognition technology, providing a gesture recognition method, a gesture recognition controller, and a medium. The gesture recognition method includes: acquiring reference posture change information and hand posture change information; determining relative posture change information based on the reference posture change information and hand posture change information; and determining gesture recognition information based on the relative posture change information. The reference posture change information represents the posture change of a body reference part, the hand posture change information represents the posture change of the hand, and the relative posture change information represents the posture change of the hand relative to the body. The relative posture change information can accurately represent the posture changes made by the user's hand relative to the body, and can accurately distinguish conscious gestures even during user movement. The gesture recognition information determined based on the relative posture change information enables accurate recognition of user gestures, ensuring the accuracy and robustness of gesture recognition.

Description

Technical Field

[0001] This application relates to the field of gesture recognition technology, and in particular to a gesture recognition method, a gesture recognition controller, and a medium. Background Technology

[0002] With the continuous development of human-computer interaction technology, gesture recognition, with its intuitive and natural advantages, has become one of the development directions to replace traditional physical controllers such as mice, keyboards, and remote controls. Currently, gesture recognition mainly includes two methods: visual image recognition and inertial measurement recognition using wearable devices.

[0003] In existing technologies, gesture recognition methods based on inertial measurement rely on fixed or preset starting postures. Users may have difficulty accurately triggering gestures during natural movement. For example, when walking or running, a user's body turning may be misidentified as a gesture turning. Furthermore, body swaying can introduce interference, making it difficult to clearly distinguish whether the hand posture change is caused by body movement or the gesture itself. Consequently, it is impossible to accurately recognize the gesture. Therefore, existing gesture recognition methods based on inertial measurement require the user to be in a stationary state to ensure the accuracy of gesture recognition, which cannot adapt to user scenarios where the user is in motion.

[0004] Therefore, how to accurately and stably recognize gestures during user movement is an urgent problem to be solved. Summary of the Invention

[0005] This application provides a gesture recognition controller, method, and medium to address the shortcomings of existing technologies in accurately recognizing gestures during user movement.

[0006] This application provides a gesture recognition control method, including: Obtain reference posture change information and hand posture change information; Based on the reference posture change information and the hand posture change information, the relative posture change information is determined; Based on the relative posture change information, gesture recognition information is determined; The reference posture change information represents the posture change of a reference body part, the hand posture change information represents the posture change of the hand, and the relative posture change information represents the posture change of the hand relative to the body.

[0007] According to a gesture recognition control method provided in this application, the reference posture change information includes a first pitch angle, a first yaw angle, and a first roll angle, and the hand posture change information includes a second pitch angle, a second yaw angle, and a second roll angle; the step of determining relative posture change information based on the reference posture change information and the hand posture change information includes: Calculate the first difference between the second pitch angle and the first pitch angle, calculate the second difference between the second yaw angle and the first yaw angle, and calculate the third difference between the second roll angle and the first roll angle; The relative attitude change information is formed based on the first difference, the second difference, and the third difference.

[0008] According to a gesture recognition control method provided in this application, determining gesture recognition information based on the relative posture change information includes: The first angular range is determined based on the first difference, the second angular range is determined based on the second difference, and the third angular range is determined based on the third difference. The gesture recognition information is determined based on the combination of the first angle range, the second angle range, and the third angle range.

[0009] This application also provides a gesture recognition controller, including: The reference wearable module includes a first control unit, a first inertial measurement unit, and a first communication unit, wherein the first control unit is connected to the first inertial measurement unit and the first communication unit respectively; The hand-worn module includes a second control unit, a second inertial measurement unit, and a second communication unit, wherein the second control unit is connected to the second inertial measurement unit and the second communication unit respectively; The first communication unit is communicatively connected to the second communication unit, and the first communication unit and / or the second communication unit are also used to communicate with a host computer or a controlled object. The first inertial measurement unit is used to detect the posture change of the reference part, and the second inertial measurement unit is used to detect the posture change of the hand. The reference wearing module and the hand wearing module are used to implement the above-mentioned gesture recognition control method.

[0010] According to a gesture recognition controller provided in this application, the reference wearing module further includes a first control unit and a first control feedback unit, both of which are connected to the first control unit. And / or, the hand-wearing module further includes a second control unit and a second control feedback unit, both of which are connected to the second control unit; And / or, the reference wearing module further includes a first energy storage power supply unit and a first interface unit, wherein the first energy storage power supply unit is connected to the first control unit, the first inertial measurement unit and the first communication unit respectively, and the first interface unit is connected to the first control unit and the first energy storage power supply unit respectively; And / or, the hand-wearing module further includes a second energy storage power supply unit and a second interface unit, wherein the second energy storage power supply unit is connected to the second control unit, the second inertial measurement unit and the second communication unit respectively, and the second interface unit is connected to the second control unit and the second energy storage power supply unit respectively.

[0011] The present invention also provides a gesture recognition control method, applied to the hand-wearing module of the gesture recognition controller described above, comprising: Establish a communication connection with the reference wearing module, and establish a communication connection with the host computer or the controlled object; In response to the start recognition command, the initial hand posture information is acquired. Obtain the current state of the second hand's posture information; Based on the first hand posture information and the second hand posture information, determine the hand posture change information; Obtain the reference posture change information of the reference wearing module; Based on the hand posture change information and the reference posture change information, the relative posture change information is determined; Based on the relative posture change information, gesture recognition information is determined; The gesture recognition information is sent to the host computer or the controlled object.

[0012] According to the gesture recognition control method provided in this application, determining hand posture change information based on the first hand posture information and the second hand posture information includes: Based on the first hand posture information and the second hand posture information, the changes in the pitch angle, yaw angle and roll angle of the hand are calculated and the hand-down conversion process is performed to obtain the hand posture change information. The hand-down conversion process is used to convert the angle change of the hand-down state into the angle change corresponding to the hand-up state.

[0013] According to the gesture recognition control method provided in this application, it further includes: In response to the gesture setting command, the third hand posture information of the initial state is obtained; In response to the gesture confirmation command, obtain the fourth hand posture information of the current state; Based on the third hand posture information and the fourth hand posture information, and based on the angular changes of the hand posture in pitch angle, yaw angle and roll angle, the angle range combination information is determined; Obtain gesture command setting information, associate the angle range combination information with the gesture command setting information, and store it.

[0014] This application also provides a gesture recognition control method, applied to a reference wearing module of the gesture recognition controller described above, comprising: Establish a communication connection with the hand-wearing module; In response to the start recognition command, the first reference attitude information of the initial state is obtained; Obtain the second reference attitude information of the current state; Based on the first reference attitude information and the second reference attitude information, the reference attitude change information is determined; The reference posture change information is sent to the hand-wearing module; The reference posture change information is used to determine the relative posture change information, and then to determine the gesture recognition information.

[0015] This application also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a gesture recognition control method as described above.

[0016] This application provides a gesture recognition method, gesture recognition controller, and medium, which have at least the following beneficial effects: By acquiring independent but related reference posture change information and hand posture change information, the reference posture change information can reveal the user's body posture changes, and the hand posture change information can reveal the user's hand posture changes. Combining the reference posture change information and hand posture change information can eliminate the influence of body posture changes on hand posture changes, thereby determining the hand posture change relative to the body. This effectively eliminates the influence of overall body movement on hand posture changes, and more accurately identifies pure gesture actions. Therefore, the relative posture change information obtained based on the combination of reference posture change information and hand posture change information can accurately characterize the user's hand posture changes relative to the body. Even during user movement, it can still accurately distinguish conscious gesture actions performed by the user. Thus, the gesture recognition information determined based on the relative posture change information can achieve accurate recognition of user gestures, ensuring the accuracy and robustness of gesture recognition. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this application or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating a gesture recognition control method provided in this application.

[0019] Figure 2 This is a schematic diagram of the structure of a gesture recognition controller provided in this application.

[0020] Figure 3 This is one of the wearable schematic diagrams of a gesture recognition controller provided in this application.

[0021] Figure 4 This is the second wearable schematic diagram of a gesture recognition controller provided in this application.

[0022] Figure 5 This is a circuit diagram of the first control unit and the second control unit in one embodiment of a gesture recognition controller provided in this application.

[0023] Figure 6 This is a circuit diagram of the first inertial measurement unit and the second inertial measurement unit in one embodiment of a gesture recognition controller provided in this application.

[0024] Figure 7 This is a schematic diagram of the pitch angle, roll angle, and yaw angle in a gesture recognition control method provided in this application.

[0025] Figure 8 This is one of the schematic diagrams of a hand-raising gesture action in a gesture recognition and control method provided in this application.

[0026] Figure 9 This is the second schematic diagram of a hand-raising gesture action in a gesture recognition control method provided in this application.

[0027] Figure 10 This is the third illustration of a hand-raising gesture action in a gesture recognition and control method provided in this application.

[0028] Figure 11 This is one of the schematic diagrams of a hand gesture in a gesture recognition and control method provided in this application.

[0029] Figure 12 This is the second schematic diagram of a hand gesture in a gesture recognition and control method provided in this application.

[0030] Figure 13 This is the third illustration of a hand gesture in a gesture recognition and control method provided in this application.

[0031] Figure 14 This is a schematic diagram of the workflow of the hand-wearing module in one embodiment of the gesture recognition control method provided in this application.

[0032] Figure 15 This is a schematic diagram of the workflow of a reference wearing module in one embodiment of a gesture recognition control method provided in this application.

[0033] Figure label: 100: Reference wearing module; 110: First control unit; 120: First inertial measurement unit; 121: First gyroscope; 122: First magnetometer; 123: First accelerometer; 130: First communication unit; 140: First control unit; 150: First control feedback unit; 160: First energy storage and power supply unit; 170: First interface unit; 200: Hand wearing module; 210: Second control unit; 220: Second inertial measurement unit; 221: Second gyroscope; 222: Second magnetometer; 223: Second accelerometer; 230: Second communication unit; 240: Second control unit; 250: Second control feedback unit; 260: Second energy storage and power supply unit; 270: Second interface unit. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0035] The following is combined with Figure 1 This application describes a gesture recognition control method, comprising: S100: Acquire reference posture change information and hand posture change information; S200: Determine relative posture change information based on the reference posture change information and the hand posture change information; S300: Determine gesture recognition information based on the relative posture change information; The reference posture change information represents the posture change of a reference body part, the hand posture change information represents the posture change of the hand, and the relative posture change information represents the posture change of the hand relative to the body.

[0036] By acquiring independent but relevant reference posture change information and hand posture change information, the reference posture change information can reveal the user's body posture changes, and the hand posture change information can reveal the user's hand posture changes. Combining the reference posture change information and hand posture change information can eliminate the influence of body posture changes on hand posture changes, thereby determining the hand posture change relative to the body. This effectively eliminates the influence of overall body movement on hand posture changes, and more accurately identifies pure hand gestures. Therefore, the relative posture change information obtained based on the combination of reference posture change information and hand posture change information can accurately represent the user's hand posture changes relative to the body, even during user movement, and can still accurately distinguish... It should be noted that by combining reference posture change information and hand posture change information to determine relative posture change information, the influence of body swaying on gesture recognition can be avoided at the same time, reducing the interference of body swaying when the user makes gestures, which is conducive to improving the accuracy of gesture recognition.

[0037] This application addresses the problem of misrecognition of gestures in motion scenarios by introducing a body reference posture as a dynamic reference benchmark and calculating the relative change between the hand and the benchmark. This fundamentally solves the problem, making gesture recognition no longer limited to static environments and greatly expanding its application potential in mobile and dynamic scenarios.

[0038] It should be further emphasized that this application achieves gesture recognition based on the change in hand posture relative to the body. Users can use different positions as the starting gesture position, without being restricted to a fixed starting gesture position.

[0039] To illustrate this more intuitively, consider this example: If a user raises their hand forward during gesture recognition, but their body turns at the same time, relying solely on the hand posture change might misinterpret it as a hand-raising and turning gesture, leading to misidentification. In this application, by simultaneously detecting posture changes in reference body parts, such as the waist and chest, the changes in body posture can be identified. By combining this with hand posture changes, the portion of the hand posture change caused by the body posture change can be removed, thus eliminating the influence of body posture changes and achieving more accurate gesture recognition.

[0040] To better understand the solution in this application, a brief description of the hardware structure is provided first: (Refer to...) Figures 2 to 6 This application also provides a gesture recognition controller, including: The reference wearable module 100 includes a first control unit 110, a first inertial measurement unit 120, and a first communication unit 130, wherein the first control unit 110 is connected to the first inertial measurement unit 120 and the first communication unit 130 respectively. The hand-wearing module 200 includes a second control unit 210, a second inertial measurement unit 220, and a second communication unit 230, wherein the second control unit 210 is connected to the second inertial measurement unit 220 and the second communication unit 230 respectively. The first communication unit 130 is communicatively connected to the second communication unit 230. The first communication unit 130 and / or the second communication unit 230 are also used to communicate with a host computer or a controlled object. The first inertial measurement unit 120 is used to detect the attitude change of the reference part, and the second inertial measurement unit 220 is used to detect the attitude change of the hand.

[0041] The reference wearing module 100 is used to generate reference posture change information to reflect the posture changes of the user's body reference parts, and the hand wearing module 200 is used to generate hand posture change information to reflect the posture changes of the user's hand, providing a basis for determining relative posture change information.

[0042] The gesture recognition control method of this application is implemented based on the gesture recognition controller described above.

[0043] In some embodiments of this application, the reference posture change information and the hand posture change information can be obtained by performing attitude calculation algorithms such as Kalman filtering and complementary filtering on the detection data of the first inertial measurement unit 120 and the second inertial measurement unit 220. This process can be completed by the reference wearing module 100 and the hand wearing module 200.

[0044] In some embodiments of this application, the inertial measurement unit includes a gyroscope, an accelerometer, and a magnetometer, as referenced. Figure 7 When attitude change information includes pitch, roll, and yaw, the attitude change information can be obtained in the following way: Calculate the deflection angle (Yaw): The Kalman filter is used to fuse the detection data from the gyroscope and magnetometer to calculate the first... k State prediction matrix at time: ; in, for k State prediction matrix at time [ , ] T ; for k The estimated deflection angle at time t, with an initial value of 0; The estimated value for gyroscope bias is given, with an initial value of 0. for k The calculated matrix at time -1 [ , ] T ; u The angle value currently read by the gyroscope; F This is the state transition matrix; B To control the input matrix.

[0045] State transition matrix F With control input matrix B for: Where, d t The time interval for reading the gyroscope and magnetometer readings, or the calculation interval for data fusion, can be the same or different. The explanation follows the case where both intervals are the same; if they are different, the calculation interval for data fusion can be selected as d. t .

[0046] Calculate the first k Covariance prediction matrix at time step : ; in, F T State transition matrix F The transpose of the matrix is ​​initialized as a diagonal matrix P0 = diag([0.1 2 0.001 2 ]), 0.1 is the initial angle error, and 0.001 is the initial drift error; Q The noise covariance is related to the noise level of the sensor itself.

[0047] In some embodiments, noise covariance Q It can be: The parameters in the matrix can be adjusted based on actual measurements.

[0048] Calculation update: in, R To observe the noise covariance, the noise value of the sensor can be used, such as 0.1. H This is the observation matrix.

[0049] Calculate the updated Kalman gain K : ; in, This is the covariance prediction matrix;H The observation matrix; S Kalman gain S .

[0050] Computation Update k State calculation matrix at time step : ; in, for k The angle value read back by the magnetometer at any given moment; H The observation matrix; For the first k The state prediction matrix for the step; K Kalman gain K .

[0051] Calculate the update error covariance: ; in, K Kalman gain K ; H The observation matrix; This is the covariance prediction matrix.

[0052] calculate k The calculated value of the deflection angle at any given time and the calculated value of the gyroscope's zero bias: in, for k The deflection angle at any moment; The gyroscope is at zero bias at time 1. for k The state calculation matrix at each time step.

[0053] Through the above process, the detection data from the gyroscope and magnetometer are combined to calculate and obtain the deflection angle (Yaw). The magnetometer provides a reference for the Earth's magnetic field, which can correct the drift of the gyroscope in the deflection angle, making the obtained deflection angle (Yaw) more accurate.

[0054] Calculate the pitch and roll angles: Using a Kalman filter to fuse gyroscope and accelerometer data, calculate k Time-state prediction matrix : Where, d t The above calculation of deflection angle θ Yaw_k The process uses d tConsistent; for k The estimated gyroscope angle at time t is used to calculate the pitch angle and the roll angle, with an initial value of 0. This is the deviation value of the gyroscope, with an initial value of 0; and For the first k The calculated value at time -1; for k The gyroscope angle value read back at all times.

[0055] Calculate the state covariance matrix : in, Q The noise covariance matrix is ​​used in conjunction with the deflection angle calculated above. θ Yaw_k The noise covariance matrix used in the process Q same; and The initial value is the identity matrix.

[0056] Calculate Kalman gain S and Kalman gain K : in, The state covariance matrix; R To observe the noise covariance, R The value of is the same as the calculated deflection angle above. θ Yaw_k The process is the same.

[0057] Calculate and update the calculated value: in, for k The pitch angle value is obtained from the accelerometer at all times, and the accelerometer is used to read the first... k time y shaft and z The measured value of the shaft, i.e. a y_k and a z_k Obtained through calculation; k The roll angle value is obtained from the accelerometer at all times, and the accelerometer is used to read the first roll angle. k time x shaft andz The measured value of the shaft, i.e. a x_k and a z_k The result is obtained through calculation; arctan2 is the arctangent function in the four quadrants. for k The state prediction matrix at time step; for k The state calculation matrix at each time step.

[0058] Calculate the updated state covariance matrix: in, K Kalman gain K ; The state covariance matrix; Calculate the output angle value and the gyroscope zero bias: in, For pitch or roll, the estimated gyroscope angle used in the prior calculation. When the pitch angle is The pitch angle is the estimated angle from the gyroscope used in the previous calculation. When the roll angle is This refers to the roll angle.

[0059] Through the above process, the detection data from the gyroscope and accelerometer are integrated to calculate and obtain the pitch and roll angles. The accelerometer provides a gravity reference, which can correct the drift of the gyroscope in the pitch and roll angles, making the obtained pitch and roll angles more accurate.

[0060] The yaw, pitch, and roll angles calculated above will be used as attitude change information, i.e., as the basis for gesture recognition. Based on the data fusion calculation method described above, k The three angle values ​​at time t are as follows: Hands: θ Pitch1_k (Angle of elevation) θ Roll1_k (Roll angle) and θ Yaw1_k (Deflection angle); Reference area: θ Pitch2_k (Angle of elevation) θ Roll2_k(Roll angle) and θ Yaw2_k (Deflection angle).

[0061] refer to Figure 3 and Figure 4 The pitch angle corresponds to the x-axis and z The plane formed by the axis corresponds to the roll angle. y shaft and z The plane formed by the axis, the deflection angle corresponds to x shaft and y The plane formed by the axis.

[0062] In some embodiments of a gesture recognition control method of this application, the reference posture change information includes a first pitch angle, a first yaw angle, and a first roll angle, and the hand posture change information includes a second pitch angle, a second yaw angle, and a second roll angle; step S200 includes: Calculate the first difference between the second pitch angle and the first pitch angle, calculate the second difference between the second yaw angle and the first yaw angle, and calculate the third difference between the second roll angle and the first roll angle; The relative attitude change information is formed based on the first difference, the second difference, and the third difference.

[0063] The posture changes of the hand and the reference part are specifically quantified into three-axis angular changes of pitch angle, yaw angle and roll angle. By calculating the difference between the hand angle and the reference part angle on each axis, the change of the hand relative to the reference part on each axis is obtained. The angle difference of the three axes forms the relative posture change information.

[0064] By calculating the three-axis angle difference, the pure hand gesture movement relative to the body reference point can be accurately quantified, thereby effectively eliminating the influence of body posture changes on gestures. This ensures that the relative posture change information can accurately reflect the user's intention in making the gesture, without being interfered with by body movement, which is beneficial to improving the accuracy and precision of gesture recognition.

[0065] It should be noted that the pitch, yaw, and roll angles in the attitude change information are determined based on the angle changes between the initial state and the current state, rather than solely based on the currently determined angle. That is, the pitch, yaw, and roll angles represent the angle changes, not the absolute values ​​of the angles, thus enabling them to characterize attitude changes.

[0066] To understand this more intuitively, the initial state k The three angle values ​​at time t are as follows: Hands: θ Pitch1_k (Angle of elevation) θRoll1_k (Roll angle) and θ Yaw1_k (Deflection angle); Reference area: θ Pitch2_k (Angle of elevation) θ Roll2_k (Roll angle) and θ Yaw2_k (Deflection angle); Current status k + n The three angle values ​​at time t are as follows: Hands: θ Pitch1_k+n (Angle of elevation) θ Roll1_k+n (Roll angle) and θ Yaw1_k+n (Deflection angle); Reference area: θ Pitch2_k+n (Angle of elevation) θ Roll2_k+n (Roll angle) and θ Yaw2_k+n (Deflection angle).

[0067] Reference points for attitude change information: First pitch angle Δ θ Pitch2_k+n = θ Pitch2_k+n - θ Pitch2_k ; First deflection angle Δ θ Yaw2_k+n = θ Yaw2_k+n - θ Yaw2_k ; First roll angle Δ θ Roll2_k+n = θ Roll2_k+n - θ Roll2_k ; Hand posture change information: Second pitch angle Δ θ Pitch1_k+n = θ Pitch1_k+n - θ Pitch1_k ; Second deflection angle Δ θ Yaw1_k+n = θ Yaw1_k+n - θYaw1_k ; Second roll angle Δ θ Roll1_k+n = θ Roll1_k+n - θ Roll1_k .

[0068] In some embodiments of a gesture recognition control method of this application, step S300 includes: The first angular range is determined based on the first difference, the second angular range is determined based on the second difference, and the third angular range is determined based on the third difference. The gesture recognition information is determined based on the combination of the first angle range, the second angle range, and the third angle range.

[0069] The three-axis relative attitude change information—namely, the first difference, the second difference, and the third difference—is matched with preset angle ranges. If the three differences match a preset angle range combination, then that combination is considered to correspond to a preset gesture, achieving the goal of gesture recognition. Therefore, for each preset gesture definition, a certain range of variation is allowed in the pitch, yaw, and roll directions, permitting deviations when the user executes the gesture. The magnitude of the deviation depends on the size of the preset angle range, thereby improving the fault tolerance of gesture recognition.

[0070] In some embodiments of this application, the first angle range, the second angle range, and the third angle range can be customized by the user to suit the user's gesture habits.

[0071] refer to Figures 8 to 13 In some embodiments of this application, the gesture recognition information is determined based on a combination of the first angle range, the second angle range, and the third angle range, which can be achieved through the following expression: in, ACT 1 to ACT 6 represents the combination of six angle ranges corresponding to the gestures, with a deviation range of ±30°, which can be modified according to user needs. θ* Pitch1 to θ* Pitch6, θ* Roll1 to θ* Roll6 , θ* Yaw1 to θ* Yaw6 As a preset gesture angle, in some embodiments of this application, the initial values ​​are set as shown in Table 1 below: Table 1 Initial values ​​for each angle Gesture movement Pitch angle Roll angle Yaw angle 1 θ* Pitch1 = 60°]]> ​ <![CDATA[ θ* Roll1 = 0°]]> <![CDATA[ θ* Yaw1 = 0°]]> 2 <![CDATA[ θ* Pitch2 = - 60°]]> <![CDATA[ θ* Roll2 = 0°]]> <![CDATA[ θ* Yaw2 = 0°]]> 3 <![CDATA[ θ* Pitch3 = 0°]]> <![CDATA[ θ* Roll3 = 60°]]> <![CDATA[ θ* Yaw3 = 0°]]> 4 <![CDATA[ θ* Pitch4 = 0°]]> <![CDATA[ θ* Roll4 = - 60°]]> <![CDATA[ θ* Yaw4 = 0°]]> 5 <![CDATA[ θ* Pitch5 = 0°]]> <![CDATA[ θ* Roll5 = 0°]]> <![CDATA[ θ* Yaw5 = 60°]]> 6 <![CDATA[ θ* Pitch6 = 0°]]> <![CDATA[ θ* Roll6 = 0°]]> <![CDATA[ θ* Yaw6 = - 60°]]> The aforementioned gesture angle values ​​can be set according to user needs, such as through user-recorded gestures. The deviation range can also be set by the user, allowing them to adjust the range to their liking. It's understood that the number of gestures is not fixed at six and can be increased or decreased based on user requirements. Figures 8 to 10 These are six hand gestures for raising your hand. Figures 11 to 13 These are six hand gestures for when your hands are hanging down.

[0072] It should be emphasized that the gesture recognition goal achieved in this application does not rely on complex artificial intelligence models for processing, has low requirements for computing power and hardware performance, and has the advantage of low implementation cost.

[0073] The following describes a gesture recognition controller provided in this application. The gesture recognition controller described below can be referred to in correspondence with the gesture recognition control method described above.

[0074] refer to Figures 2 to 6 This application also provides a gesture recognition controller, including: The reference wearable module 100 includes a first control unit 110, a first inertial measurement unit 120, and a first communication unit 130, wherein the first control unit 110 is connected to the first inertial measurement unit 120 and the first communication unit 130 respectively. The hand-wearing module 200 includes a second control unit 210, a second inertial measurement unit 220, and a second communication unit 230, wherein the second control unit 210 is connected to the second inertial measurement unit 220 and the second communication unit 230 respectively. The first communication unit 130 is communicatively connected to the second communication unit 230. The first communication unit 130 and / or the second communication unit 230 are also used to communicate with a host computer or a controlled object. The first inertial measurement unit 120 is used to detect the attitude change of the reference part, and the second inertial measurement unit 220 is used to detect the attitude change of the hand.

[0075] The gesture recognition controller includes a reference wearing module 100 and a hand wearing module 200. The reference wearing module 100 is equipped with a first inertial measurement unit 120 for detecting posture changes of reference body parts, and the hand wearing module 200 is equipped with a second inertial measurement unit 220 for detecting hand posture changes. When a user performs a gesture, the hand posture change includes not only the movement of the gesture itself but also the superimposed movement of the entire body. The reference wearing module 100 acquires the posture changes of reference body parts, such as the waist and chest, while the hand wearing module 200 acquires the hand posture changes. The hand wearing module 200 is communicatively connected to the reference wearing module 100. Since the overall body movement has a similar impact on the reference body parts and hand posture, combining the posture changes of the reference body parts with the hand posture changes can remove the influence of body posture changes on hand posture changes, thereby determining the hand posture change relative to the body.

[0076] Therefore, based on the reference wearing module 100 and the hand wearing module 200, the posture change of the hand relative to the body reference part can be obtained. Even when the user is exercising, the relative change of the user's hand posture can be accurately detected, and pure hand gestures can be more accurately identified, thereby improving the accuracy of gesture recognition and adapting to the user's exercise usage scenarios.

[0077] The difference between this and general technical methods lies in the fact that traditional single-sensor gesture recognition systems, or systems without body posture references, suffer from high false recognition rates because hand movements are often confused with body movements when the user moves, making it difficult to distinguish between a gesture and body sway. This controller, by introducing posture detection of body reference parts and co-analyzing hand postures, fundamentally solves the robustness problem of gesture recognition in motion scenarios, enabling users to perform gesture control freely and conveniently in a wider range of situations.

[0078] In some embodiments of this application, the hardware structure of the reference wearing module 100 and the hand wearing module 200 can be the same, that is, the same hardware structure is used to detect the posture changes of the reference part and the hand.

[0079] In some embodiments of this application, the first control unit 110 and the second control unit 210 may be implemented using devices such as microcontrollers and embedded chips. The first communication unit 130 and the second communication unit 230 may be implemented using devices such as Bluetooth chips and WIFI chips.

[0080] It is understandable that the controlled object can be a robot, smart home device, or other similar device.

[0081] refer to Figure 2In some embodiments of a gesture recognition controller in this application, the first inertial measurement unit 120 includes a first gyroscope 121, which is connected to the first control unit 110. The second inertial measurement unit 220 includes a second gyroscope 221, which is connected to the second control unit 210.

[0082] A gyroscope is a sensor that measures angular velocity and can detect the posture of an object. A first gyroscope 121 and a second gyroscope 221 are respectively mounted on a reference part and the hand, enabling real-time detection of the posture of the reference part and the hand. The first gyroscope 121 and the second gyroscope 221 transmit the detected data to a first control unit 110 and a second control unit 210, respectively. This allows the first control unit 110 to detect changes in the posture of the reference part and the second control unit 210 to detect changes in the posture of the hand, providing data support for subsequently determining the posture changes of the hand relative to the body.

[0083] Understandably, as a type of inertial sensor, the core function of a gyroscope is to measure angular velocity, which reflects the attitude of an object. In scenarios where high accuracy is not required, detection can be performed using only a gyroscope.

[0084] refer to Figure 2 In some embodiments of a gesture recognition controller in this application, the first inertial measurement unit 120 further includes a first magnetometer 122 and a first accelerometer 123, both of which are connected to the first control unit 110. The second inertial measurement unit 220 also includes a second magnetometer 222 and a second accelerometer 223, both of which are connected to the second control unit 210.

[0085] Accelerometers measure linear acceleration and gravity vectors, allowing for the calculation of an object's tilt angle in space. Magnetometers measure the strength and direction of the Earth's magnetic field, calculating the object's deflection angle. By introducing accelerometers and magnetometers into an inertial measurement unit (IMU), which already includes a gyroscope, the IMU can more comprehensively and accurately detect an object's attitude. The accelerometer provides a gravity reference, correcting for gyroscope drift in pitch and roll angles, while the magnetometer provides a reference to the Earth's magnetic field, correcting for gyroscope drift in deflection angles. This improves upon gyroscope drift, enabling more accurate detection of hand and reference body postures, thereby enhancing the precision of gesture recognition.

[0086] It should be noted that the changes in the posture of the reference body part and the hand can be described based on pitch angle, yaw angle and roll angle to quantify the changes in posture.

[0087] In some embodiments of this application, the first inertial measurement unit 120 and the second inertial measurement unit 220 may include an inertial measurement module chip, which integrates a gyroscope, a magnetometer, and an accelerometer. In some embodiments of this application, the first inertial measurement unit 120 and the second inertial measurement unit 220 may also be implemented using a gyroscope chip, a magnetometer chip, and an accelerometer chip, respectively.

[0088] refer to Figure 2 In some embodiments of a gesture recognition controller in this application, the reference wearing module 100 further includes a first control unit 140, which is connected to the first control unit 110; And / or, the hand-wearing module 200 further includes a second control unit 240, which is connected to the second control unit 210.

[0089] By providing a first control unit 140 on the reference wearing module 100 and / or a second control unit 240 on the hand wearing module 200, users can actively operate the first control unit 140 and the second control unit 240 to interact with the gesture recognition process according to the needs of gesture recognition. This allows users to flexibly enable or disable the gesture recognition function, helping to avoid accidental gesture recognition leading to miscontrols and improving ease of use and reliability.

[0090] It is understandable that by using the first control unit 140 and the second control unit 240 for operation and interaction, the gesture recognition function can be turned on or off as needed, thereby reducing unnecessary power consumption and processing resource occupation.

[0091] In some embodiments of a gesture recognition controller in this application, the first control unit 140 includes a first recognition switch button, which is connected to the first control unit 110; And / or, the second control unit 240 includes a second identification switch button, which is connected to the second control unit 210.

[0092] By providing a first recognition switch button on the reference wearing module 100 and / or a second recognition switch button on the hand wearing module 200, users can conveniently activate or deactivate the gesture recognition function by operating the first and / or second recognition switch buttons. This allows for easy user operation using a simple structure.

[0093] In some embodiments of this application, the first control unit 140 and the second control unit 240 may include implementations of devices such as buttons, non-contact sensing buttons, and touch screens.

[0094] refer to Figure 2 In some embodiments of a gesture recognition controller in this application, the reference wearing module 100 further includes a first control feedback unit 150, which is connected to the first control unit 110. And / or, the hand-wearing module 200 further includes a second control feedback unit 250, which is connected to the second control unit 210.

[0095] By providing a first control feedback unit 150 on the reference wearing module 100 and / or a second control feedback unit 250 on the hand wearing module 200, feedback prompts can be given when the user performs operations, allowing the user to know whether the operation is successful, whether the communication pairing is successful, and whether the gesture recognition function is enabled. This helps the user confirm the validity of the operation and understand the current operating status, thereby improving the user experience.

[0096] In some embodiments of this application, the first control feedback unit 150 and the second control feedback unit 250 may include implementations using devices such as LEDs, vibration motors, speakers, and touch screens. It is understood that when using devices such as touch screens, the control unit and the control feedback unit may be the same device.

[0097] refer to Figure 2 In some embodiments of a gesture recognition controller in this application, the reference wearing module 100 includes a first energy storage power supply unit 160, which is connected to the first control unit 110, the first inertial measurement unit 120 and the first communication unit 130 respectively. The hand-wearing module 200 includes a second energy storage power supply unit 260, which is connected to the second control unit 210, the second inertial measurement unit 220 and the second communication unit 230 respectively.

[0098] By incorporating a first energy storage power supply unit 160 in the wearable module 100 and a second energy storage power supply unit 260 in the hand-wearing module 200, the entire gesture recognition controller becomes portable, eliminating reliance on an external power source and improving user convenience during exercise. Furthermore, the built-in energy storage power supply unit structure eliminates the need for external battery power and avoids cable tangling issues, thus enhancing the user experience.

[0099] refer toFigure 2 In some embodiments of a gesture recognition controller in this application, the reference wearing module 100 further includes a first interface unit 170, which is connected to the first control unit 110 and the first energy storage power supply unit 160 respectively. The hand-wearing module 200 also includes a second interface unit 270, which is connected to the second control unit 210 and the second energy storage power supply unit 260 respectively.

[0100] The wearable module 100 is equipped with a first interface unit 170, and the wearable hand module 200 is equipped with a second interface unit 270. The first interface unit 170 and the second interface unit 270 provide a structural foundation for charging, data transmission, and maintenance. Users can charge the built-in energy storage power supply unit through the interface units to ensure continuous operation of the device. At the same time, the interface units can also be connected to external devices such as computers to realize functions such as firmware upgrades, parameter configuration, or system debugging.

[0101] In some embodiments of this application, the first interface unit 170 and the second interface unit 270 may include implementations of device structures such as USB interfaces and contact interfaces.

[0102] In some embodiments of a gesture recognition controller in this application, the first interface unit 170 includes a first magnetic interface, and the second interface unit 270 includes a second magnetic interface.

[0103] The first interface unit 170 includes a first magnetic interface, and the second interface unit 270 includes a second magnetic interface. The magnetic force is used to realize the automatic alignment and adsorption connection of the interface, which helps to simplify the interface plugging operation. In addition, the magnetic interface is usually a closed design, which has excellent waterproof and dustproof performance, which helps to improve the durability and reliability in complex environments such as humid and dusty environments, and is suitable for use in outdoor or sports scenarios.

[0104] In some embodiments of a gesture recognition controller in this application, the first energy storage power supply unit 160 includes a first battery, a first battery protection circuit and a first battery charging circuit. The first battery is connected to the first control unit 110, the first inertial measurement unit 120 and the first communication unit 130 respectively. The first battery protection circuit is connected to the first battery. The first battery charging circuit is connected to the first interface unit 170 and the first battery respectively. The second energy storage power supply unit 260 includes a second battery, a second battery protection circuit, and a second battery charging circuit. The second battery is connected to the second control unit 210, the second inertial measurement unit 220, and the second communication unit 230, respectively. The second battery protection circuit is connected to the second battery, and the second battery charging circuit is connected to the second interface unit 270 and the second battery, respectively.

[0105] The energy storage power supply unit integrates a battery, battery protection circuit, and battery charging circuit to form a complete and reliable power management structure. The battery protection circuit prevents excessive battery output current, avoiding overheating or overcurrent, thus extending battery life and improving safety. The battery charging circuit ensures an efficient and safe charging process, preventing overcharging and battery damage.

[0106] It should be noted that the process of determining gesture recognition information based on steps S100 to S300 can be implemented in the hand-wearing module 200 or in the reference-wearing module 100. In application scenarios with a host computer, it can also be implemented in the host computer, that is, the host computer obtains the hand posture change information from the hand-wearing module 200 and the reference posture change information from the reference-wearing module 100 respectively, and then determines the gesture recognition information through processing.

[0107] The following description will focus on an embodiment in which gesture recognition information is determined in the hand-wearing module 200.

[0108] The gesture recognition control method provided in this application is further described below. The gesture recognition control method described below can be referred to in correspondence with the steps S100 to S300 described above and the gesture recognition controller.

[0109] refer to Figure 14 This application also provides a gesture recognition control method, applied to the hand-wearing module 200 in the gesture recognition controller described above, comprising: Establish a communication connection with the reference wearing module 100, and establish a communication connection with the host computer or the controlled object; In response to the start recognition command, the initial hand posture information is acquired. Obtain the current state of the second hand's posture information; Based on the first hand posture information and the second hand posture information, determine the hand posture change information; Obtain the reference posture change information of the reference wearing module 100; Based on the hand posture change information and the reference posture change information, the relative posture change information is determined; Based on the relative posture change information, gesture recognition information is determined; The gesture recognition information is sent to the host computer or the controlled object.

[0110] The process of determining gesture recognition information described above is implemented by the hand-wearing module 200, which acts as the master station and the reference wearing module 100 as the slave station. In addition to communicating with the reference wearing module 100, the hand-wearing module 200 also communicates with the host computer or the controlled object. Upon receiving the start recognition command, it acquires the initial hand posture information and the current second hand posture information in real time to determine the hand posture change information. Combined with the reference posture change information acquired from the reference wearing module 100, it determines the relative posture change information of the hand, and thus determines the gesture recognition information. As the main carrier of user interaction, the hand-wearing module 200 can respond instantly to and process the start recognition command generated by the user's operation, achieving the gesture recognition goal. This helps reduce response latency and improve the timeliness of gesture recognition.

[0111] refer to Figure 14 In some embodiments of this application, during the process of establishing a communication connection between the hand-wearing module 200 and the host computer or the controlled object, pairing verification can be performed, that is, sending verification request information to the host computer or the controlled object, and establishing a communication connection when the correct verification information is returned; or, responding to the verification request information of the host computer or the controlled object, returning the verification information determined by itself, and establishing a communication connection after returning the success information.

[0112] In some embodiments of the gesture recognition control method of this application, determining the hand posture change information based on the first hand posture information and the second hand posture information includes: Based on the first hand posture information and the second hand posture information, the changes in the pitch angle, yaw angle and roll angle of the hand are calculated and the hand-down conversion process is performed to obtain the hand posture change information. The hand-down conversion process is used to convert the angle change of the hand-down state into the angle change corresponding to the hand-up state.

[0113] Users may perform gesture control with their arms either naturally hanging down or raised, i.e., in a lowered or raised hand state. By introducing a lowered hand conversion process, the angle changes in the lowered hand state are converted into angle changes in the raised hand state. This means that gesture recognition is uniformly based on the angle changes in the raised hand state, eliminating the influence of the user's initial arm state on the final gesture recognition result. Therefore, regardless of whether the user's gesture is in a lowered or raised hand state, based on the first and second hand posture information, the gesture is uniformly converted into the angle changes corresponding to the raised hand state for gesture recognition. This achieves adaptive gesture recognition, without forcing the user's initial arm state. Users can perform gestures in a wider range of more natural initial postures without consciously adjusting their initial arm posture, thus improving the naturalness and convenience of gesture recognition.

[0114] In some embodiments of this application, hand posture change information can be obtained in the following ways: First, the hand angle is calculated and converted using the following expression: in, , , These are the first hand posture information, namely the pitch angle, roll angle, and yaw angle in the initial state. , , These are the second hand posture information, namely the pitch angle, roll angle, and yaw angle of the current state.

[0115] Through the above calculations and conversions, the three changing angles are unified between -180° and 180°.

[0116] Then, the above angles are uniformly mapped to a plane with an elevation angle of 0, which is achieved through the following calculation expression: In the above calculation and conversion process, Δ θ Pitch_k+n Δ θ Roll_k+n With Δ θ Yaw_k+n It can be replaced with the Δ of the hand. θ Pitch1_k+n Δ θ Roll1_k+n and Δ θ Yaw1_k+n After conversion , , These are the second pitch angle, second roll angle, and second yaw angle, respectively, in the hand posture change information.

[0117] The above calculation and conversion process can also be replaced by Δ at the reference location. θ Pitch2_k+n Δ θ Roll2_k+n With Δ θ Yaw2_k+n This is to map the attitude changes of the reference part onto a plane with a pitch angle of 0.

[0118] After the above mapping, gesture recognition can be performed using a plane with a pitch angle of 0 as a reference, that is, by uniformly changing the angle of raising the hand.

[0119] refer to Figure 14 In some embodiments of the gesture recognition control method of this application, it further includes: In response to the gesture setting command, the third hand posture information of the initial state is obtained; In response to the gesture confirmation command, obtain the fourth hand posture information of the current state; Based on the third hand posture information and the fourth hand posture information, and based on the angular changes of the hand posture in pitch angle, yaw angle and roll angle, the angle range combination information is determined; Obtain gesture command setting information, associate the angle range combination information with the gesture command setting information, and store it.

[0120] Based on user needs, gesture actions can be customized. When the hand-wearing module 200 receives a gesture setting instruction, it means the user needs to set a gesture, and the initial third hand posture information is obtained. When a gesture confirmation instruction is received, it means the current gesture is the gesture the user needs to set, and the current fourth hand posture information is obtained. Based on the third and fourth hand posture information, the hand posture change can be determined, and then the pitch, roll, and yaw angle values ​​can be determined, as shown in Table 1 above. Combined with a preset or user-set deviation range, such as ±30°, the angle range combination information is determined. Then, the user gesture command setting information is obtained, that is, the command corresponding to the gesture that the user needs to set is obtained. The angle range combination information is associated with the gesture command setting information and stored, so that in the subsequent gesture recognition process, when the gesture corresponding to the angle range combination information is recognized, the corresponding gesture command can be sent or executed.

[0121] In this way, users can customize the correspondence between gesture actions and gesture commands according to their own habits and preferences, no longer limited to preset fixed gesture actions. This enables the learning and memorization of user-set gesture actions and corresponding gesture commands, improving the flexibility of gesture recognition and enhancing the personalized user experience, which is conducive to improving operational efficiency and convenience.

[0122] It is understandable that since users are usually in a non-moving state when setting gestures, the angle range combination information can be determined based on the third and fourth hand posture information, without needing to consider the posture changes of the reference part. In some embodiments of this application, the angle range combination information can also be determined based on the third and fourth hand posture information, combined with the posture changes of the reference part, after eliminating the influence of the posture changes of the reference part.

[0123] It should be emphasized that the above gesture setting and gesture recognition processes do not rely on neural network models, that is, they do not require the use of neural network models for processing. This helps to reduce the requirements for hardware performance and computing power, thereby reducing implementation costs.

[0124] refer to Figure 14 This application also provides a gesture recognition control method, which, after obtaining the second hand posture information of the current state, further includes: Based on the second hand posture information, if the hand angle change is determined to be within a preset range and lasts for a preset time, then enter sleep mode.

[0125] After the hand-wearing module 200 responds to the start recognition command, it monitors the hand posture in real time. If the hand posture does not change significantly for an extended period, that is, based on the second hand posture information, it determines that the hand angle change is within a preset range and continues for a preset time. For example, if the hand posture deflection angle is within 1° and continues for one minute, it exits the gesture recognition mode and enters sleep mode. Therefore, if the user's hand posture does not change significantly for an extended period, it means the user may have forgotten to turn off the gesture recognition function. Thus, by actively exiting the gesture recognition mode and entering sleep mode, power is saved.

[0126] In some embodiments of this application, before the hand-wearing module 200 determines to enter sleep mode, it may send a sleep command to the reference wearing module 100 so that the reference wearing module 100 also enters sleep mode.

[0127] refer to Figure 15 This application also provides a gesture recognition control method, applied to the reference wearing module 100 of the gesture recognition controller described above, comprising: Establish a communication connection with the hand-wearing module 200; In response to the start recognition command, the first reference attitude information of the initial state is obtained; Obtain the second reference attitude information of the current state; Based on the first reference attitude information and the second reference attitude information, the reference attitude change information is determined; The reference posture change information is sent to the hand-wearing module 200; The reference posture change information is used to determine the relative posture change information, and then to determine the gesture recognition information.

[0128] The reference wearing module 100, acting as a slave station, is responsible for providing information on changes in the posture of the reference body part. Upon receiving a start recognition command, the reference wearing module 100 acquires the first reference posture information of the initial state and, in real time, acquires the second reference posture information of the current state to determine the reference posture change information. The reference wearing module 100 sends the reference posture change information to the hand wearing module 200, which acts as the master station. This allows the hand wearing module 200 to combine the reference posture change information with its own hand posture change information to determine the relative posture change information of the hand, thereby determining the gesture recognition information and realizing the gesture recognition function. Therefore, the reference wearing module 100, acting as the provider of reference body part information for the hand wearing module 200, can have its design and implementation relatively simplified, without needing to undertake complex gesture recognition processing.

[0129] It should be noted that the above description uses the hand-wearing module 200 as the master station and the reference wearing module 100 as the slave station. That is, the hand-wearing module 200 determines the gesture recognition information and sends it to the host computer or the controlled object, while the reference wearing module 100 only provides reference posture change information. In some embodiments of this application, the hand-wearing module 200 can also be implemented as the slave station and the reference wearing module 100 as the master station. That is, the reference wearing module 100 determines the gesture recognition information and sends it to the host computer or the controlled object, while the hand-wearing module 200 only provides hand posture change information. The processing interaction process is similar to the above situation and can be determined by simple analogy, so it will not be described in detail again.

[0130] In some embodiments of this application, the hardware structure of the reference wearing module 100 and the hand wearing module 200 can be the same, that is, the same hardware structure can be used to detect the posture changes of the reference part and the hand. In some embodiments of this application, the master station and the slave station can use different processors, but other hardware structures are the same. One of them, as the slave station, can use a processor with relatively weak performance to further reduce implementation costs.

[0131] This application also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, is implemented to perform a gesture recognition control method provided by the methods described above.

[0132] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0133] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0134] In the description of this application, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0135] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0136] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A gesture recognition control method, characterized in that, include: Obtain reference posture change information and hand posture change information; Based on the reference posture change information and the hand posture change information, the relative posture change information is determined; Based on the relative posture change information, gesture recognition information is determined; The reference posture change information represents the posture change of a reference body part, the hand posture change information represents the posture change of the hand, and the relative posture change information represents the posture change of the hand relative to the body.

2. The gesture recognition control method according to claim 1, characterized in that, The reference posture change information includes a first pitch angle, a first yaw angle, and a first roll angle; the hand posture change information includes a second pitch angle, a second yaw angle, and a second roll angle; determining the relative posture change information based on the reference posture change information and the hand posture change information includes: Calculate the first difference between the second pitch angle and the first pitch angle, calculate the second difference between the second yaw angle and the first yaw angle, and calculate the third difference between the second roll angle and the first roll angle; The relative attitude change information is formed based on the first difference, the second difference, and the third difference.

3. The gesture recognition control method according to claim 2, characterized in that, The step of determining gesture recognition information based on the relative posture change information includes: The first angular range is determined based on the first difference, the second angular range is determined based on the second difference, and the third angular range is determined based on the third difference. The gesture recognition information is determined based on the combination of the first angle range, the second angle range, and the third angle range.

4. A gesture recognition controller, characterized in that, include: The reference wearable module (100) includes a first control unit (110), a first inertial measurement unit (120), and a first communication unit (130), wherein the first control unit (110) is connected to the first inertial measurement unit (120) and the first communication unit (130), respectively; The hand-wearing module (200) includes a second control unit (210), a second inertial measurement unit (220), and a second communication unit (230), wherein the second control unit (210) is connected to the second inertial measurement unit (220) and the second communication unit (230) respectively; The first communication unit (130) is communicatively connected to the second communication unit (230). The first communication unit (130) and / or the second communication unit (230) are also used to communicate with a host computer or a controlled object. The first inertial measurement unit (120) is used to detect the posture change of the reference part. The second inertial measurement unit (220) is used to detect the posture change of the hand. The reference wearing module (100) and the hand wearing module (200) are used to implement a gesture recognition control method as described in any one of claims 1 to 3.

5. A gesture recognition controller according to claim 4, characterized in that, The reference wearing module (100) further includes a first control unit (140) and a first control feedback unit (150), both of which are connected to the first control unit (110). And / or, the hand-wearing module (200) further includes a second control unit (240) and a second control feedback unit (250), both of which are connected to the second control unit (210); And / or, the reference wearing module (100) further includes a first energy storage power supply unit (160) and a first interface unit (170), wherein the first energy storage power supply unit (160) is connected to the first control unit (110), the first inertial measurement unit (120) and the first communication unit (130) respectively, and the first interface unit (170) is connected to the first control unit (110) and the first energy storage power supply unit (160) respectively; And / or, the hand-wearing module (200) further includes a second energy storage power supply unit (260) and a second interface unit (270), wherein the second energy storage power supply unit (260) is connected to the second control unit (210), the second inertial measurement unit (220) and the second communication unit (230) respectively, and the second interface unit (270) is connected to the second control unit (210) and the second energy storage power supply unit (260) respectively.

6. A gesture recognition control method, characterized in that, A hand-wearing module for a gesture recognition controller as described in claim 4 or 5, comprising: Establish a communication connection with the reference wearing module, and establish a communication connection with the host computer or the controlled object; In response to the start recognition command, the initial hand posture information is acquired. Obtain the current state of the second hand's posture information; Based on the first hand posture information and the second hand posture information, determine the hand posture change information; Obtain the reference posture change information of the reference wearing module; Based on the hand posture change information and the reference posture change information, the relative posture change information is determined; Based on the relative posture change information, gesture recognition information is determined; The gesture recognition information is sent to the host computer or the controlled object.

7. The gesture recognition control method according to claim 6, characterized in that, The step of determining hand posture change information based on the first hand posture information and the second hand posture information includes: Based on the first hand posture information and the second hand posture information, the changes in the pitch angle, yaw angle and roll angle of the hand are calculated and the hand-down conversion process is performed to obtain the hand posture change information. The hand-down conversion process is used to convert the angle change of the hand-down state into the angle change corresponding to the hand-up state.

8. The gesture recognition control method according to claim 6, characterized in that, Also includes: In response to the gesture setting command, the third hand posture information of the initial state is obtained; In response to the gesture confirmation command, obtain the fourth hand posture information of the current state; Based on the third hand posture information and the fourth hand posture information, and based on the angular changes of the hand posture in pitch angle, yaw angle and roll angle, the angle range combination information is determined; Obtain gesture command setting information, associate the angle range combination information with the gesture command setting information, and store it.

9. A gesture recognition control method, characterized in that, A reference wearable module applied to a gesture recognition controller as described in claim 4 or 5, comprising: Establish a communication connection with the hand-wearing module; In response to the start recognition command, the first reference attitude information of the initial state is obtained; Obtain the second reference attitude information of the current state; Based on the first reference attitude information and the second reference attitude information, the reference attitude change information is determined; The reference posture change information is sent to the hand-wearing module; The reference posture change information is used to determine the relative posture change information, and then to determine the gesture recognition information.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements a gesture recognition control method as described in any one of claims 1 to 3 and 6 to 9.

Images