TWI625114B - Motion recognition device and method - Google Patents

Abstract

A motion recognition device includes a sensing unit and a processing unit. The sensing unit generates a sensing signal according to a body motion occurring at a specific position on a user's body, wherein the sensing signal includes a first sensing signal portion and is different from the first sensing signal a second sensing signal portion, and the body motion belongs to a segment motion of a motion type. The processing unit processes the sensing signal to generate a motion parameter signal structure, and identifies the specific position based on the motion parameter signal structure to determine a valid reference signal for identifying the motion type, wherein the motion parameter signal structure includes the a fusion signal of the first and second sensing signal portions.

Description

Motion recognition device and method

The present disclosure relates to an identification device and method, and more particularly to a motion recognition device and method.

The main function of the conventional motion sensing device is to generate a sensing signal by sensing a back and forth motion, and interpret the sensing signal to count the number of round trips of the back and forth motion. The most important representative item on the market is the pedometer; however, this type of sensing device does not recognize the type of motion or the mode of motion, and can only simply calculate the number of movements of the back and forth motion. Since the conventional motion sensing device has the above-described operational characteristics, the portable motion sensing device is limited for recognizing the usability of various types of motion.

U.S. Patent No. 8,260,384 B2 discloses a portable mobile computer system. A method and apparatus for obtaining and analyzing mobile training materials for strength training is disclosed in US Patent Application Publication No. 2014/0336,947. U.S. Patent No. 8,787,006 B2 discloses a wrist-worn electronic device and method of use thereof. US Patent Application Publication No. 2008/0262392 A1 discloses a calibration technique for a motion sensing device. US Published Patent Application No. 2013/0044215 A1 discloses a bistable spring having a flexible display. US Patent Publication No. 2014/0172873 A1 A method and apparatus for personally characterizing data collection using a sensor is disclosed.

A motion sensing device should be able to clearly identify a variety of motion features to clearly distinguish a type of motion from other types of motion, and reward the recognition result to the fitness person; thus, the fitness effect can be effectively assisted by the fitness person. . For example, there are three types of exercise to be made by the exerciser, and the three exercise types belong to the same exercise type group. Additionally, the three types of motion can have similar motion. When the exerciser makes the three types of exercise in different exercise modes or habits, the muscle patterns created by the three exercise types will be significantly different. The three types of motion may be a pinch (inner clip/abduction) type of motion, a type of leg lift exercise, and a type of positive leg lift. In order to effectively recognize similar body motions, a motion recognition device is provided to recognize the three motion types and additionally to identify other motion types.

In addition, a fitness instructor (or senior fitness person) having fitness trainer qualifications and a fitness beginner respectively make a first body motion and a second body motion compared to the first body motion. Since the fitness instructor has received more training for a longer period of time, the first body motion is typically a predetermined standard sport. Since the fitness beginner has accepted less training for a shorter period of time, the deviation between the second body motion and the predetermined standard motion is relatively large. The reason for the deviation is that the movement posture of the second body movement deviates from the movement posture of the predetermined standard movement, or the second body movement is incomplete, or the training accepted by the fitness beginner is not sufficient, The second body movement may include an inadvertent or unconscious habitual action to form the deviation. The inadvertent or unconscious habitual action can be detected by the motion sensing device; therefore, the motion sensing device reminds the fitness beginner to improve the second body motion, and promotes the fitness beginner to make the predetermined standard motion And improve the results of sports training.

The motion sensing device is capable of detecting a detailed change in the body motion during the body movement. For example, a physical trainee uses a chair to make a physical movement of the Skull Crusher. The French referral includes an action: moving the lower arm (arm) to the top of the eye and then rotating horizontally to the vertical. When the body trainer wants to exercise muscles more effectively, the French press can be adjusted to include more difficult movements: move the lower arm to place the lower arm about perpendicular to the body, and then lift the lower arm up, Then line the lower arm straight to the upper arm. The motion sensing device is capable of recognizing subtle changes in the body movement, reporting the identified type of motion to the body trainee, and notifying the body trainee when the body movement has been completed and/or the associated muscles have been exercised .

The body trainee can use an exercise device in the gym to make the body exercise. For example, when the body trainee operates the fitness device of the driven row machine and holds the crossbar of the rowing machine with both hands to make the body movement, wearing a body part of the body trainee The motion sensing device is capable of recognizing the body motion, wherein there is a distance between the two hands of the crossbar, and the distance is relatively narrow or wide to allow for changes in the muscle group being exercised. For example, when the physical trainee does a sit-up on a lounge chair in a gym (Sit-ups) When exercised, the motion sensing device worn to a body portion of the body trainer can identify the sit-up motion based on the external condition of the recliner. For example, the recliners in the gym each have different angles of inclination, and each body trainee can prefer different angles of inclination of a recliner depending on the person. Under the external condition of the reclining chair, the motion sensing device is capable of obtaining an appropriate signal for analysis, accurately identifying the motion state, counting the motion cycle, and obtaining the analysis result.

Conventional wearable motion sensing devices have long had functional limitations; therefore, multiple conditions of body motion are needed to identify the type of motion of the body motion. The purpose of the present disclosure is to propose a motion recognition device. The motion recognition device can sense various body movements and recognize the type of exercise, count a motion cycle of body motion to obtain a number of exercise cycles, and store the number of times of data or a state of motion or the result of the body motion In a database; therefore, the athletic data stored in the database can be subsequently processed and reapplied to provide a reference motion recommendation to the user of the motion recognition device. The motion recognition device of the present disclosure is wearable and is particularly suitable for body trainers engaged in physical exercise in the gym.

An embodiment of the present disclosure provides a motion recognition apparatus. The identification device comprises a sensing unit and a processing unit. The sensing unit generates a sensing signal according to a first body motion occurring at a specific position on a user's body, wherein the sensing signal includes a first sensing signal portion and is different from the first feeling a second sensing signal portion of the test signal, and the first body motion belongs to a segment motion of a first motion type. The processing unit processes the sensing signal to generate a motion parameter signal knot And identifying the specific position based on the motion parameter signal structure to determine a valid reference signal for identifying the first motion type, wherein the motion parameter signal structure includes one of the first and second sensing signal portions Fusion signal.

Another embodiment of the present disclosure is to provide an identification device. The identification device comprises a signal generating unit and a processing unit. The signal generating unit generates a motion parameter signal structure in response to a body motion, wherein the body motion occurs at a specific position on a user's body and belongs to a first motion type. The processing unit identifies the specific location based on the motion parameter signal structure to determine a valid reference signal for identifying the first motion type.

Yet another embodiment of the present disclosure is to provide an identification method. The identification method includes the steps of: generating a motion parameter signal structure in response to a first body motion, wherein the first body motion occurs at a specific location on a user's body and belongs to a first motion type; The motion parameter signal structure identifies the particular location to determine a valid reference signal for identifying the first motion type.

Yet another embodiment of the present disclosure is to provide a motion recognition device. The motion recognition device comprises a memory module and a processing module. The memory module stores a feature function code data unit associated with a first identification value area and a second identification value area, wherein the first and second identification value areas are adjacent and respectively indicate a first A type of exercise and a second type of exercise. The processing module obtains a specific motion feature value data unit derived from a specific motion sensing signal, based on the feature function code And the material unit and the specific motion characteristic value data unit perform a calculation to generate an identification value, and determine whether the identification value belongs to one of the first and second identification value regions.

20, 30, 40, 50, 60, 61‧‧‧ sports identification devices

21‧‧‧Sensor unit

211‧‧‧Accelerometer

212‧‧‧Gyro

22, 42, 56‧ ‧ processing unit

221‧‧‧Processing module

222‧‧‧ memory module

23‧‧‧ buttons

35, 37, 51‧‧‧ operating units

36, 52‧‧‧ coupling unit

41‧‧‧Signal generating unit

411‧‧‧ conversion unit

42‧‧‧Processing unit

55‧‧‧Sports sensing unit

561‧‧‧Sports Identification Module

562‧‧‧Signal Processing Module

57‧‧‧Human Machine Interface Unit

90‧‧‧Users

91‧‧‧ body

911‧‧‧ specific body parts

A1, a2, a3, a4‧‧‧ motion eigenvalues

a _z,max ‧‧‧maximum eigenvalue

a _z,min ‧‧‧minimum eigenvalue

B1‧‧‧first maximum

B2‧‧‧ first minimum

B3‧‧‧ second maximum

B4‧‧‧ second minimum

CA‧‧‧Recognition Reference Signal Code Data Unit

CA1‧‧‧second candidate reference signal code

CA2‧‧‧ first candidate reference signal code

CA3‧‧‧ third candidate reference signal code

CB1‧‧‧ representative signal code

CH‧‧‧Sport type indicator data unit

CH2‧‧‧ sport type indicator

C _i~N ‧‧‧ eigenvalue

C _M1,i~n , C _M2,i~n , , ‧‧‧Specific eigenvalues

CP1‧‧‧ specific location code

CT1‧‧‧ sport type code

DA1‧‧‧Second motion characteristic value data unit

DA2‧‧‧First motion characteristic value data unit

DA21‧‧‧First motion characteristic value data unit part

DA21A‧‧‧First motion characteristic value data unit subsection

DA22‧‧‧Second motion characteristic value data unit

DA22A‧‧‧Second motion characteristic value data unit subsection

DA3‧‧‧ third motion characteristic value data unit

DA8‧‧‧Specific motion characteristic value data unit

DC1‧‧‧Differential data unit

DC11‧‧‧ representative extreme deviation

DC12‧‧‧ representative value difference

DFE _n ‧‧‧ sports feature values

DH1‧‧‧ first identification value

DH2‧‧‧ second identification value

DM1‧‧‧Sports Measurement Information

DQ1, DQ2, DQ6‧‧‧ main motion axis direction data unit

DR‧‧‧ Identification Reference Unit

DU1‧‧‧ first identification value area

DU2‧‧‧ second identification value area

DU3‧‧‧ third identification value area

DU4‧‧‧ fourth identification value area

DUA‧‧‧ effective identification area

DV1‧‧‧First Trust Value Area

DV2‧‧‧Second Trusted Value Area

E1‧‧‧ first difference

E2‧‧‧ second difference

F1‧‧‧first feature function code data unit

F2‧‧‧Second feature function code data unit

G1, G2, G6‧‧‧ sports groups

GA‧‧‧First Sports Group

GB‧‧‧Second Sports Group

gF‧‧‧gravity acceleration

g _x , g _y , g _z ‧‧‧gravity acceleration component

H1‧‧‧ Third sport type

H2‧‧‧ Fourth sport type

H3‧‧‧ Fifth sport type

H4‧‧‧ sixth sport type

HA1, HA2, HA6‧‧‧ sports types

HM1‧‧‧First sport type

HM2‧‧‧Second sports type

KA1‧‧‧ Orientation

L‧‧‧Number of all eigenvalues

MC1, MC2, MC9‧‧‧ body movement

ML1‧‧‧First Body Movement

ML11‧‧‧ first movement part

ML12‧‧‧Second sports section

P ₁ , P ₂ ‧‧‧Sport type cutting function

PL1‧‧‧ specific location

PL5, PL6, PL9‧‧‧ position

(q ₀ , q ₁ , q ₂ , q ₃ ) ‧ ‧ quaternion

QA1, QA2, QA6‧‧‧ main motion axis direction

QF1‧‧‧Gravity direction

QH1‧‧‧Predetermined direction

Ref _i ‧‧‧ candidate reference signal

SE1‧‧‧ sensing signal

SE11‧‧‧ accelerometer signal

SE12‧‧‧Gyroscope signal

SE1A‧‧‧First sensing signal section

SE1B‧‧‧Second sensing signal section

SE2‧‧‧Special motion sensing signal

SK1‧‧‧ Estimated angle

SP1‧‧‧motion parameter signal structure

SP11, SP12, SP13, SP18, SP19‧‧‧ motion parameter signals

SP5‧‧‧First motion parameter signal structure

SP6‧‧‧Second motion parameter signal structure

SPS0‧‧‧ fusion signal

SPS1‧‧‧ representative signal

SPS11‧‧‧First motion parameter signal section

SPS12‧‧‧Second motion parameter signal section

SR1‧‧‧second candidate reference signal

SR2‧‧‧ first candidate reference signal

SR21‧‧‧First candidate reference signal section

SR22‧‧‧Second candidate reference signal section

SR3‧‧‧ third candidate reference signal

SRG‧‧‧ candidate reference signal combination

SRP‧‧‧ valid reference signal

ST1‧‧‧ trigger signal

T21, T22, T23, T61, T62, T63, T64‧‧‧ time

UA1‧‧‧ attached coordinate system

V1, V2‧‧‧ cutting plane

W1‧‧‧First Movement Cycle

W2‧‧‧Secondary exercise cycle

X, Y, Z, x, y, z, x ^A , y ^A , z ^A , x ^K , y ^K , z ^K , x ^W , y ^W , z ^W ‧‧‧ reference axis

Φ‧‧‧ rolling angle

φ _Roll ‧‧‧ Estimated roll angle

Θ‧‧‧first angle (pitch angle)

θ _Pitch ‧‧‧ Estimated pitch angle

Ψ‧‧‧ yaw angle

ψ _Yaw ‧‧‧ Estimated yaw angle

Σ‧‧‧threshold distance

Σ1‧‧‧trust value distance

δ _n ‧‧‧ threshold

ξ _i~n ‧‧‧ threshold range

ΔC _{M2~Mi, i~n} ‧‧‧ eigenvalue difference

‧‧‧First signal

‧‧‧ Accelerometer signal ideal component

‧‧‧Accelerometer signal noise component

‧‧‧Accelerometer signal bias component

‧‧‧Accelerometer signal gravity component

, , , , , ‧‧‧Original sensing signal component

‧‧‧third signal

, ‧‧‧Acoustic signal

, ‧‧‧Acceleration differential signal

, , , , φ _{Roll , max} , φ _{Roll , min} , θ _{Pitch , max} , θ _{Pitch , min} ‧ ‧ eigenvalues

‧‧‧gravity acceleration signal component

‧‧‧Second signal

ω _x ‧‧‧ angular velocity signal

‧‧‧The ideal component of the gyro signal

‧‧‧Gyroscope signal noise component

‧‧‧Gyroscope signal bias component

‧‧‧fourth signal

The disclosure is further understood by the following detailed description of the drawings: FIG. 1 is a schematic diagram of a motion recognition device in various embodiments of the present disclosure.

Fig. 2 is a schematic view showing the movement recognizing device of Fig. 1 worn on a user.

Figure 3 is a schematic illustration of a motion recognition device in various embodiments of the present disclosure.

4A, 4B, and 4C: Schematic diagrams of types of body motion recognized by a motion recognition device worn on the wrist.

Figure 5: Schematic diagram of the type of body motion recognized by a motion recognition device worn on the upper arm.

Figure 6 is a schematic illustration of the type of body motion recognized by a motion recognition device worn on the upper arm.

Figures 7A and 7B are schematic illustrations of the types of body motion recognized by a motion recognition device worn on the upper arm.

Figure 8 is a schematic illustration of the type of body motion recognized by a motion recognition device worn on the upper arm.

Figures 9A and 9B are schematic views of the type of body motion recognized by a motion recognition device worn on the ankle.

Figures 10A and 10B are schematic views of the type of body motion recognized by a motion recognition device worn on the ankle.

Figure 11 is a schematic illustration of a reference coordinate system in various embodiments of the present disclosure.

Fig. 12 is a schematic view showing the relationship between the motion recognition device and a user's body in Fig. 2.

Figure 13 is a schematic diagram of a first candidate reference signal and a representative signal of the motion recognition device in Figure 1.

Figure 14 is a schematic illustration of the recognition of the type of motion of a first body motion in various embodiments of the present disclosure.

Fig. 15 is a flow chart showing a first motion type identification calculation program in various embodiments of the present disclosure.

Figure 16 is a flow chart showing a second motion type identification calculation program in various embodiments of the present disclosure.

Figure 17A is a block diagram showing the structure of a motion recognition device in various embodiments of the present disclosure.

Fig. 17B is a schematic view showing the appearance of the motion recognition device in Fig. 17A.

Fig. 18 is a schematic view showing the wrist recognizing device of the motion recognition device in Fig. 17A.

Fig. 19 is a view showing the upper arm of the exercise recognizing device in Fig. 17A.

This disclosure will be explained below, as indicated below, The illustrations are for illustrative purposes only and are not intended to limit the precise forms disclosed.

Please refer to Figure 1 and Figure 2. 1 is a schematic illustration of a motion recognition device 20 in various embodiments of the present disclosure. Figure 2 is a schematic illustration of the motion recognition device 20 worn on a user 90 in Figure 1. As shown in FIGS. 1 and 2, the motion recognition device 20 includes a sensing unit 21 and a processing unit 22. The sensing unit 21 generates a sensing signal SE1 according to a first body motion ML1 occurring at a specific position PL1 of the user's body 91. The sensing signal SE1 includes a first sensing signal portion SE1A, And a second sensing signal portion SE1B different from the first sensing signal portion SE1A, and the first body motion ML1 belongs to a segment motion of a first motion type HM1. The processing unit 22 processes the sensing signal SE1 to generate a motion parameter signal structure SP1, and identifies the specific position PL1 based on the motion parameter signal structure SP1 to determine a valid reference signal SRP for identifying the first motion type HM1. The motion parameter signal structure SP1 includes a fused signal SPS0 of the first and second sensing signal portions SE1A and SE1B.

The motion recognition device 20 is configured to have a certain direction KA1, a gravity direction QF1, and an attached coordinate system UA1 for determining the orientation KA1, and is worn at the specific position PL1. In some embodiments, the motion recognition device 20 has the user 90 having the user body 91 that makes the first body motion ML1, and the user body 91 has a plurality of different positions PL5, PL6. ,..., PL9. The particular location PL1 is optionally selected from the user's body 91. The number of different positions PL5, PL6, ..., PL9. The processing unit 22 is coupled to the sensing unit 21. The first motion type HM1 is selected from the complex motion types HA1, HA2, ..., HA6. The complex motion types HA1, HA2, ..., HA6 are predetermined in relation to the specific position PL1, and respectively have a plurality of main motion axis directions QA1, QA2, ..., QA6 associated with the attached coordinate system UA1. The plurality of main motion axis directions QA1, QA2, ..., QA6 are pre-detected to generate a plurality of main motion axis direction data units DQ1, DQ2, ..., DQ6 corresponding to the plurality of main motion axis directions QA1, QA2, ..., QA6, respectively. . For example, the first body movement ML1 is a gymnastics practice; the first motion type HA1 is a drill type; and the main motion axis direction QA1 is a main rotation axis direction or a main translation axis direction.

The sensing signal SE1 is generated in relation to the attached coordinate system UA1, and includes an accelerometer signal SE11 and a gyro signal SE12. The accelerometer signal SE11 and the gyro signal SE12 are the first and the first Two sensing signal sections SE1A and SE1B. For example, the fused signal SPS0 is generated by using a signal fusion operation. In a state where the orientation KA1 is directed to a predetermined direction QH1 related to the gravity direction QF1, the processing unit 22 generates a specific position code CP1 representing the specific position PL1 based on the motion parameter signal structure SP1, and thereby identifying This particular location PL1. The predetermined direction QH1 is determined based on the specific position PL1. The predetermined direction QH1 and the gravity direction QF1 have a first angle θ. The fused signal SPS0 is a signal that estimates one of the angles SK1 associated with the first angle θ. In this state, the processing unit 22 determines whether the processing unit 22 detects a trigger signal ST1 to make a first judgment. Break the result. When the first determination result is affirmative, the processing unit 22 generates the specific position code CP1 based on the estimated angle SK1. For example, the plurality of different positions PL5, PL6, ..., PL9 include a wrist position, an upper arm position, and an ankle position. For example, the specific position PL1 is the wrist position, the upper arm position, or the ankle position.

The processing unit 22 obtains an identification reference data unit DR based on the specific position code CP1. The identification reference data unit DR is predetermined based on the complex main motion axis direction data units DQ1, DQ2, ..., DQ6, and includes a candidate reference signal code data unit CA and a motion type indicator data unit CH. The candidate reference signal code data unit CA includes a first candidate reference signal code CA2. The first candidate reference signal code CA2 represents a first candidate reference signal SR2 derived from the motion parameter signal structure SP1. The motion type indicator data unit CH includes a motion type indicator CH2 corresponding to the first candidate reference signal code CA2. The motion type indicator CH2 indicates one of an invalid motion type and a second motion type HM2, and the second motion type HM2 is included in the plural motion type HA1, HA2, ..., HA6. The processing unit 22 determines the first candidate reference signal SR2 based on the motion parameter signal structure SP1 and the first candidate reference signal code CA2.

In some embodiments, the sensing unit 21 includes an accelerometer 211 and a gyroscope 212. The accelerometer 211 and the gyroscope 212 are both coupled to the processing unit 22. The accelerometer 211 generates the accelerometer signal SE11. The gyroscope 212 generates the gyroscope signal SE12. The sensing signal SE1 includes the accelerometer signal SE11 and the gyroscope signal SE12. The processing unit 22 processes the accelerometer signal SE11 and the gyro signal SE12 by performing the signal fusion operation to generate the motion parameter signal structure SP1. The signal fusion operation includes a signal for calculating the first angle θ based on the accelerometer signal SE11 and the gyro signal SE12 to form the estimated angle SK1.

In some embodiments, the motion recognition device 20 further includes a button 23 coupled to the processing unit 22. The user body 91 has a particular body portion 911 at the particular location PL1. The motion recognition device 20 is worn to the specific body portion 911. The specific body part 911 makes the first body motion ML1 to drive the motion recognition device 20. In the state in which the orientation KA1 is directed to the predetermined direction QH1 associated with the gravity direction QF1, the button 23 causes the processing unit 22 to receive the trigger signal ST1 in response to a user press from the user 90. For example, the orientation KA1 has a pointing reference axis; the user body 91 is configured to have a reference position and is configured to have a first corresponding to the specific position PL1 based on the reference position and an ergonomic principle A limited outer region is predetermined; and in the state, the pointing reference axis points to a first one of the first predetermined limited outer regions. For example, the user body 91 is configured to have a plurality of predetermined limited outer regions based on the reference position and the ergonomics, wherein the plurality of predetermined limited outer regions are different and correspond to the plurality of different positions PL5, PL6, respectively. ,..., PL9. For example, the particular body portion 911 is the wrist, upper arm or ankle of the user's body 91.

In some embodiments, the first body movement ML1 package A first moving portion ML11 and a second moving portion ML12 adjacent to the first moving portion ML11 are included. The first and second moving portions ML11 and ML12 respectively form a first motion cycle W1 and a second motion cycle W2 adjacent to the first motion cycle W1. For example, the first body movement ML1 is a back and forth movement. The motion parameter signal structure SP1 includes a first motion parameter signal structure portion SP5 and a second motion parameter signal structure portion SP6 corresponding to the first and second motion portions ML11 and ML12, respectively. The first candidate reference signal SR2 includes a first candidate reference signal portion SR21 and a second candidate reference signal portion SR22 corresponding to the first and second motion portions ML11 and ML12, wherein the second candidate reference signal portion The SR 22 is adjacent to the first candidate reference signal portion SR21. The processing unit 22 obtains a first motion feature value data unit DA2 from the motion parameter signal structure SP1 based on the first candidate reference signal SR2. For example, the first and second motion parameter signal structure portions SP5 and SP6 are generated on the first and second motion cycles W1 and W2, respectively. The first and second candidate reference signal portions SR21 and SR22 are generated on the first and second motion cycles W1 and W2, respectively.

For example, the first motion feature value data unit DA2 includes a first motion feature value data unit portion DA21 and a second motion feature value data unit portion DA22 corresponding to the first and second motion portions ML11 and ML12, respectively. The processing unit 22 obtains the first motion feature value data unit portion DA21 from the first motion parameter signal structure portion SP5 based on the first candidate reference signal portion SR21. The processing unit 22 is based on the second candidate reference signal portion SR22 from the second motion parameter The second motion feature value data unit portion DA22 is obtained in the signal structure portion SP6.

In some embodiments, the processing unit 22 generates a difference data unit DC1 based on the first and second motion feature value data unit portions DA21 and DA22, and determines whether the difference data unit DC1 is satisfied for a period of motion. A first specific condition is initially determined to make a second determination result. When the second determination result is affirmative, the processing unit 22 recognizes that the valid reference signal SRP is the first candidate reference signal SR2, and determines that the first body motion ML1 meets a predetermined period motion start condition, and based on the first A candidate reference signal code CA2 determines whether the motion type indicator CH2 indicates one of the plurality of motion types HA1, HA2, ..., HA6 to make a third determination result. When the third determination result is affirmative, the processing unit 22 recognizes that the first motion type HM1 is the second motion type HM2 based on the motion type indicator CH2, and obtains a motion representing the first motion type HM1. Type code CT1.

The motion parameter signal structure SP1 includes complex motion parameter signals SP11, SP12, SP13, ..., SP18 and SP19. The identification reference data unit DR further includes a representative signal code CB1 representing a representative signal SPS1 included in the complex motion parameter signals SP11, SP12, SP13, ..., SP18 and SP19. The processing unit 22 determines the representative signal SPS1 based on the motion parameter signal structure SP1 and the representative signal code CB1. The processing unit 22 obtains the first and second motion feature value data unit portions DA21 by comparing the first and second motion feature value data unit portions DA21 and DA22. A representative extreme value deviation between DC22 is DC11. For example, the fused signal SPS0 is the motion parameter signal SP18.

For example, the first motion feature value data unit portion DA21 includes a first motion feature value data unit sub-portion DA21A corresponding to the representative signal SPS1, wherein the first motion feature value data unit sub-portion DA21A has a first maximum a value, a first minimum value, and a first difference between the first maximum value and the first minimum value. The second motion feature value data unit portion DA22 includes a second motion feature value data unit sub-portion DA22A corresponding to the representative signal SPS1, wherein the second motion feature value data unit sub-portion DA22A has a second maximum value, a second minimum value, and a second difference between the second maximum value and the second minimum value. The processing unit 22 obtains a representative value difference DC12 from the first and the second differences. The difference data unit DC1 includes the representative extreme value deviation DC11 and the representative value difference DC12. The first specific condition includes a first sub-condition and a second sub-condition. The first sub-condition is that the representative extreme value deviation DC11 falls within a first predetermined range of values. The second sub-condition is that the representative value difference DC12 falls within a second predetermined range of values.

In some embodiments, the candidate reference signal code data unit CA further includes at least one second candidate reference signal code CA1 representing at least one second candidate reference signal SR1 (and optionally a third candidate reference signal SR3) (and A third candidate reference signal code CA3) is selectable. The second candidate reference signal code CA1 represents the second candidate reference signal SR1 originating from the motion parameter signal structure SP1. The first candidate reference signal SR2 and the at least one second candidate reference signal SR1 (and optional SR3) form a candidate reference signal combination SRG.

The processing unit 22 determines the at least one second candidate reference signal SR1 (and the selectable SR3) based on the motion parameter signal structure SP1 and the at least one second candidate reference signal code CA1 (and optional CA3), and is based on The at least one second candidate reference signal SR1 (and the selectable SR3) obtains at least one second motion corresponding to the at least one second candidate reference signal SR1 (and the selectable SR3) from the motion parameter signal structure SP1. The feature value data unit DA1 (and optionally a third motion feature value data unit DA3). When the processing unit 22 processes the first motion feature value data unit DA2, the processing unit 22 processes the at least one second motion feature value data unit DA1 (and selectable DA3) to determine whether the candidate reference signal combination SRG includes The valid reference signal SRP.

In some embodiments, when the second determination result is negative, the first body motion ML1 further has a third motion portion, and a fourth motion portion adjacent to the third motion portion, and the processing unit Re-determining a fourth candidate reference signal based on the motion parameter signal structure SP1 and the candidate reference signal code data unit CA, and obtaining a fourth motion feature value from the motion parameter signal structure SP1 based on the fourth candidate reference signal. And a data unit, and thereby determining whether the valid reference signal SRP is recognized as the fourth candidate reference signal. In some embodiments, when the determination result of the candidate reference signal combination SRG includes the valid reference signal SRP is negative, the first body motion ML1 further has a third motion portion, and is adjacent to the third motion. a fourth moving part of the part, and the The processing unit 22 re-determines an additional candidate reference signal combination based on the motion parameter signal structure SP1 and the candidate reference signal code data unit CA, and determines whether the additional candidate reference signal combination includes the valid reference signal SRP, and thereby determining the Whether the valid reference signal SRP is recognized as the fourth candidate reference signal. In some embodiments, the third moving portion is the second moving portion ML12.

In some embodiments, the second motion portion ML12 is later than the first motion portion ML11; the fourth motion portion is later than the third motion portion; and the first motion feature value data unit DA2 includes a plurality of specific motion feature values. . The processing unit 22 provides a first feature function code data unit F1 associated with the first candidate reference signal code CA2 and a first complex identification value region DU1, DU2, and DV1 in advance. The first feature function code data unit F1 is employed to indicate one of the first complex number value regions DU1, DU2, and DV1. The first complex identification value areas DU1, DU2, and DV1 include a first identification value area DU1, a second identification value area DU2 adjacent to the first identification value area DU1, and the first and second identifications. A first trust value region DV1 between the value regions DU1 and DU2. For example, the first trust value area DV1 is selectable. The first and second identification value areas DU1 and DU2 respectively indicate a third motion type H1 and a fourth motion type H2. The third and fourth motion types H1 and H2 are all included in the plurality of motion types HA1, HA2, ..., HA6.

The motion recognition device 20 pre-establishes the first feature function code data unit F1 by sensing a plurality of body motions MC1, MC2, . . . , MC9 occurring at the specific position PL1, wherein the plurality of body motions MC1, MC2, ..., MC9 are divided into complex motion groups G1, G2, ..., G6, and the complex motion groups G1, G2, ..., G6 belong to the complex motion types HA1, HA2, ..., HA6, respectively. The first feature function code data unit F1 represents a motion type cutting function and is represented based on a relationship between the complex specific motion feature values, wherein the motion type cutting function is employed to indicate the first complex identification value region One of DU1, DU2, and DV1. For example, the processing unit 22 provides the first complex identification value areas DU1, DU2, and DV1 in advance.

When the third determination result is negative, the processing unit 22 performs a first calculation to generate a first identification value DH1 based on the first feature function code data unit F1 and the first motion feature value data unit DA2, and It is determined whether the first identification value DH1 belongs to one of the first and second identification value areas DU1 and DU2 to make a fourth determination result. When the fourth determination result is affirmative, the processing unit 22 determines that the first identification value DH1 belongs to a valid identification value area DUA in the first and second identification value areas DU1 and DU2, and identifies the first The motion type HM1 is an effective motion type indicated by the valid identification value region DUA, and the motion type code CT1 corresponding to the valid identification value region DUA is obtained. The processing unit 22 generates motion measurement information DM1 related to the first body motion ML1 based on the motion type code CT1.

In some embodiments, one of the first motion feature value data unit DA2, the first motion feature value data unit portion DA21, and the second motion feature value data unit portion DA22 includes the complex specific motion feature value. When the third determination result is negative, the processing unit The first calculation is performed based on the first feature function code data unit F1 and the complex specific motion feature value to generate the first identification value DH1.

In some embodiments, the processing unit 22 further provides a second feature function code data unit F2 associated with the first candidate reference signal code CA2 and a second complex identification value region DU3, DU4, and DV2. The second feature function code data unit F2 is different from the first feature function code data unit F1 and is used to indicate one of the second complex identification value regions DU3, DU4 and DV2. The second complex identification value areas DU3, DU4, and DV2 include a third identification value area DU3, a fourth identification value area DU4 adjacent to the third identification value area DU3, and the third and fourth identifications. A second trust value region DV2 between the value regions DU3 and DU4. For example, the first trust value area DV1 is selectable. The third and fourth identification value regions DU3 and DU4 respectively indicate a fifth motion type H3 and a sixth motion type H4, and the fifth and sixth motion types H3 and H4 are included in the complex motion type HA1. HA2,..., HA6. For example, the processing unit 22 provides the second complex identification value regions DU3, DU4, and DV2 in advance.

In a second specific condition, one of the third and fourth identification value regions DU3 and DU4 at least partially overlaps with one of the first and second identification value regions DU1 and DU2. When the fourth determination result is negative, the processing unit 22 performs a second calculation based on the second feature function code data unit F2 and the first motion feature value data unit DA2 to generate a second identification value DH2, and determines Whether the second identification value DH2 belongs to the third and fourth identification value areas DU3 and DU4 First, and thereby identifying the first motion type HM1.

In the various embodiments provided in FIGS. 1 and 2, a motion recognition device 30 includes a memory module 222 and a processing module 221. The memory module 222 stores a first feature function code data unit F1 associated with a first identification value area DU1 and a second identification value area DU2, wherein the first and second identification value areas DU1 and DU2 are Adjacent, and respectively indicating a second motion type H1 and a third motion type H2. The processing module 221 obtains a specific motion feature value data unit DA8 derived from a specific motion sensing signal SE2, and performs a first calculation based on the first feature function code data unit F1 and the specific motion feature value data unit DA8. A first identification value DH1 is generated, and it is determined whether the first identification value DH1 belongs to one of the first and second identification value areas DU1 and DU2.

The motion recognition device 30 further includes a sensing unit 21. The sensing unit 21 generates the specific motion sensing signal SE2 according to a first body motion ML1, wherein the first body motion ML1 occurs at a specific position PL1 on a user's body 91 and belongs to a first motion type. A fragment of HM1 action. For example, the specific motion sensing signal SE2 is the sensing signal SE1. The memory module 222 and the processing module 221 are disposed in a processing unit 22, and the processing unit 22 is disposed in the motion recognition device 30. The processing module 221 is coupled to the memory module 222 and the sensing unit 21, and processes the specific motion sensing signal SE2 to generate a motion parameter signal structure SP1, and identifies the specific location based on the motion parameter signal structure SP1. PL1 determines one for identifying the first motion type HM1 Valid reference signal SRP. The specific motion sensing signal SE2 includes an accelerometer signal SE11 and a gyroscope signal SE12. The motion parameter signal structure SP1 includes the accelerometer signal SE11 and a fused signal SPS0 of the gyro signal SE12. For example, the fused signal SPS0 is generated by using a signal fusion operation.

The motion recognition device 30 is configured to have a certain direction KA1, a gravity direction QF1, and an attached coordinate system UA1 for determining the orientation KA1, and is worn at the specific position PL1. In some embodiments, the motion recognition device 30 has the user 90 having the user body 91 that makes the first body motion ML1, and the user body 91 has a plurality of different positions PL5, PL6. ,..., PL9. The particular location PL1 is optionally selected from the plurality of different locations PL5, PL6, ..., PL9 on the user's body 91. The first motion type HM1 is selected from the complex motion types HA1, HA2, ..., HA6. The complex motion types HA1, HA2, ..., HA6 are predetermined in relation to the specific position PL1, and respectively have a plurality of main motion axis directions QA1, QA2, ..., QA6 associated with the attached coordinate system UA1. The plurality of main motion axis directions QA1, QA2, ..., QA6 are pre-detected to generate a plurality of main motion axis direction data units DQ1, DQ2, ..., DQ6 corresponding to the plurality of main motion axis directions QA1, QA2, ..., QA6, respectively. . For example, the first body motion ML1 is a body exercise; the first motion type HA1 is a practice type; and the main motion axis direction QA1 is a main rotation axis direction or a main translation axis direction.

The specific motion sensing signal SE2 is generated in relation to the attached coordinate system UA1. In this orientation KA1 is pointed to the gravity In a state of a predetermined direction QH1 of the direction QF1, the processing module 221 generates a specific position code CP1 representing the specific position PL1 based on the motion parameter signal structure SP1, and thereby identifies the specific position PL1. The predetermined direction QH1 is determined based on the specific position PL1. The predetermined direction QH1 and the gravity direction QF1 have a first angle θ. The motion parameter signal structure SP1 includes a signal that estimates the angle SK1 associated with the first angle θ. In this state, the processing module 221 determines whether the processing module 221 detects a trigger signal ST1 to make a first determination result. When the first determination result is affirmative, the processing module 221 generates the specific position code CP1 based on the estimated angle SK1. For example, the plurality of different positions PL5, PL6, ..., PL9 include a wrist position, an upper arm position, and an ankle position. For example, the specific position PL1 is the wrist position, the upper arm position, or the ankle position.

The processing module 221 obtains an identification reference data unit DR based on the specific position code CP1. The identification reference data unit DR is predetermined based on the plurality of main motion axis direction data units DQ1, DQ2, ..., DQ6, obtained from the memory module 222, and includes a candidate reference signal code data unit CA and a motion Type indicator data unit CH. The candidate reference signal code data unit CA includes a first candidate reference signal code CA2. The first candidate reference signal code CA2 represents a first candidate reference signal SR2 derived from the motion parameter signal structure SP1. The motion type indicator data unit CH includes a motion type indicator CH2 corresponding to the first candidate reference signal code CA2. The motion type indicator CH2 indicates an invalid motion type and a second motion type HM2 First, the second motion type HM2 is included in the plurality of motion types HA1, HA2, ..., HA6. The processing module 221 determines the first candidate reference signal SR2 based on the motion parameter signal structure SP1 and the first candidate reference signal code CA2.

The motion recognition device 30 previously establishes the first feature function code data unit F1 by sensing the plurality of body motions MC1, MC2, ..., MC9 occurring at the specific position PL1. The plurality of body movements MC1, MC2, ..., MC9 are divided into a plurality of motion groups G1, G2, ..., G6, and the plurality of motion groups G1, G2, ..., G6 belong to the plurality of motion types HA1, HA2, ..., respectively HA6, and the plurality of motion types HA1, HA2, ..., HA6 comprise the second and the third motion types H1 and H2. The first feature function code data unit F1 represents a motion type cutting function and is associated with the first candidate reference signal code CA2 and a first complex identification value region DU1, DU2 and DV1. The first feature function code data unit F1 is employed to indicate one of the first complex identification value regions DU1, DU2, and DV1. The first complex identification value areas DU1, DU2, and DV1 include a first identification value area DU1, a second identification value area DU2 adjacent to the first identification value area DU1, and the first and second identifications. A first trust value region DV1 between the value regions DU1 and DU2. For example, the first trust value area DV1 is selectable.

In some embodiments, the sensing unit 21 includes an accelerometer 211 and a gyroscope 212. The accelerometer 211 and the gyroscope 212 are both coupled to the processing module 221. The accelerometer 211 generates the accelerometer signal SE11. The gyroscope 212 generates the gyroscope signal SE12. The The sensing signal SE1 includes the accelerometer signal SE11 and the gyroscope signal SE12. The processing module 221 processes the accelerometer signal SE11 and the gyroscope signal SE12 by performing the signal fusion operation to generate the motion parameter signal structure SP1. The signal fusion operation includes a signal for calculating the first angle θ based on the accelerometer signal SE11 and the gyro signal SE12 to form the estimated angle SK1.

In some embodiments, the motion recognition device 20 further includes a button 23 coupled to the processing module 221. The user body 91 has a particular body portion 911 at the particular location PL1. The motion recognition device 30 is worn to the specific body portion 911. The specific body part 911 makes the first body motion ML1 to drive the motion recognition device 20. In the state in which the orientation KA1 is directed to the predetermined direction QH1 associated with the gravity direction QF1, the button 23 causes the processing module 221 to receive the trigger signal ST1 in response to user pressing from the user 90. For example, the orientation KA1 has a pointing reference axis; the user body 91 is configured to have a reference position and is configured to have a first corresponding to the specific position PL1 based on the reference position and an ergonomic principle A limited outer region is predetermined; and in the state, the pointing reference axis points to a first one of the first predetermined limited outer regions. For example, the user body 91 is configured to have a plurality of predetermined limited outer regions based on the reference position and the ergonomics, wherein the plurality of predetermined limited outer regions are different and correspond to the plurality of different positions PL5, PL6, respectively. ,..., PL9. For example, the particular body portion 911 is the wrist, upper arm or ankle of the user's body 91.

In some embodiments, after the trigger signal ST1 is received, the first body motion ML1 includes a first motion portion ML11 and a second motion portion ML12 adjacent to the first motion portion ML11. The first and second moving portions ML11 and ML12 respectively form a first motion cycle W1 and a second motion cycle W2 adjacent to the first motion cycle W1. The motion parameter signal structure SP1 includes a first motion parameter signal structure portion SP5 and a second motion parameter signal structure portion SP6 corresponding to the first and second motion portions ML11 and ML12, respectively. The first candidate reference signal SR2 includes a first candidate reference signal portion SR21 and a second candidate reference signal portion SR22 corresponding to the first and second motion portions ML11 and ML12, wherein the second candidate reference signal portion The SR 22 is adjacent to the first candidate reference signal portion SR21. The processing module 221 obtains a first motion feature value data unit DA2 from the motion parameter signal structure SP1 based on the first candidate reference signal SR2.

For example, the first motion feature value data unit DA2 includes a first motion feature value data unit portion DA21 and a second motion feature value data unit portion DA22 corresponding to the first and second motion portions ML11 and ML12, respectively. The processing module 221 obtains the first motion feature value data unit portion DA21 from the first motion parameter signal structure portion SP5 based on the first candidate reference signal portion SR21. The processing module 221 obtains the second motion feature value data unit portion DA22 from the second motion parameter signal structure portion SP6 based on the second candidate reference signal portion SR22.

In some embodiments, the processing module 221 is based on the first And the second motion feature value data unit sections DA21 and DA22 to generate a difference data unit DC1, and determine whether the difference data unit DC1 satisfies a first specific condition for a periodic motion start determination to make a The second judgment result. When the second determination result is affirmative, the processing module 221 recognizes that the valid reference signal SRP is the first candidate reference signal SR2, and determines that the first body motion ML1 meets a predetermined period motion start condition, and based on the The first candidate reference signal code CA2 determines whether the motion type indicator CH2 indicates one of the plurality of motion types HA1, HA2, ..., HA6 to make a third determination result. When the third determination result is affirmative, the processing module 221 identifies that the first motion type HM1 is the second motion type HM2 based on the motion type indicator CH2, and obtains a representative of the first motion type HM1. Motion type code CT1.

The motion parameter signal structure SP1 includes complex motion parameter signals SP11, SP12, SP13, ..., SP18 and SP19. The identification reference data unit DR further includes a representative signal code CB1 representing a representative signal SPS1 included in the complex motion parameter signals SP11, SP12, SP13, ..., SP18 and SP19. The processing module 221 determines the representative signal SPS1 based on the motion parameter signal structure SP1 and the representative signal code CB1. The processing module 221 obtains a representative pole between the first and second motion feature value data unit portions DA21 and DA22 by comparing the first and second motion feature value data unit portions DA21 and DA22. Value deviation DC11. For example, the fused signal SPS0 is the motion parameter signal SP18.

For example, the first motion feature value data unit portion The DA21 includes a first motion feature value data unit sub-portion DA21A corresponding to the representative signal SPS1, wherein the first motion feature value data unit sub-portion DA21A has a first maximum value, a first minimum value, and A first difference between the first maximum value and the first minimum value. The second motion feature value data unit portion DA22 includes a second motion feature value data unit sub-portion DA22A corresponding to the representative signal SPS1, wherein the second motion feature value data unit sub-portion DA22A has a second maximum value, a second minimum value, and a second difference between the second maximum value and the second minimum value. The processing module 221 obtains a representative value difference DC12 from the first and the second differences. The difference data unit DC1 includes the representative extreme value deviation DC11 and the representative value difference DC12. The first specific condition includes a first sub-condition and a second sub-condition. The first sub-condition is that the representative extreme value deviation DC11 falls within a first predetermined range of values. The second sub-condition is that the representative value difference DC12 falls within a second predetermined range of values.

In some embodiments, the candidate reference signal code data unit CA further includes at least one second candidate reference signal code CA1 representing at least one second candidate reference signal SR1 (and optionally a third candidate reference signal SR3) (and A third candidate reference signal code CA3) is selectable. The second candidate reference signal code CA1 represents the second candidate reference signal SR1 originating from the motion parameter signal structure SP1. The first candidate reference signal SR2 and the at least one second candidate reference signal SR1 (and optional SR3) form a candidate reference signal combination SRG.

The processing module 221 is based on the motion parameter signal structure SP1 and the at least one second candidate reference signal code CA1 (and optional CA3) to determine the at least one second candidate reference signal SR1 (and selectable SR3), and based on the at least one second candidate reference signal SR1 ( And selecting at least SR2) to obtain at least one second motion characteristic value data unit DA1 corresponding to the at least one second candidate reference signal SR1 (and optional SR3) from the motion parameter signal structure SP1 (and optionally A third motion characteristic value data unit DA3). When the processing module 221 processes the first motion feature value data unit DA2, the processing module 221 processes the at least one second motion feature value data unit DA1 (and the selectable DA3) to determine the candidate reference signal combination SRG. Whether to include the valid reference signal SRP.

In some embodiments, the second motion portion ML12 is later than the first motion portion ML11; and the first motion feature value data unit DA2 includes a plurality of specific motion feature values. For example, the specific motion feature value data unit DA8 is the first motion feature value data unit DA2. When the third determination result is negative, the processing module 221 performs the first calculation to generate the first identification value DH1 based on the first feature function code data unit F1 and the complex specific motion feature value, and determines the Whether the first identification value DH1 belongs to one of the first and second identification value areas DU1 and DU2 to make a fourth determination result. In some embodiments, one of the first motion feature value data unit DA2, the first motion feature value data unit portion DA21, and the second motion feature value data unit portion DA22 includes the complex specific motion feature value.

When the fourth determination result is positive, the processing module 221 determines that the first identification value DH1 belongs to the first and second identification values. An effective identification value area DUA of the areas DU1 and DU2 identifies that the first motion type HM1 is an effective motion type indicated by the valid identification value area DUA, and obtains the motion corresponding to the valid identification value area DUA. Type code CT1. The processing module 221 generates motion measurement information DM1 related to the first body motion ML1 based on the motion type code CT1.

In some embodiments, the memory module 222 further stores a second feature function code data unit F2 associated with the first candidate reference signal code CA2 and a second complex identification value region DU3, DU4, and DV2. The second feature function code data unit F2 is different from the first feature function code data unit F1 and is used to indicate one of the second complex identification value regions DU3, DU4 and DV2. The second complex identification value areas DU3, DU4, and DV2 include a third identification value area DU3, a fourth identification value area DU4 adjacent to the third identification value area DU3, and the third and fourth identifications. A second trust value region DV2 between the value regions DU3 and DU4. For example, the first trust value area DV1 is selectable. The third and fourth identification value regions DU3 and DU4 respectively indicate a fifth motion type H3 and a sixth motion type H4, and the fifth and sixth motion types H3 and H4 are included in the complex motion type HA1. HA2,..., HA6. For example, the memory module 222 provides the first complex identification value regions DU1, DU2, and DV1, and the second complex identification value regions DU3, DU4, and DV2 in advance.

In a second specific condition, one of the third and fourth identification value regions DU3 and DU4 and the first and second identification value regions One of the domains DU1 and DU2 at least partially overlaps. When the fourth determination result is negative, the processing module 221 performs a second calculation based on the second feature function code data unit F2 and the first motion feature value data unit DA2 to generate a second identification value DH2. It is determined whether the second identification value DH2 belongs to one of the third and fourth identification value regions DU3 and DU4, and thereby identifies the first motion type HM1.

Please refer to FIG. 3, which is a schematic diagram of a motion recognition device 40 in various embodiments of the present disclosure. Please refer to Figure 2 for additional details. As shown in FIGS. 2 and 3, the motion recognition device 40 includes a signal generating unit 41 and a processing unit 42. The signal generating unit 41 generates a motion parameter signal structure SP1 according to a first body motion ML1, wherein the first body motion ML1 occurs at a specific position PL1 on a user's body 91 and belongs to a first motion type HM1. . The processing unit 42 identifies the specific location PL1 based on the motion parameter signal structure SP1 to determine a valid reference signal SRP for identifying the first motion type HM1.

The motion recognition device 40 is configured to have a certain direction KA1, a gravity direction QF1, and an attached coordinate system UA1 for determining the orientation KA1, and is worn at the specific position PL1. The specific position PL1 is a plurality of different positions PL5, PL6, ..., PL9 that are selectively selected from the user's body 91. In some embodiments, the signal generating unit 41 is coupled to the processing unit 42 and includes a sensing unit 21 and a converting unit 411 coupled to the sensing unit 21. The sensing unit 21 generates a sensing signal SE1 according to the first body motion ML1, wherein the first body motion ML1 belongs to one of the first motion types HM1. The conversion unit 411 is coupled to the processing unit 42 and generates the motion parameter signal structure SP1 according to the sensing signal SE1.

In some embodiments, the sensing unit 21 includes an accelerometer 211 and a gyroscope 212. The accelerometer 211 and the gyroscope 212 are both coupled to the conversion unit 411. The accelerometer 211 generates an accelerometer signal SE11. The gyroscope 212 generates a gyroscope signal SE12. The sensing signal SE1 includes the accelerometer signal SE11 and the gyroscope signal SE12. The conversion unit 411 processes the accelerometer signal SE11 and the gyro signal SE12 to generate the motion parameter signal structure SP1. The motion parameter signal structure SP1 includes the accelerometer signal SE11 and a fused signal SPS0 of the gyro signal SE12. For example, the fused signal SPS0 is generated by using a signal fusion operation.

The first motion type HM1 is selected from the complex motion types HA1, HA2, ..., HA6. The complex motion types HA1, HA2, ..., HA6 are predetermined in relation to the specific position PL1, and respectively have a plurality of main motion axis directions QA1, QA2, ..., QA6 associated with the attached coordinate system UA1. The plurality of main motion axis directions QA1, QA2, ..., QA6 are pre-detected to generate a plurality of main motion axis direction data units DQ1, DQ2, ..., DQ6 corresponding to the plurality of main motion axis directions QA1, QA2, ..., QA6, respectively. . The motion parameter signal structure SP1 is generated in relation to the attached coordinate system UA1. In a state where the orientation KA1 is directed to a predetermined direction QH1 related to the gravity direction QF1, the processing unit 42 generates a specific position code CP1 representing the specific position PL1 based on the motion parameter signal structure SP1, and thereby identifying This particular location PL1. The sport The identification device 40 further includes a button 23 coupled to the processing unit 42. In the state in which the orientation KA1 is directed to the predetermined direction QH1 associated with the gravity direction QF1, the button 23 causes the processing unit 42 to receive a trigger signal ST1 in response to a user press from the user 90. The processing unit 42 generates the specific location code CP1 in response to the trigger signal ST1.

The processing unit 42 obtains an identification reference unit DR based on the specific position code CP1. The identification reference data unit DR is predetermined based on the complex main motion axis direction data units DQ1, DQ2, ..., DQ6, and includes a candidate reference signal code data unit CA and a motion type indicator data unit CH. The candidate reference signal code data unit CA includes a first candidate reference signal code CA2. The first candidate reference signal code CA2 represents a first candidate reference signal SR2 derived from the motion parameter signal structure SP1. The motion type indicator data unit CH includes a motion type indicator CH2 corresponding to the first candidate reference signal code CA2. The motion type indicator CH2 indicates one of an invalid motion type and a second motion type HM2, and the second motion type HM2 is included in the plural motion type HA1, HA2, ..., HA6. The processing unit 42 determines the first candidate reference signal SR2 based on the motion parameter signal structure SP1 and the first candidate reference signal code CA2.

In some embodiments, the first body motion ML1 includes a first motion portion ML11 and a second motion portion ML12 adjacent to the first motion portion ML11. The first and second moving portions ML11 and ML12 respectively form a first motion cycle W1 and a second motion cycle W2 adjacent to the first motion cycle W1. The processing unit 42 is based on the The first candidate reference signal SR2 obtains a first motion feature value data unit DA2 from the motion parameter signal structure SP1. The first motion feature value data unit DA2 includes a first motion feature value data unit portion DA21 and a second motion feature value data unit portion DA22 corresponding to the first and second motion portions ML11 and ML12, respectively.

The processing unit 22 generates a difference data unit DC1 based on the first and second motion feature value data unit portions DA21 and DA22. The processing unit 42 determines whether the difference data unit DC1 satisfies a specific condition for a one-cycle motion start determination to make a first determination result. When the first determination result is affirmative, the processing unit 42 recognizes that the valid reference signal SRP is the first candidate reference signal SR2, and determines whether the motion type indicator CH2 is indicated based on the first candidate reference signal code CA2. One of the plural motion types HA1, HA2, ..., HA6 to make a second judgment result. When the second determination result is affirmative, the processing unit 22 recognizes that the first motion type HM1 is the second motion type HM2 based on the motion type indicator CH2, and obtains a motion representing the first motion type HM1. Type code CT1.

In some embodiments, the second moving portion ML12 is later than the first moving portion ML11. The first motion feature value data unit DA2 includes a plurality of specific motion feature values. The processing unit 22 provides a first feature function code data unit F1 in advance. The first feature function code data unit F1 is represented based on a relationship between the complex specific motion feature values, and is related to the first candidate reference signal code CA2, a first identification value region DU1, and adjacent to the relationship a second identification value area of the first identification value area DU1 Domain DU2. The first and second identification value areas DU1 and DU2 respectively indicate a third motion type H1 and a fourth motion type H2. The third and fourth motion types H1 and H2 are all included in the plurality of motion types HA1, HA2, ..., HA6. When the second determination result is negative, the processing unit 42 performs a calculation based on the feature function code data unit F1 and the first motion feature value data unit DA2 to generate a first identification value DH1, and determines the first identification. The value DH1 belongs to a valid identification value area DUA in the first and second identification value areas DU1 and DU2, and the first motion type HM1 is identified as an effective motion type indicated by the valid identification value area DUA. And obtaining the motion type code CT1 corresponding to the valid identification value area DUA.

In various embodiments according to the first, second, and third figures, a motion recognition method includes the following steps: generating a motion parameter signal structure SP1 in response to a first body motion ML1, wherein The first body movement ML1 occurs at a specific position PL1 on a user's body 91 and belongs to a first motion type HM1; and the specific position PL1 is identified based on the motion parameter signal structure SP1 to determine the first position A valid reference signal SRP for motion type HM1.

The motion recognition method further includes a step of providing a motion recognition device 20, 30 or 40. The motion recognition device 20, 30 or 40 is configured to have a certain direction KA1, a gravity direction QF1, and an attached coordinate system UA1 for determining the orientation KA1, and is worn at the specific position PL1. The first body motion ML1 belongs to one of the first motion types HM1. The specific location PL1 is optionally selected from the user The plurality of positions PL5, PL6, ..., PL9 on the body 91 are different. The motion parameter signal structure SP1 is generated in response to the first body motion ML1. The step comprises the following substeps: generating a sensing signal SE1 associated with the body coordinate system UA1 in response to the first body motion ML1, wherein the sensing The signal SE1 includes an accelerometer signal SE11 and a gyro signal SE12; and the sensing signal SE1 is processed to generate the motion parameter signal structure SP1, wherein the motion parameter signal structure SP1 includes the accelerometer signal SE11 and the gyro signal SE12 A fusion signal SPS0. For example, the fused signal SPS0 is generated by using a signal fusion operation.

The first motion type HM1 is selected from the complex motion types HA1, HA2, ..., HA6. The complex motion types HA1, HA2, ..., HA6 are predetermined in relation to the specific position PL1, and respectively have a plurality of main motion axis directions QA1, QA2, ..., QA6 associated with the attached coordinate system UA1. The plurality of main motion axis directions QA1, QA2, ..., QA6 are pre-detected to generate a plurality of main motion axis direction data units DQ1, DQ2, ..., DQ6 corresponding to the plurality of main motion axis directions QA1, QA2, ..., QA6, respectively. . For example, the first body movement ML1 is a gymnastics practice; the first motion type HA1 is a drill type; and the main motion axis direction QA1 is a main rotation axis direction or a main translation axis direction. The motion recognition method further includes a step of generating a specific position code CP1 representing the specific position PL1 based on the motion parameter signal structure SP1 in a state in which the orientation KA1 is directed to a predetermined direction QH1 related to the gravity direction QF1. And thereby identifying the specific location PL1.

For example, the predetermined direction QH1 is based on the specific position PL1 And was determined. The predetermined direction QH1 and the gravity direction QF1 have a first angle θ. The fused signal SPS0 is a signal that estimates one of the angles SK1 associated with the first angle θ. The step of generating the specific location code CP1 includes the following sub-steps: in the state, determining whether a trigger signal ST1 is detected to make a first determination result; in this state, corresponding to the user 90 A user pressed to receive the trigger signal ST1; and when the first determination result is affirmative, the specific position code CP1 is generated based on the estimated angle SK1.

The motion recognition method further includes a step of obtaining an identification reference data unit DR based on the specific position code CP1. The identification reference data unit DR is predetermined based on the complex main motion axis direction data units DQ1, DQ2, ..., DQ6, and includes a candidate reference signal code data unit CA and a motion type indicator data unit CH. The candidate reference signal code data unit CA includes a first candidate reference signal code CA2. The first candidate reference signal code CA2 represents a first candidate reference signal SR2 derived from the motion parameter signal structure SP1. The motion type indicator data unit CH includes a motion type indicator CH2 corresponding to the first candidate reference signal code CA2. The motion type indicator CH2 indicates one of an invalid motion type and a second motion type HM2, and the second motion type HM2 is included in the plural motion type HA1, HA2, ..., HA6. The motion recognition method further includes a step of determining the first candidate reference signal SR2 based on the motion parameter signal structure SP1 and the first candidate reference signal code CA2.

The motion parameter signal structure SP1 includes a complex motion parameter Number signals SP11, SP12, SP13, ..., SP18 and SP19. After the trigger signal ST1 is provided, the first body motion ML1 includes a first motion portion ML11 and a second motion portion ML12 adjacent to the first motion portion ML11. The first and second moving portions ML11 and ML12 respectively form a first motion cycle W1 and a second motion cycle W2 adjacent to the first motion cycle W1. The motion parameter signal structure SP1 includes a first motion parameter signal structure portion SP5 and a second motion parameter signal structure portion SP6 corresponding to the first and second motion portions ML11 and ML12, respectively. The first candidate reference signal SR2 includes a first candidate reference signal portion SR21 and a second candidate reference signal portion SR22 corresponding to the first and second motion portions ML11 and ML12, wherein the second candidate reference signal portion The SR 22 is adjacent to the first candidate reference signal portion SR21. The identification reference data unit DR further includes a representative signal code CB1 representing a representative signal SPS1 included in the complex motion parameter signals SP11, SP12, SP13, ..., SP18 and SP19. For example, the fused signal SPS0 is the motion parameter signal SP18.

In some embodiments, the motion recognition method further includes a step of obtaining a first motion feature value data unit DA2 from the motion parameter signal structure SP1 based on the first candidate reference signal SR2. The first motion feature value data unit DA2 includes a first motion feature value data unit portion DA21 and a second motion feature value data unit portion DA22 corresponding to the first and second motion portions ML11 and ML12, respectively. The step of obtaining the first motion characteristic value data unit DA2 includes the following sub-steps: based on the first candidate reference signal portion SR21, the first motion parameter is received. The first motion feature value data unit portion DA21 is obtained in the number structure portion SP5; and the second motion feature value data unit portion DA22 is obtained from the second motion parameter signal structure portion SP6 based on the second candidate reference signal portion SR22. .

In some embodiments, the motion recognition method further comprises the steps of: generating a difference data unit DC1 based on the first and second motion feature value data unit portions DA21 and DA22; determining whether the difference data unit DC1 is satisfied for a first specific condition of the first motion start determination to determine a second determination result; when the second determination result is positive, the valid reference signal SRP is identified as the first candidate reference signal SR2, and the The first body motion ML1 satisfies a predetermined period motion start condition, and based on the first candidate reference signal code CA2, determines whether the motion type indicator CH2 indicates one of the plurality of motion types HA1, HA2, ..., HA6, To make a third determination result; and when the third determination result is affirmative, the processing unit 22 recognizes that the first motion type HM1 is the second motion type HM2 based on the motion type indicator CH2, and obtains A motion type code CT1 representing the first motion type HM1.

The step of generating the difference data unit DC1 includes the following substeps: obtaining the first and second motion characteristic value data unit portions DA21 by comparing the first and second motion characteristic value data unit portions DA21 and DA22 A representative extreme value deviation DC11 between the DA and the DA22; and determining the representative signal SPS1 based on the motion parameter signal structure SP1 and the representative signal code CB1. For example, the first sport The levy data unit portion DA21 includes a first motion feature value data unit sub-portion DA21A corresponding to the representative signal SPS1, wherein the first motion feature value data unit sub-portion DA21A has a first maximum value and a first minimum value. a value, and a first difference between the first maximum value and the first minimum value. The second motion feature value data unit portion DA22 includes a second motion feature value data unit sub-portion DA22A corresponding to the representative signal SPS1, wherein the second motion feature value data unit sub-portion DA22A has a second maximum value, a second minimum value, and a second difference between the second maximum value and the second minimum value. The step of generating the difference data unit DC1 further comprises a substep of obtaining a representative value difference DC12 from the first and the second differences, wherein the difference data unit DC1 comprises the representative extreme value deviation DC11 and the representative The difference in sex value is DC12.

In some embodiments, the first particular condition includes a first sub-condition and a second sub-condition. The first sub-condition is that the representative extreme value deviation DC11 falls within a first predetermined range of values. The second sub-condition is that the representative value difference DC12 falls within a second predetermined range of values. The candidate reference signal code data unit CA further includes at least one second candidate reference signal code CA1 (and optionally a third) representing at least one second candidate reference signal SR1 (and optionally a third candidate reference signal SR3). Candidate reference signal code CA3). The second candidate reference signal code CA1 represents the second candidate reference signal SR1 originating from the motion parameter signal structure SP1. The first candidate reference signal SR2 and the at least one second candidate reference signal SR1 (and optional SR3) form a candidate reference signal combination SRG.

The motion recognition method further includes the following steps: based on the operation The dynamic parameter signal structure SP1 and the at least one second candidate reference signal code CA1 (and optional CA3) determine the at least one second candidate reference signal SR1 (and selectable SR3); based on the at least one second candidate reference Signal SR1 (and optional SR3) to obtain at least one second motion characteristic value data unit DA1 corresponding to the at least one second candidate reference signal SR1 (and optional SR3) from the motion parameter signal structure SP1 (and Selectable a third motion feature value data unit DA3); and when the first motion feature value data unit DA2 is processed, processing the at least one second motion feature value data unit DA1 (and selectable DA3) to determine Whether the candidate reference signal combination SRG includes the valid reference signal SRP.

In some embodiments, the second motion portion ML12 is later than the first motion portion ML11; and the first motion feature value data unit DA2 includes a plurality of specific motion feature values. The motion recognition method further includes a step of providing a first feature function code data unit F1 in advance by sensing the plurality of body motions MC1, MC2, ..., MC9 occurring at the specific position PL1. The first feature function code data unit F1 is related to the first candidate reference signal code CA2 and a first complex identification value area DU1, DU2 and DV1.

The first feature function code data unit F1 is employed to indicate one of the first complex number value regions DU1, DU2, and DV1. The first complex identification value areas DU1, DU2, and DV1 include a first identification value area DU1, a second identification value area DU2 adjacent to the first identification value area DU1, and the first and second identifications. A first trust value region DV1 between the value regions DU1 and DU2. For example, the first trust value Area DV1 is optional. The first and second identification value areas DU1 and DU2 respectively indicate a third motion type H1 and a fourth motion type H2. The third and fourth motion types H1 and H2 are all included in the plurality of motion types HA1, HA2, ..., HA6. The plurality of body movements MC1, MC2, ..., MC9 are divided into a plurality of motion groups G1, G2, ..., G6, and the plurality of motion groups G1, G2, ..., G6 belong to the plurality of motion types HA1, HA2, ..., respectively HA6. The first feature function code data unit F1 represents a motion type cutting function and is represented based on a relationship between the complex specific motion feature values, wherein the motion type cutting function is employed to indicate the first complex identification value region One of DU1, DU2, and DV1.

The motion recognition method further includes the following steps: when the third determination result is negative, performing a first calculation based on the first feature function code data unit F1 and the first motion feature value data unit DA2 to generate a first Identifying the value DH1, and determining whether the first identification value DH1 belongs to one of the first and second identification value areas DU1 and DU2 to make a fourth determination result; when the fourth determination result is positive Determining that the first identification value DH1 belongs to a valid identification value area DUA in the first and second identification value areas DU1 and DU2, and identifying that the first motion type HM1 is indicated by the valid identification value area DUA An effective motion type, and obtaining the motion type code CT1 corresponding to the valid identification value region DUA; generating motion measurement information DM1 related to the first body motion ML1 based on the motion type code CT1; and providing a first Two feature function code data unit F2.

For example, the second feature function code data unit F2 is related to the first candidate reference signal code CA2 and a second complex number value region DU3, DU4 and DV2. The second feature function code data unit F2 is different from the first feature function code data unit F1 and is used to indicate one of the second complex identification value regions DU3, DU4 and DV2. The second complex identification value areas DU3, DU4, and DV2 include a third identification value area DU3, a fourth identification value area DU4 adjacent to the third identification value area DU3, and the third and fourth identifications. A second trust value region DV2 between the value regions DU3 and DU4. For example, the second trust value region DV2 is selectable. The third and fourth identification value regions DU3 and DU4 respectively indicate a fifth motion type H3 and a sixth motion type H4, and the fifth and sixth motion types H3 and H4 are included in the complex motion type HA1. HA2,..., HA6.

In a second specific condition, one of the third and fourth identification value regions DU3 and DU4 at least partially overlaps with one of the first and second identification value regions DU1 and DU2. The motion recognition method further includes a step of: performing a second calculation to generate a second based on the second feature function code data unit F2 and the first motion feature value data unit DA2 when the fourth determination result is negative The identification value DH2 is determined to determine whether the second identification value DH2 belongs to one of the third and fourth identification value regions DU3 and DU4, and thereby the first motion type HM1 is recognized.

In some embodiments, the motion recognition devices 20, 30, and 40 each have a wearable motion sensing device structure, a motion recognition operating device, and a motion recognition algorithm, and include a user wear device. Techniques for setting, user operating devices, measuring devices, and motion recognition devices. The motion recognition devices 20 and 30 each include an operating unit 35 and a coupling unit 36 coupled to the operating unit 35. The operating unit 35 includes the sensing unit 21 and the processing unit 22. The motion recognition device 40 includes an operating unit 37 and a coupling unit 36 coupled to the operating unit 37. The operation unit 37 includes the signal generating unit 41 and the processing unit 42 , wherein the signal generating unit 41 includes the sensing unit 21 . Any of the motion recognition devices 20, 30, and 40 can function as a wearable device and be coupled to the particular body portion 911 of the user 90 by using the coupling unit 36. For example, the coupling unit 36 is a wear assembly. For example, the coupling unit 36 is a type of strap.

In some embodiments, the wear assembly has at least two stable states, and the wear assembly uses the wear assembly to couple the wear device to the particular body portion 911 of the user 90. The sensing unit 21 of the wearing device can be a motion sensing module. The wear assembly can have a deformation such that the wear assembly has an extended steady state and a curled stable state. When the specific body portion 911 making the first body motion ML1 is at least partially surrounded by the wearing assembly, the wearing assembly is deformed to be in the curled stable state and is fixed to the specific body portion 911. . The wearing assembly has a stable quality in the extended steady state; and in the extended stable state, the operating unit 35 or 37 is placed to the wearing assembly.

Please refer to Table 1, which is a listing of various types of body movements in various embodiments of the present disclosure. In some embodiments, the wearables The position is worn at the specific position PL1 or worn to the specific body part 911. The particular location PL1 is optionally selected from the plurality of different locations PL5, PL6, ..., PL9 on the user's body 91. The plurality of different positions PL5, PL6, ..., PL9 include the wrist position, the upper arm position, and the ankle position. The wear device is worn to the particular body portion 911 with the wearer assembly, the particular body portion 911 being selected from a plurality of body portions. The plurality of body parts include the wrist, the upper arm and the ankle, so the wearing device can be used to sense body movement (such as fitness) related to at least one of the following body parts: palm, lower arm, upper arm, Shoulder, chest, back, waist, hips, thighs, knees, calves, soles, etc. For example, the plural physical exercise types are listed in Table 1.

In some embodiments, the user 90 typically exercises the body motion to exercise body muscles by operating an exercise device or load, wherein the load includes a dumbbell, a barbell, or the like. The most commonly used exercises include wrist movements, upper arm movements, and leg movements, and are classified as Various types of body movements as shown in Table 1 were formed. The user 90 performs the exercises to exercise and strengthen the muscles in the plurality of body parts including the hand, the upper arm, the foot, the chest, the back, the shoulder, the waist, and the buttocks to achieve exercise. The purpose of health and body sculpting. When the user 90 makes physical movements in the body parts of the hands, legs, etc., the related wrists, upper arms and ankles and the like have significant athletic performance, so the body movement is accurately detected. Considering the operation, as well as the comfort of wearing the human body position and the natural acceptance of long-wearing habits, the wrist position, the upper arm position and the ankle position are all detectable movements. The motion recognition device 20 is most suitable for wearing. s position.

Please refer to FIGS. 4A, 4B and 4C, which are schematic diagrams of the types of body motion recognized by the motion recognition device 20 worn on the wrist. Please refer to FIG. 5, which is a schematic diagram of the type of body motion recognized by the motion recognition device 20 worn on the upper arm. Please refer to FIG. 6, which is a schematic diagram of the type of body motion recognized by the motion recognition device 20 worn on the upper arm. Please refer to FIGS. 7A and 7B, which are schematic diagrams of the types of body motion recognized by the motion recognition device 20 worn on the upper arm. Please refer to Fig. 8, which is a schematic diagram of the type of body motion recognized by the motion recognition device 20 worn on the upper arm. Please refer to FIGS. 9A and 9B, which are schematic diagrams of the types of body motion recognized by the motion recognition device 20 worn on the ankle. Please refer to FIGS. 10A and 10B, which are schematic diagrams of the types of body motion recognized by the motion recognition device 20 worn on the ankle.

In some embodiments, the sensing unit 21 has a sense Sensing Orientation. The sensing unit 21 or the motion sensing module (Motion Sensing Module) has a complex motion sensing axis direction and a complex sensing sensitivity respectively in the direction of the complex motion sensing axis, and is sensed on the user's body 91. The first body motion ML1 that occurs at the particular location PL1. The user body 91 has the specific body portion 911 at the specific position PL1. The motion sensing module can be applied to sense different motions that occur on different body parts, such as the wrist, the upper arm, and the ankle. When the motion sensing module senses a specific motion occurring at the specific position PL1 on the user's body 91, the motion sensing module has a motion sensing axis direction corresponding to the specific position PL1. A maximum sensing sensitivity. For example, when sensing a wrist motion, the motion sensing module has a highest sensing sensitivity in the x motion sensing axis direction; therefore, the sensing signal component level generated in the x motion sensing axis direction Higher than that produced in the direction of the y motion sensing axis or the z motion sensing axis direction. In some embodiments, the motion sensing module outputs a first sensing signal component, a first in the x motion sensing axis direction, the y motion sensing axis direction, and the z motion sensing axis direction. The second sensing signal component and a third sensing signal component. For example, the maximum levels of the first, second, and third sensing signal components, and the level ranges of the first, second, and third sensing signal components are used to identify the first The body movement ML1 or the specific body part 911.

In some embodiments, the sensing unit 21 or the motion sensing module has an Arrangement Orientation. When the specific position PL1 is changed, the setting orientation of the sensing unit 21 is changed. change. When the specific position PL1 is changed to a different position, the set orientation can be changed to a different set orientation. For example, in a specific state, when the motion recognition device 20 determines that the z motion sensing axis direction is the same as the gravity direction of the motion recognition device 20, the motion recognition device 20 recognizes the specific position PL1. It is the wrist position and it is necessary to recognize a wrist movement. In the specific state, when the motion recognition device 20 makes a determination result that the x motion sensing axis direction is the same as the gravity direction of the motion recognition device 20, the motion recognition device 20 recognizes that the specific position PL1 is The upper arm position is to recognize an upper arm movement. In the specific state, when the motion recognition device 20 makes a determination that the negative x motion sensing axis direction is the same as the gravity direction of the motion recognition device 20, the motion recognition device 20 recognizes that the specific position PL1 is The ankle position and the identification of an ankle movement.

In some embodiments, the set orientation of the motion sensing module is set based on the particular location PL1. In some embodiments, the motion recognition device 20 detects an angular relationship between a specific motion sensing axis direction of the motion sensing module and the gravity direction to identify that the specific position PL1 is a body part position. . For example, the motion recognition device 20 is configured to have a certain direction KA1, a gravity direction QF1, and an attached coordinate system UA1 for determining the orientation KA1, and is worn at the specific position PL1. In a state where the orientation KA1 is directed to a predetermined direction QH1 related to the gravity direction QF1, the processing unit 22 generates a specific position code CP1 representing the specific position PL1 based on the motion parameter signal structure SP1, and thereby identifying This particular location PL1.

In some embodiments, any of the motion recognition devices 20, 30, and 40 are worn as a wearable motion sensing device at a particular location PL1 on a user's body 91 and are sensed by A first body motion ML1 occurring at the specific position PL1 generates a sensing signal SE1, wherein the first body motion ML1 belongs to a first motion type HM1, and the specific position PL1 may be a selected multiple position, the plural Different positions include a wrist position, an upper arm position, and an ankle position. The wearable motion sensing device generates a specific position code CP1 representing the specific position PL1 based on the motion parameter signal structure SP1, and determines a first candidate reference signal SR2 based on the motion parameter signal structure SP1 and the specific position code CP1. Obtaining a first motion feature value data unit DA2 from the motion parameter signal structure SP1 based on the first candidate reference signal SR2, and identifying the first motion type HM1 based on the first motion feature value data unit DA2. A motion type code CT1 representing the first motion type HM1, so the wearable motion sensing device recognizes a direction of motion of the first body motion ML1 and counts the motion cycle of the first body motion ML1.

The wearable motion sensing device recognizes the first motion type HM1 by performing a first motion recognition calculation program. The sensing unit 21 includes a six-axis inertial sensing element. For example, the sensing unit 21 includes an accelerometer 211 that is a three-axis accelerometer, and a gyroscope 212 that is a three-axis gyroscope. The wearable motion sensing device generates the sensing signal SE1 by using the sensing unit 21 in response to the first body motion ML1 of the exercise, and performs the first based on the sensing signal SE1. Motion recognition The program automatically recognizes the specific position PL1 and the first motion type HM1. For example, the user body 91 has a specific body portion 911 at the specific position PL1, and any one of the motion recognition devices 20, 30, and 40 is worn on the specific body portion 911.

The first motion recognition calculation program includes sensing signal processing, wearing position recognition, and motion type identification. The motion type identification includes the following features to determine the first candidate reference signal SR2, obtain the first motion feature value data unit DA2, determine whether the first body motion ML1 meets a one-cycle motion start condition or a one-cycle motion termination condition, And based on the first feature function code data unit F1 to perform a calculation to identify the first motion type HM1. For example, the processing unit 22 executes the first motion recognition algorithm. For example, the conversion unit 411 and the processing unit 42 execute the first motion recognition calculation program. In the following description, the processing unit 22 executes the first motion recognition calculation program as an example.

Please refer to FIG. 11, which is a schematic diagram of a reference coordinate system in various embodiments of the present disclosure. Please refer to Figure 1 for additional details. Figure 11 shows a reference coordinate system having three reference coordinate axes X, Y and Z, and the attached coordinate system UA1 having three reference coordinate axes x, y and z. The attached coordinate system UA1 is represented in relation to the gravity direction QF1. The processing unit 22 receives the sensing signal SE1, and the sensing signal SE1 is generated in relation to the attached coordinate system UA1.

The accelerometer 211 of the sensing unit 21 generates the accelerometer signal SE11 (or a first signal) ). The gyroscope 212 of the sensing unit 21 generates the gyroscope signal SE12 (or a second signal) ). The sensing signal SE1 includes the first signal And the second signal . The first signal Containing three original sensed signal components associated with the three reference axes x, y, and z, respectively , , . The second signal Containing three original sensed signal components associated with the three reference axes x, y, and z, respectively , , . The first signal Contains a gravitational acceleration signal component . The processing unit 22 obtains the motion parameter signal structure SP1 by using a six-axis fusion algorithm (or a signal fusion operation), and the motion parameter signal structure SP1 includes a quaternion (q ₀ , q ₁ , q ₂ , q _{3 ).} ), an estimated roll angle φ _Roll associated with a Euler angle, an estimated pitch angle θ _Pitch associated with a Euler angle, and a third signal And a fourth signal . For example, the processing unit 22 is for the first signal Performing a first filtering to generate the third signal And for the second signal Performing a second filtering to generate the fourth signal .

The estimation of the roll angle φ associated with the attached coordinate system UA1 can be expressed as φ _Roll =tan ^-1 (g _y /g _z )

The estimate of the pitch angle θ associated with the attached coordinate system UA1 can be expressed as

The quaternion (q ₀ , q ₁ , q ₂ , q ₃ ) can be expressed as the following equation

q ₀ =cos(μ/2);q ₁ =λ ₁ sin(μ/2);q ₂ =λ ₂ sin(μ/2);q ₃ =λ ₃ sin(μ/2)

The quaternion (q ₀ , q ₁ , q ₂ , q ₃ ) should satisfy its normalization constant q ₀ ² +q ₁ ² +q ₂ ² +q ₃ ² =1

When the object continuously changes, the quaternion (q ₀ , q ₁ , q ₂ , q ₃ ) and an attached angular velocity satisfy the following relationship:

When the motion recognition device 20 makes a rotation in the order of yaw → pitch → roll, then from the quaternion (q ₀ , q ₁ , q ₂ , q ₃ ) to the estimated Euler angle (including one The conversion relationship between the estimated roll angle φ _Roll , an estimated pitch angle θ _Pitch , and an estimated yaw angle ψ _Yaw is listed as follows:

When the motion recognizing device 20 makes the rotation, the conversion relationship from the estimated Euler angles to the quaternions (q ₀ , q ₁ , q ₂ , q ₃ ) is listed as follows: q ₀ =cos( Φ/2)cos(θ/2)cos(ψ/2)+sin(φ/2)sin(θ/2)sin(ψ/2) q ₁ =sin(φ/2)cos(θ/2) Cos(ψ/2)-cos(φ/2)sin(θ/2)sin(ψ/2) q ₂ =cos(φ/2)sin(θ/2)cos(ψ/2)+sin(φ /2)cos(θ/2)sin(ψ/2) q ₃ =cos(φ/2)cos(θ/2)sin(ψ/2)-sin(φ/2)sin(θ/2)cos (ψ/2) Equation 5

The processing unit 22 obtains the motion parameter signal structure SP1 by substituting Equations 1 to 5 by using a special fusion algorithm architecture, such as an Amplified Kalman Filter (EKF). That is, the processing unit 22 processes the sensing signal SE1 to generate the motion parameter signal structure SP1. The motion parameter of a signal structure comprising a plurality of motion parameter signals SP1 SP11, SP12, SP13, ..., SP18, SP19; and motion parameters of the plurality of signals SP11, SP12, SP13, ..., SP 1 8, SP 1 9 including the third signal The fourth signal a signal of the estimated roll angle φ _Roll , a signal of the estimated pitch angle θ _Pitch , and a signal of the quaternion (q ₀ , q ₁ , q ₂ , q ₃ ). For example, the motion parameter signal structure SP1 is generated by using the signal fusion operation.

Please refer to FIG. 12, which is a schematic diagram showing the relationship between the motion recognition device 20 and the user's body 91 in FIG. According to the muscle group mainly trained by the body movement, the plural body motion type is divided into a plurality of body motion type groups corresponding to the plural positions PL5, PL6, ..., PL9, respectively. For example, the three positions PL5, PL6, and PL9 are respectively predetermined as the wrist position, the upper arm position, and the ankle position. In a particular state, when the motion recognition device 20 is worn at the position PL5, the orientation KA1 of the motion recognition device 20 is directed to a first predetermined direction; in a particular state, when the motion recognition device 20 When the position PL6 is worn, the orientation KA1 of the motion recognition device 20 is directed to a second predetermined direction; and in a specific state, when the motion recognition device 20 is worn at the position PL9, the motion The orientation KA1 of the identification device 20 is directed to a third predetermined direction.

The first, the second, and the third predetermined direction are different according to the gravity direction QF1, and may be determined according to a design configuration. The design configuration is provided to best suit the operating context, habits, and ergonomics of the user 90. The specific position PL1 is the plurality of different positions PL5, PL6, ..., PL9 selected from the user's body 91. For the first body motion ML1 occurring at the specific position PL1, in a state where the orientation KA1 is directed to a predetermined direction QH1 related to the gravity direction QF1, the processing unit 22 generates a representative based on the motion parameter signal structure SP1. A specific position code CP1 of the specific position PL1, and thereby identifying the specific position PL1.

In Fig. 12, the motion recognition device 20 has the attached coordinate system UA1 for determining the orientation KA1. When the motion recognition device 20 is worn at the position PL5 (the wrist position), the three reference axes x, y and z of the attached coordinate system UA1 use three reference axes x ^W , y ^W and z respectively ^W is indicated. When the motion recognition device 20 is worn at the position PL6 (the upper arm position), the three reference axis axes x, y and z of the body coordinate system UA1 use three reference axes x ^A , y ^{A , respectively} And z ^A is expressed. When the motion recognition device 20 is worn at the position PL9 (the ankle position), the three reference axes x, y and z of the attached coordinate system UA1 use three reference axes x ^K , y ^K and z respectively ^K is indicated.

For example, the motion recognition device 20 has a gravitational acceleration gF having a gravitational acceleration direction that is the same as the gravitational direction QF1. The third signal Has a gravitational acceleration signal component The gravity acceleration signal component Is related to the gravitational acceleration gF or the gravitational acceleration signal component And is a three-axis signal. The processing unit 22 regards the gravity direction QF1 as the main reference axis direction, and mainly uses the gravity acceleration signal component. To get the specific location code CP1. When the processing unit 22 additionally adopts the estimated pitch angle θ _Pitch , the estimated roll angle φ _Roll , and the quaternion (q ₀ , q ₁ , q ₂ , q ₃ ) generated by the six-axis signal fusion operation. In at least one of the cases, the processing unit 22 can improve the recognition accuracy for the specific position PL1. For example, the predetermined direction QH1 (or the reference axis of the attached coordinate system UA1) and the gravity direction QF1 have the first angle θ (or a pitch angle).

As shown in Fig. 12, the user's body 91 has a specific body portion 911 that makes the first body motion ML1 at the specific position PL1. For example, the particular body portion 911 can be the wrist, the upper arm, or the ankle. In the state in which the orientation KA1 is directed to the predetermined direction QH1 related to the gravity direction QF1, the button 23 causes the processing unit 22 to receive a trigger signal ST1 in response to a user press from the user 90. The processing unit 22 determines whether the processing unit 22 detects the trigger signal ST1 to make a first determination result. For example, when the first determination result is affirmative, the processing unit 22 is based on the gravity acceleration signal component. Or the signal of the pitch angle θ _Pitch is estimated to obtain the specific position code CP1. For example, the signal of the estimated angle SK1 is the signal of the estimated pitch angle θ _Pitch .

The processing unit 22 obtains an identification reference data unit DR based on the specific position code CP1. The identification reference data unit DR is predetermined based on the complex main motion axis direction data units DQ1, DQ2, ..., DQ6, and includes a candidate reference signal code data unit CA and a motion type indicator data unit CH. Candidate reference signal code data sheet The meta CA includes a first candidate reference signal code CA2. The first candidate reference signal code CA2 represents a first candidate reference signal SR2 derived from the motion parameter signal structure SP1. The motion type indicator data unit CH includes a motion type indicator CH2 corresponding to the first candidate reference signal code CA2. The motion type indicator CH2 indicates one of an invalid motion type and a second motion type HM2, and the second motion type HM2 is included in the plural motion type HA1, HA2, ..., HA6. The processing unit 22 determines the first candidate reference signal SR2 based on the motion parameter signal structure SP1 and the first candidate reference signal code CA2.

In some embodiments, according to ergonomic principles, the complex identification reference units respectively corresponding to the complex different positions PL5, PL6, . . . , PL9 are different, wherein the complex identification reference unit includes the identification reference unit DR . The processing unit 22 determines at least one candidate reference signal based on the motion parameter signal structure SP1 and the identification reference data unit DR, wherein the at least one candidate reference signal may include a valid reference signal. For example, the processing unit 22 is based on the third signal The fourth signal The signal of the estimated roll angle φ _Roll , the signal of the estimated pitch angle θ _Pitch , the signal of the quaternion (q ₀ , q ₁ , q ₂ , q ₃ ), and the candidate reference signal code data unit The CA determines to establish the following candidate reference signals: Ref _i ( , , φ _Roll , θ _Pitch , (q ₀ , q ₁ , q ₂ , q ₃ )), where the symbol i is equal to one of 1 to 3, i=1 represents the wrist position, i=2 represents the upper arm position, and i=3 represents the position of the ankle.

In some embodiments, the first candidate reference signal SR2 represented by the first candidate reference signal code CA2 is the candidate reference signals Ref _i ( , One of φ _Roll , θ _Pitch , (q ₀ , q ₁ , q ₂ , q ₃ )). The processing unit 22 obtains a first motion feature value data unit DA2 from the motion parameter signal structure SP1 based on the first candidate reference signal SR2. For example, the first body motion ML1 is a back and forth motion and includes a first motion portion ML11 forming a first motion cycle W1. Is the third signal of the three-axis signal Having a first signal portion corresponding to the first motion portion ML11; and the fourth signal that is a three-axis signal There is a second signal portion corresponding to the first moving portion ML11. The processing unit 22 is based on the candidate reference signals Ref _i ( , One of φ _Roll , θ _Pitch , (q ₀ , q ₁ , q ₂ , q ₃ )) to obtain a complex eigenvalue having a number N from at least the first and second signal portions, wherein N is greater than 1. For example, the N feature values include the third signal The fourth signal The signal of the estimated roll angle φ _Roll , the signal of the estimated pitch angle θ _Pitch , and the maximum value and each minimum value of the signal of the quaternion (q ₀ , q ₁ , q ₂ , q ₃ ) Etc., and is expressed as the eigenvalue C _i~N : C _i~N ( , , φ _Roll , θ _Pitch , (q ₀ , q ₁ , q ₂ , q ₃ )).

Please refer to FIG. 13 , which is a schematic diagram of the first candidate reference signal SR2 and the representative signal SPS1 of the motion recognition device 20 in FIG. 1 . In some embodiments, the processing unit 22 determines whether the first body motion ML1 is effectively initiated or effectively ends based on the first candidate reference signal SR2. After the trigger signal ST1 is received, the first body motion ML1 includes a first motion portion ML11 and a second motion portion ML12 adjacent to the first motion portion ML11. The first and second moving portions ML11 and ML12 respectively form a first motion cycle W1 and a second motion cycle W2 adjacent to the first motion cycle W1. The motion parameter signal structure SP1 includes a first motion parameter signal structure portion SP5 and a second motion parameter signal structure portion SP6 corresponding to the first and second motion portions ML11 and ML12, respectively. The first candidate reference signal SR2 includes a first candidate reference signal portion SR21 and a second candidate reference signal portion SR22 corresponding to the first and second motion portions ML11 and ML12, wherein the second candidate reference signal portion The SR 22 is adjacent to the first candidate reference signal portion SR21.

The first motion feature value data unit DA2 includes a first motion feature value data unit portion DA21 and a second motion feature value data unit portion DA22 corresponding to the first and second motion portions ML11 and ML12, respectively. The processing unit 22 obtains the first motion feature value data unit portion DA21 from the first motion parameter signal structure portion SP5 based on the first candidate reference signal portion SR21. The processing unit 22 obtains the second motion feature value data unit portion DA22 from the second motion parameter signal structure portion SP6 based on the second candidate reference signal portion SR22. The processing unit 22 determines whether the first body motion ML1 has been effectively initiated based on the first and second motion feature value data unit portions DA21 and DA22. For example, the processing unit 22 generates a difference data unit DC1 based on the first and second motion feature value data unit portions DA21 and DA22, and determines whether the difference data unit DC1 satisfies one for a periodic motion start determination. The first specific condition is to make a second judgment result. When the second determination result is affirmative, the processing unit 22 recognizes that the valid reference signal SRP is the first candidate reference signal SR2, and determines that the first body motion ML1 satisfies a predetermined period motion start condition. E.g, The second motion portion ML12 is later than the first motion portion ML11; and the second motion feature value data unit portion DA22 includes a plurality of specific motion feature values.

In the prior art, the judgment of each of the start of the periodic motion and the end of the periodic motion generally uses "activity stationary" as a reference index. However, human physical activity is too numerous to enumerate. If the "activity is still" is used as the judgment condition, it is easy to cause the purpose of the non-user or the sports product, and it is difficult to avoid misjudgment. In some embodiments, the processing unit 22 separately obtains the first complex specific feature value C _M1,i~n and the second complex specific feature value C on the adjacent first and second motion cycles W1 and W2. _{M2, i~n} , and performing a comparison to obtain a difference between the first and second complex specific eigenvalues C _M1,i~n and C _M2,i~n . For example, the first motion feature value data unit portion DA21 includes the complex specific feature value C _M1,i~n , and the second motion feature value data unit portion DA22 includes the complex specific feature value C _M2,i~n . When the first and second complex specific characteristic values C _M1,i~n and C _M2,i~n have eigenvalue differences ΔC _M2~M1,i~n respectively fall within the corresponding threshold range ξ _i~n Internally, the processing unit 22 determines that the first and second plurality of specific feature values C _M1,i~n and C _M2,i~n can be applied to a primary recognition function for performing the first body motion A motion type identification of ML1. After completing the motion type identification for the first body motion ML1, the processing unit 22 determines the first when an additionally obtained specific feature value difference falls outside a corresponding threshold range ξ _i~n Body movement ML1 has been terminated.

As shown in FIG. 13, the first body motion ML1 has the first motion portion ML11 between time T21 and time T22; The first body motion ML1 has the second motion portion ML12 between the time T22 and the time T23. Between the time T21 and the time T22, the first moving portion ML11 forms the first motion cycle W1; and between the time T22 and the time T23, the second moving portion ML12 forms the second motion cycle W2 . The first candidate reference signal portion SR21 corresponding to the first motion portion ML11 forms a first signal loop between the time T21 and the time T22; and the second candidate reference corresponding to the second motion portion ML12 The signal portion SR22 forms a second signal loop between the time T22 and the time T23.

For example, the representative signal SPS1 is the motion parameter signal SP13 in the motion parameter signal structure SP1. The representative signal SPS1 has a first motion parameter signal portion SPS11 between the time T21 and the time T22, and has a second motion parameter signal portion SPS12 between the time T22 and the time T23. The first motion parameter signal structure portion SP5 includes the first motion parameter signal portion SPS11, and the second motion parameter signal structure portion SP6 includes the second motion parameter signal portion SPS12.

The first motion feature value data unit portion DA21 includes a first motion feature value data unit sub-portion DA21A corresponding to the representative signal SPS1 (such as the motion parameter signal SP13). The second motion feature value data unit portion DA22 includes a second motion feature value data unit sub-portion DA22A corresponding to the representative signal SPS1 (such as the motion parameter signal SP13). The processing unit 22 obtains the first transport from the first motion parameter signal portion SPS11 based on the first candidate reference signal portion SR21. The moving feature value data unit sub-portion DA21A obtains the second motion feature value data unit sub-portion DA22A from the second motion parameter signal portion SPS12 based on the second candidate reference signal portion SR22.

The first motion feature value data unit sub-section DA21A includes the motion feature value a1 at time T61 and the motion feature value a2 at time T62. The second motion feature value data unit sub-section DA22A includes the motion feature value a3 at time T63 and the motion feature value a4 at time T64. For example, the motion feature values a1 and a3 are both maximum values, and the motion feature values a2 and a4 are all minimum values. The first motion characteristic value data unit sub-section DA21A has a first maximum value b1, a first minimum value b2, and a first difference e1 between the first maximum value b1 and the first minimum value b2. The second motion characteristic value data unit sub-section DA22A has a second maximum value b3, a second minimum value b4, and a second difference e2 between the second maximum value b3 and the second minimum value b4. For example, the first and second maximum values b1 and b3 are the motion feature values a1 and a3, respectively; and the first and second minimum values b2 and b4 are the motion feature values a2 and a4, respectively.

In some embodiments, the processing unit 22 obtains between the first and second motion feature value data unit portions DA21 and DA22 by comparing the first and second motion feature value data unit portions DA21 and DA22. A representative extreme deviation DC11. For example, the representative extreme value deviation DC11 is selected from the absolute value |a2-a1|, ..., and the absolute value |a4-a3|. The processing unit 22 obtains a representative value difference DC12 from the first and the second differences e1 and e2. The difference data unit DC1 package The representative extreme value deviation DC11 and the representative value difference DC12 are included. The first specific condition includes a first sub-condition and a second sub-condition. The first sub-condition is that the representative extreme value deviation DC11 falls within a first predetermined range of values. The second sub-condition is that the representative value difference DC12 falls within a second predetermined range of values.

Please refer to FIG. 14, which is a schematic diagram of the motion type identification of the first body motion ML1 in various embodiments of the present disclosure. In some embodiments, the processing unit 22 provides the first feature function code data unit F1 associated with the first candidate reference signal code CA2 and a first complex identification value region DU1, DU2, and DV1 in advance. The first feature function code data unit F1 is employed to indicate one of the first complex number value regions DU1, DU2, and DV1. The first complex identification value areas DU1, DU2, and DV1 include a first identification value area DU1, a second identification value area DU2 adjacent to the first identification value area DU1, and the first and second identifications. A first trust value region DV1 between the value regions DU1 and DU2. For example, the first trust value area DV1 is selectable. The first and second identification value areas DU1 and DU2 indicate the third motion type H1 and the fourth motion type H2, respectively. For example, the first trust value region DV1 has a trust value distance σ1 between the first and second identification value regions DU1 and DU2. The first feature function code data unit F1 represents a motion type cutting function and is represented based on a relationship between the complex specific motion feature values, wherein the motion type cutting function is employed to indicate the first complex identification value region One of DU1, DU2, and DV1.

Whether the difference data unit DC1 satisfies the first After the second determination result of a specific condition is affirmative, the processing unit 22 performs a first calculation based on the first feature function code data unit F1 and the first motion feature value data unit DA2 to generate a first identification value. DH1, determining that the first identification value DH1 belongs to a valid identification value area DUA in the first and second identification value areas DU1 and DU2, and identifying that the first motion type HM1 is determined by the effective identification value area DUA An effective motion type is indicated, and the motion type code CT1 corresponding to the valid identification value region DUA is obtained. The processing unit 22 generates motion measurement information DM1 related to the first body motion ML1 based on the motion type code CT1.

In some embodiments, the processing unit 22 provides in advance a set of feature function code data units F1, F2, etc. representing a set of motion type cutting functions P _{1 - K} , respectively, the set of feature function code data units F1, F2, etc. It is established by calculating a large amount of data in the database. For example, the processing unit 22 performs a calculation to effectively identify the first motion based on the third complex specific feature value, and at least one of the set of feature function code data units F1, F2, and the like. Type HM1, wherein the third complex specific feature value is selected from the first complex specific feature value C _M1,i~n and the second complex specific feature value C _M2,i~n . The set of motion type cutting functions P _1~K has a set of cutting planes and a number of levels K, wherein the number of levels K is a natural number. When the number of levels K is greater than 1, the set of motion type cutting functions P _{1 - K} are arranged based on a predetermined order and are at least partially employed to identify the first motion type HM1. For example, the set of motion type cutting functions P _1~K includes linear or non-linear multi-dimensional feature functions based on a first complex motion eigenvalue And a second complex motion eigenvalue And trained, where L represents the number of records in the database for establishing the motion feature values of the set of motion type cutting functions P _1~K . For example, the set of motion type cutting functions P _1~K are expressed as follows: , i=1~K, where σ represents a threshold distance of a trust value region; for example, the trust value distance σ1 of the first trust value region DV1 is equal to 2σ.

As shown in FIG. 14, the first and second identification value regions DU1 and DU2 respectively indicate the third motion type H1 and the fourth motion type H2, and are both two-dimensional identification value regions. The motion type cutting function P ₁ has a cutting plane V1 that separates the first and second identification value regions DU1 and DU2. The motion type cutting function P ₂ has a cutting plane V2 that separates the third and fourth identification value regions DU3 and DU4. For example, the first and second feature function code data units F1 and F2 represent the motion type cutting functions P ₁ and P _{2 , respectively} .

Please refer to FIG. 15 , which is a flow chart of a first motion type identification calculation program 110 in various embodiments of the present disclosure. In step 1102, the sensing unit 21 generates a sensing signal SE1 according to a first body motion ML1 occurring at a specific position PL1 on a user's body 91. The first body motion ML1 belongs to a first motion type HM1; and the sensing signal SE1 is generated in relation to an attached coordinate system UA1 of the sensing unit 21.

In step 1104, the processing unit 22 or the conversion unit 411 processes the sensing signal SE1 to generate a motion parameter signal structure SP1, wherein the sensing signal SE1 includes an accelerometer signal SE11 and a gyroscope signal SE12, and the The motion parameter signal structure SP1 can include the accelerometer signal SE11 and a fused signal SPS0 of the gyro signal SE12.

In step 1106, the processing unit 22 determines whether the processing unit 22 detects a trigger signal ST1 to make a first determination result. When the first determination result is negative, the step subsequent to step 1106 returns to step 1106. When the first determination result is affirmative, the step subsequent to step 1106 is step 1108.

In step 1108, the processing unit 22 generates a specific location code CP1 representative of the particular location PL1 based on the motion parameter signal structure SP1. The user body 91 is worn with the motion recognition device 20 at the specific position PL1, wherein the motion recognition device 20 includes the attached coordinate system UA1, the sensing unit 21, and the processing unit 22.

In step 1110, the processing unit 22 obtains an identification reference unit DR based on the specific location code CP1. The identification reference data unit DR is predetermined based on the attached coordinate system UA1 and a main motion axis direction of the first motion type HM1, and includes a first candidate reference signal code CA2. The first candidate reference signal code CA2 represents a first candidate reference signal SR2 derived from the motion parameter signal structure SP1. In step 1110, the processing unit 22 determines the first candidate parameter based on the motion parameter signal structure SP1 and the first candidate reference signal code CA2. Test number SR2.

In step 1112, the processing unit 22 obtains a first motion feature value data unit DA2 from the motion parameter signal structure SP1 based on the first candidate reference signal SR2. The first body motion ML1 includes a first motion portion ML11 and a second motion portion ML12 adjacent to the first motion portion ML11. The first and second moving portions ML11 and ML12 respectively form a first motion cycle W1 and a second motion cycle W2 adjacent to the first motion cycle W1. The first motion feature value data unit DA2 includes a first motion feature value data unit portion DA21 and a second motion feature value data unit portion DA21 respectively corresponding to the first motion portion ML11 and the second motion portion ML12, respectively.

In step 1114, the processing unit 22 generates a difference data unit DC1 based on the first and second motion feature value data unit portions DA21 and DA21, and uses the difference data unit DC1 to determine whether the first body motion ML1 is Has effectively ended or started. The difference data unit DC1 includes a representative extreme value deviation DC11 which is generated by comparing the first and second motion characteristic value data unit portions DA21 and DA22.

In step 1116, the processing unit 22 determines whether the difference data unit DC1 satisfies a first specific condition for a periodic motion start determination to make a second determination result. When the second determination is negative, the step subsequent to step 1116 returns to step 1118. When the second determination result is affirmative, the step subsequent to step 1116 is step 1122.

In step 1118, the processing unit 22 determines the difference. Whether the data unit DC1 satisfies a second specific condition for the one-cycle motion end judgment to make a third judgment result. When the third determination result is negative, the step subsequent to step 1118 returns to step 1120. When the third determination result is affirmative, the step subsequent to step 1118 returns to step 1110.

In step 1120, the processing unit 22 determines whether the number of current negative determination results of the third determination result reaches a predetermined threshold number to make a fourth determination result. When the fourth determination result is negative, the step subsequent to step 1120 returns to step 1112. When the fourth determination result is affirmative, the step after step 1120 returns to step 1110. For example, under the condition that the predetermined threshold number is equal to 1, the current negative determination result number is only 1, and the fourth determination result is affirmative.

In step 1122, the processing unit 22 recognizes that the valid reference signal SRP is the first candidate reference signal SR2, determines that the first body motion ML1 satisfies a predetermined period motion start condition, and based on the first motion feature value data. The unit DA2 and the at least one first feature function code data unit F1 perform a calculation to identify the first motion type HM1.

In step 1124, the processing unit 22 determines whether the first motion type HM1 is successful by performing the calculation to make a fifth determination result. When the fifth determination result is negative, the step subsequent to step 1124 returns to step 1110. When the fifth determination result is affirmative, the step after step 1124 is step 1126.

In step 1126, the processing unit 22 obtains a motion type code CT1 representing the first motion type HM1, and generates a motion measurement related to the first body motion ML1 based on the motion type code CT1. Quantity information DM1.

In each of the various embodiments disclosed in FIGS. 1 , 2 , and 3 , any of the motion recognition devices 20 , 30 , and 40 employs the sensing unit 21 (as a six-axis inertial sensing). Component) to achieve this type of motion recognition. The sensing unit 21 includes an accelerometer 211 that is a three-axis accelerometer and a gyroscope 212 that is a three-axis gyroscope. For any of a variety of body motion types known to be commonly found in fitness centers, any of the motion recognition devices 20, 30, and 40 automatically recognize their wearing position, and the type of motion of any body motion. For example, for the purposes of the following description, the wearing position is selected from the wrist position, the upper arm position, and the ankle position. Most body movement types are used separately to exercise specific muscle groups. In order to avoid the occurrence of muscle compensatory phenomenon, a body exercise type only exercises a related muscle group of a practitioner to improve the training effect of the muscle group. Therefore, in some embodiments of the present disclosure, the wearing positions of the respective types of movement are divided into the three wearing positions in advance according to the main muscle groups exercised by the respective types of exercise.

Briefly, for exercise of the hand muscle group, the wearing position is configured as the wrist position. For exercise of the chest, back, waist, and abdominal muscle groups, the wearing position is configured as the upper arm position. For exercise of the foot muscle group, the wearing position is configured as the ankle position. According to this division, it is possible to have a large system extensibility, that is, a more recognizable type of body movement. The identification result of the wearing position is the main basis of the motion type identification, wherein the wearing position is variable according to the body motion type.

The second motion type recognition program executed by any one of the motion recognition devices 20, 30, and 40 includes: processing the signal by using six The axis sensing signal identifies the wearing position, determines the start/end of the periodic motion, determines a valid reference signal, and identifies the type of motion. Next, the motion recognition device 20 is taken as an example to describe the second motion type recognition program.

The sensing unit 21 includes the accelerometer 211 and the gyroscope 212, and the sensing signal SE1 is generated according to the first body motion ML1 at the specific position PL1 on the user's body 91, wherein the first body motion ML1 belongs to the segment action of the first motion type HM1. The processing unit 22 receives the sensing signal SE1 from the sensing unit 21, and generates the motion parameter signal structure SP1 according to the sensing signal SE1. The accelerometer 211 generates the accelerometer signal SE11 (or a first signal) as a first triaxial sensing signal; and the gyroscope 212 generates the gyroscope signal SE12 as a second triaxial sensing signal ( Or a second signal). For example, the sensing signal SE1 is a six-axis sensing signal, and includes the accelerometer signal SE11 and the gyro signal SE12; and the motion parameter signal structure SP1 includes a fusion of the accelerometer signal SE11 and the gyro signal SE12. Signal SPS0. The fused signal SPS0 is generated by using a signal fusion operation and is a signal for estimating the angle SK1. The signal fusion operation includes a signal for calculating the first angle θ based on the accelerometer signal SE11 and the gyro signal SE12 to form the estimated angle SK1.

The accelerometer signal SE11 is expressed as follows:

The gyroscope signal SE12 is expressed as follows:

In a first state in which the motion recognition device 20 is stationary, the processing unit 22 roughly obtains an estimated pitch angle and an estimated roll angle based on the accelerometer signal SE11. A method for calculating the estimated pitch angle and the estimated roll angle is as follows: in the first state, the accelerometer signal SE11 is expressed as . Based on the accelerometer signal SE11, the estimation of the roll angle of the motion recognition device 20 is expressed as φ _Roll =tan ^-1 (g _y /g _z ). Based on the accelerometer signal SE11, the estimation of the pitch angle of the motion recognition device 20 is expressed as .

The processing unit 22 performs a filtering and correcting operation on the accelerometer signal SE11 to generate a third signal. The estimation of the roll angle of the motion recognition device 20 and the estimation of the pitch angle are expressed as follows:

In the first state in which the motion recognition device 20 is stationary, the gyro signal SE12 is expressed as ,among them Should be 0. then, . Under this condition, the bias of the gyroscope 212 can be recalibrated, thereby suppressing Any errors caused in the attitude calculation.

When the motion of the motion recognition device 20 continuously changes, the roll angle estimation and pitch of the motion recognition device 20 are based on the following dynamic equation including the measured appendage angular velocity and attitude angle (including the roll angle φ and the pitch angle θ). The estimate of the angle can be calculated.

In the signal fusion operation, when the motion recognition device 20 is relatively stationary (ie, 0, 0 o'clock), the roll angle and the pitch angle satisfy the following equation:

The obtained roll angle and the obtained pitch angle form an integral initial value of the above dynamic equation, and are substituted into the above dynamic equation to obtain the estimated roll angle And estimated pitch angle .

In the signal fusion operation, the processing unit 22 obtains the estimated roll angle by performing the following operations. And estimated pitch angle :

When the motion recognition device 20 is relatively stationary, the value of α tends to 1.

Conversely, when the motion recognition device 20 is moved, the value of α tends to zero. The estimated pitch angle θ _Pitch is expressed similar to the estimated roll angle φ _Roll and is as follows:

The motion parameter signal structure SP1 includes complex motion parameter signals SP11, SP12, SP13, ..., SP18, SP19, and the complex motion parameter signals SP11, SP12, SP13, ..., SP18, SP19 include the third signal. The fourth signal The estimated roll angle The signal, and the signal of the estimated pitch angle θ _Pitch . For example, the estimated roll angle The signal of the estimated pitch angle θ _Pitch and the estimated pitch angle θ _Pitch are respectively a first fused signal and a second fused signal; and the signal of the estimated angle SK1 is the signal of the estimated pitch angle θ _Pitch . The user body 91 contains the particular body part 911 at the particular location PL1, the particular body part 911 having the first body movement ML. For example, the particular body portion 911 is the wrist, the upper arm or the ankle.

The motion recognition device 20 further includes the button 23, which is coupled to the processing unit 22 and acts as a motion initiation button. In the state in which the orientation KA1 is directed to the predetermined direction QH1 related to the gravity direction QF1, the button 23 causes the processing unit 22 to receive the trigger signal ST1 in response to a user press from the user 90. The processing unit 22 determines whether the processing unit 22 detects the trigger signal ST1 to make the first determination result. For example, when the first determination result is affirmative, the processing unit 22 is based on the signal of the estimated pitch angle θ _Pitch and the optional estimated roll angle. The signal to get the specific location code CP1. For example, the trigger signal ST1 indicates the initiation of the first body motion ML1. The motion recognition device 20 is worn to the particular body portion 911, and the particular body portion 911 is determined based on a design configuration that is based on the operational context, operational habits, and ergonomics that best fits the user 90. The principle is determined.

For example, the motion recognition device 20 has a gravitational acceleration gF having a gravitational acceleration direction that is the same as the gravitational direction QF1. As shown in FIG. 12, in the state in which the orientation KA1 is directed to the predetermined direction QH1 related to the gravity direction QF1, the button 23 causes the processing unit 22 to receive the trigger signal ST1 according to the user's pressing. The processing unit 22 determines whether the processing unit 22 detects the trigger signal ST1 to make the first determination result. When the first determination result is affirmative, the processing unit 22 recognizes the specific position PL1 based on the estimated pitch angle θ _Pitch included in the motion parameter signal structure SP1 at this time. For example, the recognition result of the specific position PL1 is classified into the following types. When the estimated pitch angle θ _Pitch satisfies the relationship of -60° < θ _Pitch < 60°, the recognition result obtained by the processing unit 22 indicates that the wearing position is the wrist position, and the specific position code CP1 represents the Position PL5. When the estimated pitch angle θ _Pitch satisfies the relationship of -120° < θ _Pitch < -60°, the recognition result obtained by the processing unit 22 indicates that the wearing position is the upper arm position, and the specific position code CP1 Represents the position PL6. When the estimated pitch angle θ _Pitch satisfies -60° < θ _Pitch In the 135° relationship, the recognition result obtained by the processing unit 22 indicates that the wearing position is the ankle position, and the specific position code CP1 represents the position PL9. When the identification of the wearing position is completed, the first stage of the motion type identification is completed.

In some embodiments, the first motion type HM1 is selected from the complex motion types HA1, HA2, ..., HA6. The complex motion types HA1, HA2, ..., HA6 are predetermined in relation to the specific position PL1, and respectively have a plurality of main motion axis directions QA1, QA2, ..., QA6 associated with the attached coordinate system UA1. The plurality of main motion axis directions QA1, QA2, ..., QA6 are detected in advance to generate a direction corresponding to the plurality of main motion axes QA1, QA2, ..., QA6 correspond to the plurality of main motion axis direction data units DQ1, DQ2, ..., DQ6, respectively. The processing unit 22 obtains an identification reference data unit DR based on the specific position code CP1. The identification reference data unit DR is predetermined based on the complex main motion axis direction data units DQ1, DQ2, ..., DQ6, and includes a candidate reference signal code data unit CA and a motion type indicator data unit CH.

The candidate reference signal code data unit CA includes a first candidate reference signal code CA2, a second candidate reference signal code CA1, and a third candidate reference signal code CA3. The first, second and third candidate reference signal codes CA2, CA1 and CA3 respectively represent a first candidate reference signal SR2, a second candidate reference signal SR1 and a first one derived from the motion parameter signal structure SP1. Three candidate reference signals SR3. The motion type indicator data unit CH includes a first motion type indicator, a second motion type indicator, and a first corresponding to the first, second, and third candidate reference signal codes CA2, CA1, and CA3, respectively. a third motion type indicator, wherein the first motion type indicator is a motion type indicator CH2. The motion type indicator CH2 indicates one of an invalid motion type and a second motion type HM2, and the second motion type HM2 is included in the plural motion type HA1, HA2, ..., HA6. The processing unit 22 determines the first, second, and third candidate reference signals SR2 based on the motion parameter signal structure SP1 and the first, second, and third candidate reference signal codes CA2, CA1, and CA3. SR1 and SR3.

In some embodiments, the first, second, and third candidate reference signals SR2, SR1, and SR3 form a candidate reference signal combination. SRG. The identification reference data unit DR further includes a representative signal code CB1 representing a representative signal SPS1 in the complex motion parameter signals SP11, SP12, SP13, ..., SP18, SP19. The processing unit 22 determines the representative signal SPS1 based on the motion parameter signal structure SP1 and the representative signal code CB1.

For example, the specific location code CP1 represents the location PL9 (or the ankle position). For example, the first candidate reference signal SR2 is an acceleration differential signal. The acceleration differential signal Is an acceleration signal derived from the motion parameter signal structure SP1 . The second candidate reference signal SR1 is a sum signal ( + ), the sum signal is an acceleration differential signal And an acceleration differential signal Sum. Acceleration differential signals with Is derived from the acceleration signals included in the motion parameter signal structure SP1 with . The third candidate reference signal SR3 are included in the structure of the motion parameter signal SP1 angular velocity signal ω _x. The representative signal SPS1 is the acceleration signal included in the motion parameter signal structure SP1. .

The processing unit 22 obtains a complex motion feature value from the motion parameter signal structure SP1 based on the first, second, and third candidate reference signals SR2, SR1, and SR3. After the wearing position is determined, the processing unit 22 determines the candidate reference signal combination SRG based on the identification reference data unit DR. For example, when the wearing position is recognized as the pedal position, the processing unit 22 simultaneously triggers the first, second, and third candidate reference signals SR2, SR1, and SR3 (eg, , + )with ) to obtain the complex motion feature value. The identification reference unit DR is predetermined and employed to identify the first motion type HM1.

For example, the first body motion ML1 is a back and forth motion and includes a first motion portion ML11 and a second motion portion ML12 adjacent to the first motion portion ML11. The first and second moving portions ML11 and ML12 respectively form a first motion cycle W1 and a second motion cycle W2 adjacent to the first motion cycle W1. The processing unit 22 obtains the complex motion feature value from the motion parameter signal structure SP1 based on the first, the second, and the third candidate reference signals SR2, SR1, and SR3 during the second motion cycle W2. The main motion feature values in the complex motion feature values include , , , , φ _{Roll , max} , φ _{Roll , min} , θ _{Pitch , max} and θ _{Pitch , min} . The movement feature value is a complex motion feature value represents the DFE _n, where n is a natural number and 1 <n 27.

Each of the first, second, and third candidate reference signals SR2, SR1, and SR3 includes a first motion reference signal portion and a corresponding one of the first motion portion ML11 and the second motion portion ML12, respectively. The second motion reference signal portion. The processing unit 22 separately obtains a first motion feature value data unit DA2, a second motion characteristic value data unit DA1, and a third from the motion parameter signal structure SP1 based on the complex candidate reference signals SR1, SR2, and SR3. Motion feature value data unit DA3. For example, the first candidate reference signal SR2 includes a first candidate reference signal portion SR21 and a second candidate reference signal portion SR22 corresponding to the first motion portion ML11 and the second motion portion ML12, respectively. The second candidate reference The signal portion SR22 is adjacent to the first candidate reference signal portion SR21.

For example, the first motion feature value data unit DA2 includes a first motion feature value data unit portion DA21 and a second motion feature value data unit portion DA22 corresponding to the first and second motion portions ML11 and ML12, respectively. The processing unit 22 obtains the first motion feature value data unit portion DA21 from the first motion parameter signal structure portion SP5 based on the first candidate reference signal portion SR21. The processing unit 22 obtains the second motion feature value data unit portion DA22 from the second motion parameter signal structure portion SP6 based on the second candidate reference signal portion SR22. The processing unit 22 generates a difference data unit DC1 based on the first and second motion characteristic value data unit portions DA21 and DA22, and determines whether the difference data unit DC1 satisfies a first one for a periodic motion start determination. Specific conditions to make a judgment result. The processing unit 22 obtains a representative extreme value between the first and second motion feature value data unit portions DA21 and DA22 by comparing the first and second motion feature value data unit portions DA21 and DA22. Deviation DC11.

For example, the first motion feature value data unit portion DA21 includes a first motion feature value data unit sub-portion DA21A corresponding to the representative signal SPS1, wherein the first motion feature value data unit sub-portion DA21A has a first maximum a value, a first minimum value, and a first difference between the first maximum value and the first minimum value. The second motion feature value data unit portion DA22 includes a second motion feature value data unit sub-portion DA22A corresponding to the representative signal SPS1, wherein the second motion feature value data unit sub-portion DA22A has a second maximum value, a second minimum value, and at the second maximum value and the second minimum value A second difference between the two. The processing unit 22 obtains a representative value difference DC12 from the first and the second differences. The difference data unit DC1 includes the representative extreme value deviation DC11 and the representative value difference DC12. The first specific condition includes a first sub-condition and a second sub-condition. The first sub-condition is that the representative extreme value deviation DC11 falls within a first predetermined range of values. The second sub-condition is that the representative value difference DC12 falls within a second predetermined range of values.

For example, the second determination result includes a first sub-judgment result and a second sub-judgment result. The first sub-judgment result is made in relation to whether the representative extremum deviation DC11 falls within the first predetermined value range, or is related to the first and second motion characteristic value data unit portions DA21 and DA22 Whether each of the errors or deviations between the corresponding feature values is less than a threshold δ _n is made. When the first sub-judgment result is affirmative, the second sub-judgment result is made in relation to whether the representative value difference DC12 falls within the second predetermined value range. When the second sub-judgment result is affirmative, the second determination result is affirmative, and the processing unit 22 recognizes that the valid reference signal SRP is the first candidate reference signal SR2, and determines that the first body motion ML1 satisfies Determining a periodic motion start condition, and determining, based on the first candidate reference signal code CA2, whether the motion type indicator CH2 indicates one of the plurality of motion types HA1, HA2, ..., HA6 to make a third determination result. When the third determination result is affirmative, the processing unit 22 recognizes that the first motion type HM1 is the second motion type HM2 based on the motion type indicator CH2, and obtains a motion representing the first motion type HM1. Type code CT1. When the second sub-judgment result is negative, the processing unit 22 determines that the first body motion ML1 satisfies a predetermined period motion termination condition, or determines that the first body motion ML1 does not satisfy the predetermined period motion start condition.

In some embodiments, the specific position PL1 is configured as the ankle position; and the plural motion type HA1, HA2, ..., HA6 includes a pinch (inner clip/abduction) motion type, a side leg lift motion type, and A type of leg lift exercise. When the first body motion ML1 belongs to the type of the pinch motion at the ankle position, the processing unit 22 identifies that the valid reference signal SRP is the third candidate reference signal SR3 (or the angular velocity) based on the second determination result. Signal ω _x ). When the first body motion ML1 belongs to one of the side leg movement type and the positive leg movement type at the ankle position, the processing unit 22 identifies that the valid reference signal SRP is based on the second determination result. The first candidate reference signal SR1 (or the acceleration differential signal) ).

For example, when the first candidate reference signal SR2 (or the acceleration differential signal) When the first motion characteristic value data unit DA2 is used to obtain the first sub-judgment result is positive, and the representative value difference DC12 is equal to or greater than 2500, the third determination result is made negative. And the first motion type HM1 is recognized as one of the side leg movement type and the positive leg movement type based on the first feature function code data unit F1. In another aspect, when the first candidate reference signal SR2 (or the acceleration differential signal) When the first motion feature value data unit DA2 is used to obtain the first sub-judgment result is positive, and the representative value difference DC12 is less than 2500, the second judgment result is made negative. For example, the third motion type indicator indicates the type of foot motion. When the third candidate reference signal SR3 (or the angular velocity signal ω _x ) is used to obtain the third motion characteristic value data unit DA3, and the representative value difference DC12 is less than 2500, the third determination result is made. To be sure, and the first motion type HM1 is recognized as the type of pinch motion.

On the other hand, when the third candidate reference signal SR3 is used to obtain the third motion feature value data unit DA3, and the representative value difference DC12 is equal to or greater than 2500, the second determination result is made as No, yes. When a specific motion feature value data unit obtained based on a specific candidate reference signal satisfies a predetermined condition for the periodic motion start determination, the following feature function is performed based on the specific motion feature value data unit. At this time, the processing unit 22 recognizes that the valid reference signal SRP is the specific candidate reference signal, and stops triggering other candidate reference signals.

In some embodiments, the processing unit 22 provides a first feature function code data unit F1 associated with the first candidate reference signal code CA2 and a first complex identification value region DU1, DU2, and DV1 in advance. The first feature function code data unit F1 is employed to indicate one of the first complex number value regions DU1, DU2, and DV1. The first complex identification value areas DU1, DU2, and DV1 include a first identification value area DU1, a second identification value area DU2 adjacent to the first identification value area DU1, and the first and second identifications. A first trust value region DV1 between the value regions DU1 and DU2. For example, the first trust value area DV1 is selectable. The first and second identification value regions DU1 and DU2 respectively refer to A third motion type H1 and a fourth motion type H2 are shown. The third and fourth motion types H1 and H2 are all included in the plurality of motion types HA1, HA2, ..., HA6.

When the third determination result is negative, the processing unit 22 performs a first calculation to generate a first identification value DH1 based on the first feature function code data unit F1 and the first motion feature value data unit DA2, and It is determined whether the first identification value DH1 belongs to one of the first and second identification value areas DU1 and DU2 to make a fourth determination result. When the fourth determination result is affirmative, the processing unit 22 determines that the first identification value DH1 belongs to a valid identification value area DUA in the first and second identification value areas DU1 and DU2, and identifies the first The motion type HM1 is an effective motion type indicated by the valid identification value region DUA, and the motion type code CT1 corresponding to the valid identification value region DUA is obtained. The processing unit 22 generates motion measurement information DM1 related to the first body motion ML1 based on the motion type code CT1.

In some embodiments, the processing unit 22 further provides a second feature function code data unit F2 different from the first feature function code data unit F1, and the second feature function code data unit F2 is related to the first candidate. The reference signal code CA2, a third identification value area DU3, and a fourth identification value area DU4 adjacent to the third identification value area DU3. When the fourth determination result is negative, the processing unit 22 performs a second calculation based on the second feature function code data unit F2 and the first motion feature value data unit DA2 to generate a second identification value DH2, and determines Whether the second identification value DH2 belongs to the third and fourth identification value areas One of the fields DU3 and DU4, and thereby identifying the first motion type HM1.

In some embodiments, the processing unit 22 provides in advance a set of feature function code data units F1, F2, etc. representing a set of motion type cutting functions P _{1 - K} , respectively, the set of feature function code data units F1, F2, etc. It is established by calculating a large amount of data in the database, and is determined to satisfy the periodic motion initiation condition of the complex motion types HA1, HA2, ..., HA6. For example, the processing unit 22 is based on at least one of the complex motion feature values DFE _n (including the second motion feature value data unit portion DA22), and the set of feature function code data units F1, F2, and the like. The data unit performs a calculation to effectively identify the first motion type HM1. The set of motion type cutting functions P _1~K has a set of cutting planes and a number of levels K, wherein the number of levels K is a natural number. When the number of levels K is greater than 1, the set of motion type cutting functions P _{1 - K} are arranged based on a predetermined order and are at least partially employed to identify the first motion type HM1. For example, the processing unit 22, by using this type of cutting motion function P ₁ to identify the first type of motion is the HM1 and one side leg lift exercise type wherein the positive leg is sufficient.

For example, the set of motion type cutting functions P _1~K are expressed as follows: , i=1~K, where σ represents a threshold distance of a trust value region; for example, the trust value distance σ1 of the first trust value region DV1 is equal to 2σ.

For example, the motion type cutting function P ₁ is expressed as follows: P ₁ ={abs(a _z,Max +6000)/1000}‧{abs(a _z,Min +6000)/1000}‧{abs(θ _{Pitch, Max} +90)/100}‧{abs(θ _Pitch,min +180)/100}±20

For example, the processing unit 22 performs the first calculation by substituting the complex motion feature value DFE _n into the motion type cutting function P ₁ to generate a calculation result value (or the first identification value DH1), wherein the motion type The cutting function P ₁ is represented by a relationship between the complex motion characteristic values DFE _n . When the calculated result value is equal to or less than 130, the processing unit 22 recognizes that the first motion type HM1 is the side lift leg motion type. When the calculated result value is greater than 130, the processing unit 22 recognizes that the first motion type HM1 is the positive leg motion type. For example, the first and second feature function code data units F1 and F2 represent the motion type cutting functions P ₁ and P _{2 , respectively} .

In some embodiments, the processing unit 22 provides a complex feature value range table respectively corresponding to the complex motion types HA1, HA2, . . . , HA6, wherein the complex feature value range table includes one corresponding to the first motion type HM1. A list of specific feature value ranges. The complex motion feature values DFE _n respectively correspond to a range of complex feature values in the particular feature value range table. When identifying the first motion type HM1 successful, the value processing unit 22 for each value of the plurality of DFE _n motion feature, determining whether the plurality of moving respectively feature value DFE _n falling within the plurality of feature value ranges to do A fifth judgment result. When the fifth determination result is negative, the processing unit 22 determines that the identification of the first motion type HM1 is unsuccessful, and re-triggers the candidate reference signal combination SRG.

Please refer to Figure 16, which is a variety of examples in this disclosure. A flow chart of a second motion type identification calculation program 120 in the embodiment. In step 1202, the sensing unit 21 generates a sensing signal SE1 according to a first body motion ML1 occurring at a specific position PL1 on a user's body 91. The first body motion ML1 belongs to a first motion type HM1; and the sensing signal SE1 is generated in relation to an attached coordinate system UA1 of the sensing unit 21.

In step 1204, the processing unit 22 or the conversion unit 411 processes the sensing signal SE1 to generate a motion parameter signal structure SP1, wherein the sensing signal SE1 includes an accelerometer signal SE11 and a gyroscope signal SE12, and the The motion parameter signal structure SP1 can include the accelerometer signal SE11 and a fused signal SPS0 of the gyro signal SE12.

In step 1206, the processing unit 22 determines whether the processing unit 22 detects a trigger signal ST1 to make a first determination result. When the first determination result is negative, the step subsequent to step 1206 returns to step 1206. When the first determination result is affirmative, the step subsequent to step 1206 is step 1208.

In step 1208, the processing unit 22 generates a specific location code CP1 representative of the particular location PL1 based on the motion parameter signal structure SP1. The user body 91 is worn with the motion recognition device 20 at the specific position PL1, wherein the motion recognition device 20 includes the attached coordinate system UA1, the sensing unit 21, and the processing unit 22.

In step 1210, the processing unit 22 obtains an identification reference unit DR based on the specific location code CP1. Identification reference The data unit DR is predetermined based on the attached coordinate system UA1 and a main motion axis direction of the first motion type HM1, and includes a candidate reference signal code data unit CA. The candidate reference signal code data unit CA includes a complex candidate reference signal code (such as a first candidate reference signal code CA2, a second candidate reference signal code CA1, and a third candidate reference signal code CA3). The complex candidate reference signal codes (such as the first, the second, and the third candidate reference signal codes CA2, CA1, and CA3) respectively represent a plurality of candidate reference signals (such as a first one) derived from the motion parameter signal structure SP1. The candidate reference signal SR2, a second candidate reference signal SR1, and a third candidate reference signal SR3). In step 1210, the processing unit 22 determines the complex number based on the motion parameter signal structure SP1 and the complex candidate reference signal code (such as the first, the second, and the third candidate reference signal codes CA2, CA1, and CA3). Candidate reference signals (such as the first, second and third candidate reference signals SR2, SR1 and SR3).

In step 1212, the processing unit 22 obtains correspondingly from the motion parameter signal structure SP1 based on the complex candidate reference signals (such as the first, the second, and the third candidate reference signals SR2, SR1, and SR3). a complex motion feature value data unit of the plurality of candidate reference signals (such as the first, second, and third candidate reference signals SR2, SR1, and SR3) (such as a first motion feature value data unit DA2, a second motion) The feature value data unit DA1 and a third motion feature value data unit DA3). The first body motion ML1 includes a first motion portion ML11 and a second motion portion ML12 adjacent to the first motion portion ML11. The first and second moving portions ML11 and ML12 respectively form a first motion A cycle W1 and a second motion cycle W2 adjacent to the first motion cycle W1. Each of the plurality of motion feature value data units (such as the first, second, and third motion feature value data units DA2, DA1, and DA3) includes the first and second motion portions ML11 and ML12 Corresponding to a first motion feature value data unit portion and a second motion feature value data unit portion respectively.

In step 1214, the processing unit 22 generates a plurality of motion feature value data units (such as the first, the second, and the third) based on the respective first and respective second motion feature value data unit portions. The motion feature value data units DA2, DA1, and DA3) respectively correspond to the plurality of difference data units, and use the complex difference data unit to determine whether the first body motion ML1 is effectively ended or started. The complex difference data units respectively comprise a plurality of representative extreme value deviations, each deviation of the complex representative extreme value deviations being generated by comparing the respective first and respective second motion characteristic value data unit portions.

In step 1216, the processing unit 22 determines whether the complex difference data unit includes a first difference data unit to make a second determination result, and the first difference data unit is optimally satisfied for a periodic motion start. A first specific condition of judgment. When the second determination result is negative, the step subsequent to step 1216 is step 1218. When the second determination result is affirmative, the step subsequent to step 1216 is step 1222.

In step 1218, the processing unit 22 determines whether any of the data units of the plurality of difference data units satisfy a second specific condition for the one-cycle motion end determination to make a third determination result. When When the third determination result is negative, the step after step 1218 is step 1220. When the third determination result is affirmative, the step subsequent to step 1218 returns to step 1210.

In step 1220, the processing unit 22 determines whether a current negative number of steps 1218 has reached a predetermined threshold to make a fourth determination result. When the fourth determination result is negative, the step subsequent to step 1220 returns to step 1212. When the fourth determination result is affirmative, the step subsequent to step 1220 returns to step 1210.

In step 1222, the processing unit 22 recognizes that a fourth candidate reference signal corresponding to the first difference data unit is a valid reference signal SRP, and determines that the first body motion ML1 meets a predetermined period motion initiation condition. The fourth candidate reference signal is included in the complex candidate reference signal.

In step 1224, the processing unit 22 identifies the first motion type HM1 based on a fourth motion feature value data unit and at least one first feature function code data unit F1, wherein the fourth motion feature value data unit corresponds to the The first difference data unit is included in the complex motion feature value data unit.

In step 1226, the processing unit 22 determines whether the first motion type HM1 is successful by performing the calculation to make a fifth determination result. When the fifth determination result is negative, the step subsequent to step 1226 returns to step 1210. When the fifth determination result is affirmative, the step subsequent to step 1226 is step 1228.

In step 1228, the processing unit 22 obtains the representative A motion type code CT1 of the first motion type HM1, and based on the motion type code CT1, generates motion measurement information DM1 related to the first body motion ML1.

Please refer to Figures 17A and 17B. Figure 17A is a block diagram showing the structure of a motion recognition device 50 in various embodiments of the present disclosure. Fig. 17B is a schematic view showing the appearance of the motion recognition device 50 in Fig. 17A. The motion recognition device 50 is a wearable motion sensing device and includes an operating unit 51 and a coupling unit 52 coupled to the operating unit 51. For example, the coupling unit 52 is a wearing unit. The operating unit 51 includes a motion sensing unit 55, a processing unit 56, and a human interface unit 57. The processing unit 56 includes a motion recognition module 561 and a signal processing module 562 coupled to the motion sensing unit 55 and the human interface unit 57. The signal processing module 562 is coupled to the motion recognition module. 561 and the human interface unit 57.

In some embodiments, the motion sensing unit 55 is configured to include at least one motion sensor, such as an accelerometer, gyroscope, or geomagnetic instrument. The coupling unit 52 is employed to mount the motion sensing unit 55 and is worn on a body part of the exerciser such that the motion sensing unit 55 is adapted to sense motion. The motion recognition module 561 includes at least one of a motion recognition function code data unit, a motion recognition specification, an identification parameter, an identification reference value, a threshold value, and the like, and a related algorithm. The signal processing module 562 is configured to receive an input from the fitness person, output, display, or access related fitness materials. For example, the signal processing module 562 is employed to display or access the sensed motion state, the number of cycles of body motion, and the health Body results, energy consumption, and more. The human interface unit 57 is capable of communicating with a smart mobile phone, configured to have a wireless communication device (such as a built-in Bluetooth transmission function), coupled to the motion sensing unit 55 for transmitting data, and coupled to the signal processing. Module 562 is for receiving, processing, computing, storing data, or the like. For example, the human interface unit 57 includes a button 23 coupled to the signal processing module 562. The motion sensing unit 55 and the processing unit 56 are similar to the sensing unit 21 and the processing unit 22 in FIG. 1, respectively. Further, the motion recognition device 50 is further operated based on the function of the motion recognition device 20 in Fig. 1.

In some embodiments according to FIGS. 17A and 17B, a motion recognition device 60 is a wearable device configured to measure a body motion and includes a motion sensing unit 55, a coupling unit 52, and a motion. The identification module 561, a signal processing module 562 and a human interface unit 57 are provided. The motion sensing unit 55 includes at least an accelerometer, a gyroscope or a geophone, and senses the body motion to generate a motion sensing signal. The coupling unit 52 includes an engagement structure and a wearing device coupled to the engagement structure. The motion sensing unit 55 is fixed to the joint structure. The wearing device is a body part that is worn to a user. The motion recognition module 561 is electrically connected to the motion sensing unit 55, and has a signal identification function code data unit, and the motion sensing signal is processed by using the signal identification function code data unit, and thereby determining the The body motion conforms to a first specific motion specification, wherein the signal identification function code data unit represents a signal recognition function. The signal processing module 562 is electrically connected to the motion recognition unit 55 and has a signal processing function code data unit. The signal processing function code data unit identifies the body motion that conforms to the first specific motion specification as a first specific motion, and generates motion information related to the body motion, wherein the motion information forms a record, and the signal The processing function code data unit represents a signal processing function. The human interface unit 57 is electrically connected to the signal processing module 562 and outputs the motion information or receives specification data related to a second specific motion.

In some embodiments according to Figures 17A and 17B, a measurement method for measuring a body motion is disclosed. The measuring method comprises the steps of: sensing a body motion occurring at a specific position on a user to generate a motion sensing signal; and processing the motion sensing signal by using a signal recognition function code data unit to confirm The body motion conforms to a first specific motion specification; the body motion conforming to the first specific motion specification is recognized as a first specific motion by using a signal processing function code data unit, and a motion related to the body motion is generated. Motion information, wherein the motion information forms a record; and outputs the motion information or receives specification data related to a second specific motion.

In some embodiments according to the 17A and 17B, a motion recognition device 61 is a wearable device configured to measure a body motion and includes a motion sensing unit 55, a coupling unit 52, and a The motion recognition module 561, a signal processing module 562 and a human interface unit 57. The motion sensing unit 55 has at least one accelerometer, a gyroscope or a geophone, and senses the body motion to generate a motion sensing signal. The coupling unit 52 includes an engagement structure and a wearing device coupled to the engagement structure. The motion sensing unit 55 is fixed to the joint structure. The Pei The wearing device is a body part that is worn to a user. The motion recognition module 561 is electrically connected to the motion sensing unit 55, and has a signal identification function code data unit, and the motion sensing signal is processed by using the signal identification function code data unit, and thereby determining the The motion sensing unit 55 is a specific location that is worn on the user, wherein the signal identification function code data unit represents a signal identification function. The signal processing module 562 is electrically connected to the motion recognition unit 55 and has a signal processing function code data unit. The body motion recognition function conforming to a first specific motion specification is adopted by using the signal processing function code data unit. And being a first specific motion, and generating motion information related to the body motion, wherein the motion information forms a record, and the signal processing function code data unit represents a signal processing function. The human interface unit 57 is electrically connected to the signal processing module 562 and outputs the motion information or receives specification data related to a second specific motion.

In some embodiments according to Figures 17A and 17B, a measurement method for measuring a body motion is disclosed. The measuring method comprises the steps of: sensing a body motion occurring at a specific position on a user to generate a motion sensing signal; processing the motion sensing signal by using a signal recognition function code data unit to match The specific position wearing a motion sensing unit is recognized; the body motion conforming to a first specific motion specification is recognized as a first specific motion by using a signal processing function code data unit, and The motion information of the body movement, wherein the motion information forms a record; and outputs the motion information or receives the specification data related to a second specific motion.

Please refer to Fig. 18, which is a schematic view of the motion recognition device 50 worn on the wrist in Fig. 17B. Please refer to Fig. 19, which is a schematic view of the exercise recognizing device 50 worn on the upper arm in Fig. 17B.

Numerous variations and other embodiments of the present invention are disclosed herein, and those skilled in the art will appreciate the benefit of the teachings presented in the above description and related drawings. Therefore, it is to be understood that the invention is not limited to the specific embodiments disclosed, and the modifications and other embodiments are intended to be included within the scope of the following claims.

Claims (11)

A motion recognition device includes: a sensing unit, which is mounted on a specific position on a user's body, and generates a sensing signal according to a first body motion that occurs at a specific position, wherein the sensing signal includes a a first sensing signal portion, and a second sensing signal portion different from the first sensing signal portion, and the first body motion belongs to a segment motion of a first motion type; and a processing unit that processes the Sensing the signal to generate a motion parameter signal structure, and identifying the specific position based on the motion parameter signal structure to determine a valid reference signal for identifying the first motion type, wherein the motion parameter signal structure includes the first sum a fused signal of the second sensing signal portion, wherein the first sensing signal portion is an accelerometer signal, the second sensing signal portion is a gyro signal, and the processing unit generates the signal based on the motion parameter signal structure A specific location code representing the particular location and thereby identifying the particular location. The motion recognition device of claim 1, configured to have a certain orientation, a direction of gravity, and an attachment coordinate system for determining the orientation, and being worn in the specific position, wherein: the sensing unit The method includes: an accelerometer that provides the accelerometer signal to the processing unit; and a gyroscope that provides the gyroscope signal to the processing unit; the first motion type is a selected self-complex motion type; the complex motion type is related to Predetermined for the specific location Having a plurality of main motion axis directions associated with the attached coordinate system, wherein the plurality of main motion axis directions are pre-detected to generate a plurality of main motion axis direction data units respectively corresponding to the plurality of main motion axis directions; Is in a state of pointing to a predetermined direction related to the direction of gravity; the processing unit obtains an identification reference unit based on the specific position code, wherein: the identification reference unit is predetermined based on the plurality of main motion axis direction data units And including a candidate reference signal code data unit and a motion type indicator data unit; the candidate reference signal code data unit includes a first candidate reference signal code; the first candidate reference signal code represents the motion parameter signal a first candidate reference signal of the structure; the motion type indicator data unit includes a motion type indicator corresponding to the first candidate reference signal code; and the motion type indicator indicates an invalid motion type and a second motion type One of the second types of motion is included in the plural Type; and the processing unit determines the first candidate reference signal based on the motion parameter signal structure of the first candidate and the code reference signal. The motion recognition device of claim 2, wherein: the specific location is a plurality of different locations selected from the user's body; The processing unit is coupled to the sensing unit; the sensing signal is generated in relation to the attached coordinate system; the predetermined direction is determined based on the specific position; and the first angle is between the predetermined direction and the gravity direction The fusion signal is a signal for estimating an angle associated with the first angle; in the state, the processing unit determines whether the processing unit detects a trigger signal to make a determination result; and when the determination result To be sure, the processing unit generates the particular location code based on the estimated angle. The motion recognition device of claim 2, wherein: the first body motion comprises a first motion portion, and a second motion portion adjacent to the first motion portion; the first and the second motion portion Forming a first motion cycle and a second motion cycle adjacent to the first motion cycle; the motion parameter signal structure includes a first motion parameter signal structure respectively corresponding to the first and second motion portions And a second motion parameter signal structure portion; the first candidate reference signal includes a first candidate reference signal portion and a second candidate reference signal portion respectively corresponding to the first motion portion and the second motion portion; The processing unit obtains a first motion feature value data unit from the motion parameter signal structure based on the first candidate reference signal, where the first motion feature value data unit includes the first and the first a first motion feature value data unit portion and a second motion feature value data unit portion respectively corresponding to the two motion portions; the processing unit obtains the first motion parameter signal structure portion based on the first candidate reference signal portion a first motion feature value data unit portion; and the processing unit obtains the second motion feature value data unit portion from the second motion parameter signal structure portion based on the second candidate reference signal portion; the processing unit is based on the first And the second motion feature value data unit portion to generate a difference data unit, and determine whether the difference data unit satisfies a first specific condition for a periodic motion start determination to make a first determination result; When the first determination result is affirmative, the processing unit recognizes that the valid reference signal is the first candidate reference signal, determines that the first body motion satisfies a predetermined period motion start condition, and is based on the first candidate reference signal code. Determining whether the motion type indicator indicates one of the plural motion types to make one Determining a result; when the second determination result is affirmative, the processing unit recognizes that the first motion type is the second motion type based on the motion type indicator, and obtains a motion type representative of the first motion type The motion parameter signal structure includes a plurality of motion parameter signals; the identification reference data unit further includes a representative signal code, the representative signal code representing a representative signal included in the complex motion parameter signal; The processing unit determines the representative signal based on the motion parameter signal structure and the representative signal code; the processing unit obtains the first and the first by comparing the first and the second motion feature value data unit portions a representative extreme value deviation between the two motion feature value data unit portions; the first motion feature value data unit portion includes a first motion feature value data unit sub-portion corresponding to the representative signal, wherein the first motion The feature value data unit subsection has a first maximum value, a first minimum value, and a first difference between the first maximum value and the first minimum value; the second motion feature value data unit portion includes Corresponding to a second motion feature value data unit sub-portion of the representative signal, wherein the second motion feature value data unit sub-portion has a second maximum value, a second minimum value, and a sum of the second maximum value a second difference between the second minimum values; the processing unit obtains a representative value difference from the first and the second differences; the difference data unit includes the representative a value deviation and the representative value difference; the first specific condition includes a first sub-condition and a second sub-condition; the first sub-condition is that the representative extremum deviation falls within a first predetermined value range; a second sub-condition is that the representative value difference falls within a second predetermined value range; the second motion portion is later than the first motion portion; the first motion feature value data unit includes a plurality of specific motion features a first feature function code data unit associated with the first candidate reference signal code and a first complex identification value region, wherein the first feature function code data unit is used to indicate the first One of the plurality of identification value regions, the first complex identification value region including a first identification value region, a second identification value region adjacent to the first identification value region, and the first and the first a first trust value region between the two identification value regions; and the first and second identification value regions respectively indicating a third motion type and a fourth motion type, the third and the fourth motion types respectively In the plural motion type; the motion recognition device pre-establishes the first feature function code data unit by sensing a plurality of body motions occurring at the specific position, wherein the plurality of body motions are segmented into a plurality of motion groups, The complex motion group belongs to the complex motion type respectively; the first feature function code data unit represents a motion type cutting function, and is based on the complex number specific a relationship between the dynamic feature values is used, wherein the motion type cutting function is employed to indicate the one of the first complex identification value regions; and when the second determination result is negative, the processing unit is based on the The first feature function code data unit and the first motion feature value data unit perform a first calculation to generate a first identification value, and determine whether the first identification value belongs to the first and second identification value regions. One of them to make a third determination result; when the third determination result is affirmative, the processing unit determines that the first identification value belongs to an effective identification value area in the first and second identification value areas, and identifies the first The motion type is an effective motion type indicated by the valid identification value region, and obtains the motion type code corresponding to the valid identification value region; the processing unit generates a motion related to the first body motion based on the motion type code Motion measurement information; the processing unit further provides a second feature function code data unit associated with the first candidate reference signal code and a second complex identification value region, wherein: the second feature function code data unit is different from the The first characteristic function code data unit is used to indicate one of the second complex identification value regions, and the second complex identification value region includes a third identification value region adjacent to the third identification value region a fourth identification value area, and a second trust value area between the third and the fourth identification value area; the third and the fourth identification The regions respectively indicate a fifth motion type and a sixth motion type, the fifth and the sixth motion types are all included in the complex motion type; and in a second specific condition, the third and the fourth identification One of the value regions at least partially overlaps with one of the first and second identification value regions; when the third determination result is negative, the processing unit is based on the second feature function code data unit and the First motion characteristic value data unit Performing a second calculation to generate a second identification value, determining whether the second identification value belongs to one of the third and fourth identification value regions, and thereby identifying the first motion type; the candidate reference signal The code data unit further includes at least one second candidate reference signal code representing the at least one second candidate reference signal, wherein: the second candidate reference signal code represents the second candidate reference signal originating from the motion parameter signal structure; The first candidate reference signal and the at least one second candidate reference signal form a candidate reference signal combination; the processing unit determines the at least one second candidate reference signal based on the motion parameter signal structure and the at least one second candidate reference signal code And obtaining, according to the at least one second candidate reference signal, at least one second motion feature value data unit corresponding to the at least one second candidate reference signal from the motion parameter signal structure; and when the processing unit processes the first The motion feature value data unit, the processing unit processes the at least one second motion feature value data unit to determine Whether the candidate reference signal comprising a combination of the active reference signal. A motion recognition device includes: a signal generating unit that is worn at a specific position on a user's body and generates a motion parameter signal structure according to a body motion of the user, wherein the body motion belongs to a first a type of motion; and a processing unit that identifies the specific location based on the motion parameter signal structure to determine a valid reference signal for identifying the first motion type, wherein the motion parameter signal structure includes a fused signal, the fused signal packet The accelerometer signal and a gyroscope signal are included, and the processing unit generates a specific location code representing the specific location based on the motion parameter signal structure, and thereby identifies the specific location. The motion recognition device of claim 5, configured to have a certain orientation, a direction of gravity, and an attachment coordinate system for determining the orientation, and being worn at the specific location, wherein: the specific location is Selecting a plurality of different positions on the body of the user; the signal generating unit includes a sensing unit and a converting unit coupled to the sensing unit; the sensing unit generates a sensing signal due to physical movement, and The method includes: an accelerometer, the accelerometer signal is provided to the conversion unit; and a gyroscope that provides the gyroscope signal to the conversion unit; the conversion unit generates the motion parameter signal structure according to the sensing signal; the processing unit Coupled to the signal generating unit; the first motion type is a selected self-complex motion type; the complex motion type is predetermined with respect to the specific position, and has a plurality of main motion axis directions associated with the attached coordinate system, respectively The direction of the plurality of main motion axes is pre-detected to generate a plurality of main motion axis directions corresponding to the direction of the plurality of main motion axes Data unit; the motion parameter signal structure is related to the attached coordinate system In the state in which the orientation is directed to a predetermined direction related to the direction of gravity; the processing unit obtains an identification reference unit based on the specific location code, wherein: the identification reference unit is based on the complex main motion axis Determining a direction data unit, and including a candidate reference signal code, and a motion type indicator corresponding to the candidate reference signal code; the candidate reference signal code representing a candidate reference signal derived from the motion parameter signal structure; The motion type indicator indicates one of an invalid motion type and a second motion type, the second motion type is included in the complex motion type; and the processing unit is based on the motion parameter signal structure and the candidate reference signal code To determine the candidate reference signal. The motion recognition device of claim 6, wherein: the body motion comprises a first moving portion, and a second moving portion adjacent to the first moving portion; the first and the second moving portions are respectively formed a first motion cycle, and a second motion cycle adjacent to the first motion cycle; the processing unit obtains a motion feature value data unit from the motion parameter signal structure based on the candidate reference signal, wherein the motion feature The value data unit includes a first motion feature value data unit portion and a second motion feature value data unit portion respectively corresponding to the first and second motion portions The processing unit generates a difference data unit based on the first and second motion feature value data unit portions; the processing unit determines whether the difference data unit satisfies a first specific condition for a periodic motion start determination The processing unit identifies that the valid reference signal is the candidate reference signal, and determines whether the motion type indicator is based on the candidate reference signal code, and determines a result of the first determination. Determining one of the plural motion types to make a second determination result; when the second determination result is positive, the processing unit identifies that the first motion type is the second based on the motion type indicator a motion type, and obtaining a motion type code representing the first motion type; the second motion portion is later than the first motion portion; the motion feature value data unit includes a plurality of specific motion feature values; the processing unit provides a feature in advance a function code data unit, wherein: the feature function code data unit is based on the complex number of specific motion feature values And associated with the candidate reference signal code, a first identification value region, and a second identification value region adjacent to the first identification value region; and the first and second identification value regions Determining a third motion type and a fourth motion type, respectively, the third and fourth motion types are included in the complex motion type; and when the second determination result is negative, the processing unit is based on the feature function a code data unit and the motion feature value data unit to perform a calculation And generating an identification value, determining that the identification value belongs to an effective identification value region in the first and second identification value regions, and identifying that the first motion type is an effective motion indicated by the valid identification value region Type, and obtain the motion type code corresponding to the valid identification value area. A motion recognition method includes the steps of: generating a motion parameter signal structure in response to a first body motion, wherein the first body motion occurs at a specific position on a user's body and belongs to a first motion type; Generating a specific location code representative of the specific location based on the motion parameter signal structure, and thereby identifying the specific location to determine a valid reference signal for identifying the first motion type, wherein the motion parameter signal structure includes a The fusion signal includes an accelerometer signal and a gyroscope signal. The motion recognition method of claim 8, further comprising the step of providing a motion recognition device, wherein: the motion recognition device is configured to have a certain direction, a gravity direction, and an attached coordinate system for determining the orientation. And being worn at the particular location; the first type of motion is a selection of a self-complex motion type; the complex motion type is predetermined relative to the particular location and has a plurality of primary motions associated with the attached coordinate system, respectively An axis direction, wherein the plurality of main motion axis directions are pre-detected to generate a plurality of main motion axis direction data units respectively corresponding to the plurality of main motion axis directions; and the motion recognition method further comprises the following steps: And in the state that the orientation is directed to a predetermined direction related to the direction of gravity; obtaining an identification reference unit based on the specific position code, wherein: the identification reference unit is predetermined based on the plurality of main motion axis direction data units And including a candidate reference signal code data unit and a motion type indicator data unit; the candidate reference signal code data unit includes a first candidate reference signal code, and the first candidate reference signal code represents the motion parameter signal a first candidate reference signal of the structure; the motion type indicator data unit includes a motion type indicator corresponding to the first candidate reference signal code; and the motion type indicator indicates an invalid motion type and a second motion type One of the second motion types is included in the complex motion type; and determining the first candidate reference signal based on the motion parameter signal structure and the first candidate reference signal code. The motion recognition method of claim 9, wherein: the first body motion belongs to a segment motion of the first motion type; the specific location is a plurality of different positions selected from the user's body; The step of motion generating the motion parameter signal structure includes the following sub-steps: generating a correlation coordinate system corresponding to the first body motion a sensing signal, wherein the sensing signal includes the accelerometer signal and the gyroscope signal; and processing the sensing signal to generate the motion parameter signal structure, wherein the motion parameter signal structure includes the accelerometer signal and the a fused signal of the gyro signal; the predetermined direction is determined based on the specific position; the predetermined direction and the direction of gravity have a first angle; the fused signal is a signal of an estimated angle associated with the first angle And the step of generating the specific location code includes the following sub-steps: determining, in the state, whether a trigger signal is detected to make a determination result; and when the determination result is positive, based on the estimation angle To generate the specific location code. The motion recognition method of claim 9, wherein: the motion parameter signal structure comprises a plurality of motion parameter signals; the first body motion comprises a first motion portion, and a second motion adjacent to the first motion portion The first and the second moving portions respectively form a first motion cycle and a second motion cycle adjacent to the first motion cycle; the motion parameter signal structure includes the first and the second motion a first motion parameter signal structure portion and a second motion parameter signal structure portion respectively corresponding to the portion; the first candidate reference signal includes the first motion portion and the second portion The moving part corresponds to a first candidate reference signal part and a second candidate reference signal part respectively; the identification reference data unit further comprises a representative signal code, the representative signal code representing one of the complex motion parameter signals The motion recognition method further includes the following steps: obtaining a first motion feature value data unit from the motion parameter signal structure based on the first candidate reference signal, wherein: the first motion feature value data unit includes The first motion feature value data unit portion and the second motion feature value data unit portion respectively corresponding to the first and the second motion portions; and the step of obtaining the first motion feature value data unit includes the following sub-steps: Obtaining the first motion feature value data unit portion from the first motion parameter signal structure portion based on the first candidate reference signal portion; and from the second motion parameter signal structure portion based on the second candidate reference signal portion Obtaining the second motion feature value data unit portion; based on the first and second The moving feature value data unit portion is configured to generate a difference data unit; determining whether the difference data unit satisfies a first specific condition for a periodic motion start determination to make a first determination result; when the first determination result When it is affirmative, identify the valid reference The signal is the first candidate reference signal, determining that the first body motion satisfies a predetermined period motion start condition, and determining, based on the first candidate reference signal code, whether the motion type indicator indicates one of the plurality of motion types And determining a second determination result; and when the second determination result is positive, identifying, according to the motion type indicator, the first motion type is the second motion type, and obtaining the representative first motion type a motion type code; the step of generating the difference data unit includes the following substeps: obtaining the first and second motion feature value data units by comparing the first and second motion feature value data unit portions a representative extreme value deviation between the portions; determining the representative signal based on the motion parameter signal structure and the representative signal code, wherein: the first motion characteristic value data unit portion includes one corresponding to the representative signal a first motion feature value data unit subsection, wherein the first motion feature value data unit subsection has a first maximum value, a first minimum value, and a first difference between the first maximum value and the first minimum value; and the second motion characteristic value data unit portion includes a second motion characteristic value corresponding to the representative signal a data unit subsection, wherein the second motion feature value data unit subsection has a second maximum value, a second minimum value, and a second difference between the second maximum value and the second minimum value; And obtaining a representative value difference from the first and the second difference, The difference data unit includes the representative extreme value deviation and the representative value difference; the first specific condition includes a first sub-condition and a second sub-condition; the first sub-condition is the representative extreme deviation In a first predetermined value range; the second sub-condition is that the representative value difference falls within a second predetermined value range; the second motion portion is later than the first motion portion; the first motion characteristic value data The unit includes a plurality of specific motion feature values; the motion recognition method further includes the step of: providing a first feature function code data unit in advance by sensing a plurality of body motions occurring at the specific location, wherein: the first feature function code The data unit is associated with the first candidate reference signal code and a first complex identification value region; the first feature function code data unit is configured to indicate one of the first complex identification value regions; the first complex identification value The area includes a first identification value area, a second identification value area adjacent to the first identification value area, and the first and second identification value areas a first trust value area; the first and the second identification value areas respectively indicate a third motion type and a fourth motion type, and the third and the fourth motion types are all included in the plural motion type; The plural body motion is divided into a plurality of motion groups, The complex motion groups respectively belong to the complex motion type; and the first feature function code data unit represents a motion type cutting function and is represented based on a relationship between the complex specific motion feature values, wherein the motion type cutting function And being used to indicate the one of the first plurality of identification value regions; when the second determination result is negative, performing a first based on the first feature function code data unit and the first motion feature value data unit Calculating to generate a first identification value, and determining whether the first identification value belongs to one of the first and second identification value regions to make a third determination result; when the third determination result is affirmative Determining that the first identification value belongs to a valid identification value area in the first and second identification value areas, and identifying that the first motion type is an effective motion type indicated by the valid identification value area, And obtaining the motion type code corresponding to the valid identification value region; generating a motion measurement related to the first body motion based on the motion type code a second feature function code data unit associated with the first candidate reference signal code and a second complex identification value region, wherein: the second feature function code data unit is different from the first feature function code data a unit, and is used to indicate one of the second complex identification value regions; the second complex identification value region includes a third identification value region, and a fourth identification value region adjacent to the third identification value region And in a second trust value region between the third and fourth identification value regions; the third and fourth identification value regions respectively indicating a fifth motion type and a sixth motion type, the fifth and the third a six motion type is included in the complex motion type; and in a second specific condition, one of the third and fourth identification value regions and one of the first and second identification value regions At least partially overlapping; and when the third determination result is negative, performing a second calculation based on the second feature function code data unit and the first motion feature value data unit to generate a second identification value, determining the Whether the second identification value belongs to one of the third and fourth identification value regions, and thereby identifying the first motion type; the candidate reference signal code data unit further includes at least one second candidate reference signal a second candidate reference signal code, wherein: the second candidate reference signal code represents the second candidate reference signal originating from the motion parameter signal structure; and the first candidate reference signal and The at least one second candidate reference signal constitutes a candidate reference signal combination; and the motion recognition method further comprises the steps of: determining the at least one second candidate reference signal based on the motion parameter signal structure and the at least one second candidate reference signal code Obtaining, according to the at least one second candidate reference signal, at least one second motion feature value data unit corresponding to the at least one second candidate reference signal from the motion parameter signal structure; When the first motion feature value data unit is processed, the at least one second motion feature value data unit is processed to determine whether the candidate reference signal combination includes the valid reference signal.