KR20180086547A - Wearable device for gesture recognition and control and gesture recognition control method using the same - Google Patents

Abstract

The present invention relates to a wearable device to recognize and control a motion, capable of correctly recognizing the motion, and a motion recognition control method using the same. In particular, the wearable device to recognize and control a motion comprises: an electromyogram (EMG) sensor module measuring an EMG signal in accordance with muscle movement of a user and converting the EMG signal into a digital signal to output the digital signal; an inertial motion unit (IMU) sensor module measuring and converting a motion signal of the user into a digital signal to output the digital signal; a control module to receive the EMG signal of the EMG sensor module and the motion signal of the IMU sensor module, and to perform filtering for removing a motion artifact included in the EMG signal based on the motion signal; and a computer device unit receiving the EMG signal, from which the motion artifact is removed, and the motion signal from the control module through wireless short-range communication to recognize a motion of the EMG signal through feature extraction, feature classification, and data selection mapping, and performing control corresponding to the motion.

Description

TECHNICAL FIELD [0001] The present invention relates to a wearable device for recognizing and controlling a motion, and a motion recognition control method using the wearable device.

The present invention relates to a wearable device for motion recognition and control and a motion recognition control method using the same. More particularly, the present invention relates to a wearable device for motion recognition and control using an EMG-based control interface The present invention relates to a wearable device for recognizing and controlling movement and a motion recognition control method using the wearable device.

Due to the development of electronic technology, the interaction between the machine and the user has become more diverse. In particular, as the interest in human-computer interaction (HCI) technology has increased recently, various researches are being conducted on interaction that can enhance user's intellectual, emotional, and life experience. These HCIs are making great efforts to develop user-friendly interfaces that use voice, vision, gestures, or other innovative input / output channels. One of the most difficult approaches in this area of research is to connect the human neural signals to the computer using the electrical properties of the human nervous system. A variety of biomedical signals can be used to connect these neurons to computers, most of which can be obtained from cell tissues, such as specific body tissues, organs or nervous systems.

The biological signal refers to an electrical or magnetic signal generated in the human body and typically includes signals such as EMG, ECG, EDG, EOG, and GSR . Such bio-signals have been mainly used for medical treatment or rehabilitation purposes in the medical field. In recent years, however, they have been used for inferring the intention of the user in the human-computer interface (HCI) field and controlling the operation of a computer, a machine, And its application range is widening. In order to control a computer, a machine, a robot, or the like using such a bio-signal, it is very important to accurately not only detect a bio-signal accurately but also accurately infer the user's operation intention from the detected bio-signal.

The simplest method of inferring the user's operation intention is to determine that the user has started the operation when the specific factor value of the biological signal exceeds the preset threshold value and perform the control operation corresponding to the operation . However, bio-signals (EMG signals) contain complex types of noise caused by device-specific noise, electromagnetic radiation, motion noise, and interaction with other tissues. Thus, there is a problem that it is difficult to set the accurate threshold value because the noise included in the bio-signals such as EMG continuously changes according to the surrounding environment, the physical condition of the user, and the contact state of the sensor and the body. It is necessary to provide a method for accurately determining whether or not the operation of the vehicle is performed. Korean Patent Laid-Open Publication No. 10-2016-0133306 and Korean Patent Registration No. 10-1000869 are disclosed in the prior art documents.

The present invention has been proposed in order to solve the above-mentioned problems of the previously proposed methods, and has an EMG sensor module for measuring an EMG signal, an IMU sensor module for measuring a motion signal, a control module for performing pre- And a computer device unit for recognizing and controlling the operation of the gesture based on the EMG signal, thereby recognizing and controlling the motion of the robot, removing the noise included in the EMG signal used for motion recognition, And an operation recognition control method using the wearable apparatus.

Also, according to the present invention, an EMG sensor module, an IMU sensor module, and a control module are integrated into a single case, a single-channel EMG sensor is positioned on an arm band of a user, and an IMU sensor is positioned on an arm, Based control interface is wearable and can be easily used for attaching and detaching. In addition, it is possible to recognize motion and control that accurately recognize a subtle gesture without movement in a manner that an EMG signal and an IMU signal are used in combination And an operation recognition control method using the wearable apparatus.

According to an aspect of the present invention, there is provided a wearable device for recognizing and controlling operation,

1. A wearable device for motion recognition and control,

An EMG (electromyogram) sensor module for measuring an EMG signal according to a user's muscle movement, converting the measured EMG signal to digital, and outputting the EMG signal;

An IMU (Inertial Motion Unit) sensor module that measures a user's motion signal, converts the measured motion signal into a digital signal, and outputs the digital signal;

A control (MCU) module that receives an EMG signal of the EMG sensor module and a motion signal of the IMU sensor module and performs filtering to remove dynamic artifacts included in the EMG signal based on the motion signal; And

And receives the EMG signal and the motion signal from which the dynamic noise has been removed from the control module through wireless short distance communication and recognizes the operation of the EMG signal through the feature extraction, classification, and data selection mapping process and performs control corresponding thereto And a computer device unit.

Preferably, the EMG sensor module includes:

And a single channel EMG sensor positioned in the user's arm band.

Preferably, the EMG sensor module includes:

Raw EMG signal measured according to the user's muscle movement is attached to the user's arm and then the rectified signal is amplified and the filtered electromyogram signal is converted into a digital signal using an A / . ≪ / RTI >

More preferably, the EMG sensor module includes:

A differential amplifier can be used as an amplifier for amplifying the rectified signal.

Preferably, the IMU sensor module further comprises:

The IMU sensor is located at the forearm of the user and measures the degree of user's motion. The IMU sensor may be an accelerometer, a gyroscope, or a magnetometer.

More preferably, the motion signal includes:

And may be used as a reference noise signal for correcting dynamic artifacts included in an EMG signal measured by the EMG sensor module.

Advantageously, the control module comprises:

And a Bluetooth module for transmitting an electromyogram signal and a motion signal, which have undergone a preprocessing process for removing unnecessary noise including dynamic noise included in the EMG signal, to the computer unit.

More preferably, the control module includes:

And a Lily Pad Simblee BLE board integrated with the low power Bluetooth 4.0 for full power communication with the computer unit.

Preferably,

The EMG sensor module, the IMU sensor module, and the control (MCU) module may be constructed in a wearable type which is accommodated in a cylindrical case and fastened to a user's forearm.

More preferably, the cylindrical case includes:

And can be manufactured as a 3D printer.

More preferably, the cylindrical case includes:

And a fastening member capable of being detachably attached to the forearm of the user.

More preferably, the computer device unit comprises:

(EMU) signal received through the Bluetooth communication from the control module to classify the gesture according to the motion of the user and classify the gesture taken by the user using the motion (IMU) signal Data selection and mapping can be performed with reference to a preset data sheet.

Still more preferably, the computer device unit further comprises:

Judges and recognizes the gesture of the electromyogram signal based on the data classified through the feature extraction of the gesture and the result obtained through the data selection and mapping processing with reference to the preset data sheet, Can be executed on the control application.

Even more preferably,

By combining the EMG signal and the motion (IMG) signal, it is possible to detect isometric muscle activity that does not appear to be an actual motion, classify subtle gestures that do not follow the motion, .

Even more preferably,

As a feature extraction of the gesture according to the user's muscle movement using the EMG signal, wavelet transform of time-frequency analysis can be applied.

Even more preferably, the gesture classification comprises:

hand close gestures, hand open gestures, and hand extention gestures.

Even more preferably,

The classification of the feature extraction of the gesture according to the user's muscle movement using the EMG signal can be classified using the K-nearest neighbor (KNN) algorithm.

According to another aspect of the present invention, there is provided a method of controlling an operation recognition using a wearable device,

An operation recognition control method using a wearable device having an EMG sensor module, an IMU sensor module, a control (MCU) module, and a computer unit,

(1) receiving the EMG signal converted into a digital signal by measuring the movement of the user's forearm through the EMG sensor module from the control module;

(2) receiving the motion (IMU) signal of which the degree of motion of the user is measured from the IMU sensor module and converted into a digital signal by the control module;

(3) The control module performs a preprocessing process for removing unnecessary noise including dynamic noise included in the EMG signal based on the motion signal, and outputs the preprocessed EMG signal and the motion signal To the computer unit; And

(4) The computer device receives the dynamic noise-free electromyogram signal and the motion signal through the wireless local area communication from the control module through the step (3), and performs the feature extraction, classification, and data selection mapping processing And recognizing an operation of the electromyogram signal and performing control corresponding thereto.

Preferably, in the step (4)

The computer device extracts and classifies gesture characteristics according to the movement of the user's muscles using an EMG signal received through the Bluetooth communication from the control module, A data selection and mapping process is performed with reference to a preset data sheet to distinguish the gestures and the result obtained through data selection and mapping process with reference to the data classified through the feature extraction of the gesture and the preset data sheet And recognizes the gesture of the electromyogram signal and recognizes the gesture of the electromyogram signal, and executes an operation process corresponding to the determined gesture on the control application.

More preferably, the computer device unit comprises:

A wavelet transform of time-frequency analysis is applied to extract a gesture according to a user's muscle movement using the electromyogram (EMG) signal, and a feature extraction of a gesture according to a user's muscle movement using the EMG signal And can be classified using the K-nearest neighbor (KNN) algorithm.

According to the wearable device for motion recognition and control proposed in the present invention and the motion recognition control method using the same, an EMG sensor module for measuring an EMG signal, an IMU sensor module for measuring a motion signal, A control module, and a computer unit for recognizing and controlling the operation of the gesture based on the EMG signal, thereby making it possible to recognize the motion and remove the noise included in the EMG signal used for the control to enable more accurate motion recognition .

In addition, according to the present invention, the EMG sensor module, the IMU sensor module, and the control module are integrated into a single case, a single-channel EMG sensor is positioned on the arm band of the user, and an IMU sensor is positioned on the forearm, The EMG-based control interface is wearable and can be conveniently used for attaching and detaching. In addition, the EMG signal and the IMU signal can be used in combination to accurately recognize subtle gestures without motion.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a functional block diagram illustrating a configuration of a wearable device for recognizing and controlling operations according to an embodiment of the present invention; FIG.
BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a wearable device for recognizing and controlling operation of a wearable device.
3 is a view showing a hand gesture recognized as an operation in a wearable device for recognizing and controlling a motion according to an embodiment of the present invention;
4 is a view showing an operation state of an arm in a wearing state of a wearable device for recognizing and controlling operation according to an embodiment of the present invention;
FIG. 5 is a view showing another operation state of the wearable wearable device for recognition and control of operation according to an embodiment of the present invention; FIG.
6 is a view showing another operation state of the wearable wearable wearable device for recognizing and controlling operation according to an embodiment of the present invention.
FIG. 7 illustrates training data of a wearable device for recognition and control of movement according to an embodiment of the present invention; FIG.
FIG. 8 is a graphical representation of gesture recognition scores of a wearable device for recognizing and controlling actions according to an embodiment of the present invention; FIG.
9 is a flowchart illustrating an operation of the wearable apparatus according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings, in order that those skilled in the art can easily carry out the present invention. In the following detailed description of the preferred embodiments of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear. In the drawings, like reference numerals are used throughout the drawings.

In addition, in the entire specification, when a part is referred to as being 'connected' to another part, it may be referred to as 'indirectly connected' not only with 'directly connected' . Also, to "include" an element means that it may include other elements, rather than excluding other elements, unless specifically stated otherwise.

FIG. 1 is a functional block diagram illustrating a configuration of a wearable device for recognizing and controlling operations according to an exemplary embodiment of the present invention. FIG. 2 is a block diagram of a wearable device for recognizing and controlling operations according to an exemplary embodiment of the present invention. FIG. 3 is a view showing a hand gesture recognized as an operation in a wearable device for recognizing and controlling a movement according to an embodiment of the present invention, and FIGS. 4 to 6 are views FIG. 8 is a view showing various operating states of an arm in a wearing state of a wearable device for recognizing and controlling operations according to an embodiment. 1 and 2, the wearable apparatus 100 for recognizing and controlling operations according to an embodiment of the present invention includes an EMG sensor module 110, an IMU sensor module 120, a control unit (MCU) ) Module 130, and a computer device unit 140. [0034]

The EMG sensor module 110 measures an electromyogram (EMG) signal according to the movement of the user's muscles, converts the measured electromyogram signal into a digital signal, and outputs the digital signal to the control module 130. The EMG (electromyogram) sensor module 110 may include a single-channel EMG sensor positioned in an arm band of a user, as shown in FIGS. Here, the EMG sensor module 110 is attached to a user's arm and rectifies a measured Raw EMG signal according to a movement of the user's muscle, amplifies the rectified signal, and then outputs the filtered EMG signal It can be converted into a digital signal by using an A / D converter. At this time, the EMG sensor module 110 can use a differential amplifier as an amplifier for amplifying the rectified signal.

The IMU sensor module 120 measures a user's motion signal, converts the measured motion signal into a digital signal, and outputs the digital signal to the control module 130. As shown in FIGS. 4 to 6, the IMU sensor module 120 is an IMU sensor for measuring the degree of user's movement, which is located at the forearm of a user and includes an accelerometer, a gyroscope ), And a magnetometer, as shown in FIG. Here, the motion signal may be used as a reference noise signal for correcting the dynamic artifact included in the EMG signal measured by the EMG sensor module 110.

The control (MCU) module 130 receives the EMG signal of the EMG sensor module 110 and the motion signal of the IMU sensor module 120, and receives a dynamic artifact included in the EMG signal based on the motion signal Gt; processing module < / RTI > The control module 130 includes a Bluetooth module (not shown) for transmitting an EMG signal and a motion signal, which are subjected to a preprocessing process for eliminating unnecessary noise including dynamic noise included in an EMG signal, (131). 2, the control module 130 is implemented as a Lily Pad Simblee BLE board integrated with the low power Bluetooth 4.0 for full power communication with the computer unit 140 .

4 to 6, the EMG sensor module 110, the IMU sensor module 120, and the control (MCU) module 130 are accommodated in a cylindrical case 101, And can be configured to be wearable. The cylindrical case 101 may be made of a 3D printer and may further include a fastening member 102 that can be detachably attached to a user's forearm.

The computer unit 140 receives the EMG signal and the motion signal from which the dynamic noise has been removed from the control module 130 through wireless short distance communication and performs the operation of the EMG signal through the feature extraction, And performs control corresponding thereto. The computer unit 140 extracts and classifies the gesture characteristics according to the user's muscle movement using the EMG signal received through the Bluetooth communication from the control module 130 and uses the motion (IMU) signal And perform data selection and mapping with reference to a preset data sheet to distinguish gestures taken by the user.

The computer device unit 140 determines and recognizes the gesture of the EMG signal on the basis of the data classified through the feature extraction of the gesture and the result obtained through the data selection and mapping performing process referring to the preset data sheet, An operation process corresponding to the judged and recognized gesture can be executed on the control application. This computer unit 140 combines an EMG signal and a motion (IMG) signal to detect isometric muscle activity that does not appear to be an actual motion, classifies subtle gestures that have no motion, The control can be performed in a state in which no disturbance occurs.

In addition, the computer device 140 may extract wavelet transform of time-frequency analysis as feature extraction of a gesture according to a user's muscle movement using an EMG signal. At this time, as shown in FIG. 3, the gesture classification may be classified into three classes: a hand close gesture, a hand open gesture, and a hand extention gesture. The wavelet transform (WT) is an easy time-frequency analysis method for analyzing an abnormal signal such as an EMG signal. An advantage obtained by using the wavelet transform is that it is easy to set the dimension of the feature vector, It is possible to obtain almost the same result as in the case of using the feature vector of the dimension. That is, the wavelet transform is developed from a fourier analysis and applied for a short time interval, and the size of the window can be modified through real time frequency analysis. As a result, the analysis of a signal having a high frequency can yield the same result as the analysis in a low frequency band. In order to distinguish the gestures taken by the user, a clear feature must be taken from each matched pattern, so that several features can be extracted. Among them, general statistical features such as maximum value, minimum value, root mean square .

In addition, the computer device unit 140 may classify the feature extraction using the K-nearest neighbor (KNN) algorithm as a classification of the gesture feature extraction according to the user's muscle movement using the EMG signal. In other words, SVM (Support Vector Machine) classification algorithm can be used as a linear or nonlinear method using Kerenel, but there is a disadvantage that many data learning is required for accurate results. In the present invention, the result of KNN is compared with the result of SVM using KNN algorithm, and it is found through experiments that KNN is better than SVM. Because KNN itself is nonlinear, it can easily detect linearly or nonlinearly distributed data and requires fewer data points to get good results.

FIG. 7 is a view showing training data of a wearable device for recognizing and controlling a movement according to an embodiment of the present invention, and FIG. 8 is a graph showing the training data of the wearable device for gesture recognition and control according to an embodiment of the present invention And a recognition score in a graph. FIG. 7 is a table showing each gesture group corresponding to hands 0, 1, and 2 of a hand close gesture, a hand open gesture, and a hand extension gesture after preprocessing and feature extraction as training data. After learning the data (MAX, MIN, MAV, RMS, STD), select which function will produce the best results. FIG. 8 shows the scores of the model to confirm the accuracy of the gesture recognition, and the variance with the scores.

9 is a flowchart illustrating an operation recognition control method using a wearable device according to an embodiment of the present invention. 9, an operation recognition control method using a wearable device according to an embodiment of the present invention includes a step S110 of receiving an EMG signal from a control module 130, A step S130 of receiving the motion (IMU) signal (S120), transmitting the electromyogram signal and the motion signal, which have been subjected to the preprocessing process, to the computer unit 140 by the control and control module 130, (S140) of recognizing the operation of the EMG signal through the feature extraction, classification, and data selection mapping processing using the EMG signal and the motion signal from which the dynamic noise is removed, and performing control corresponding thereto .

1 to 6, the wearable device 100 used in the motion recognition control method of the present invention includes an EMG sensor module 110, an IMU sensor module 120, (MCU) module 130 and a computer device unit 140. [ Hereinafter, the operation recognition control method using the wearable device 100 described above will be described in detail.

In step S110, the control module 130 receives EMG signals, which are measured from the EMG sensor module 110 through movement of the muscles of the user's forearm and converted into digital signals. Here, the EMG (Electromyogram) sensor module 110 may include a single-channel EMG sensor positioned in an arm band of a user, as shown in FIGS. Here, the EMG sensor module 110 is attached to a user's arm and rectifies a measured Raw EMG signal according to a movement of the user's muscle, amplifies the rectified signal, and then outputs the filtered EMG signal It can be converted into a digital signal by using an A / D converter. At this time, the EMG sensor module 110 can use a differential amplifier as an amplifier for amplifying the rectified signal.

In step S120, the control module 130 receives the motion (IMU) signal, which is measured by the user's motion level from the IMU sensor module 120 and converted into a digital signal. As shown in FIGS. 4 to 6, the IMU sensor module 120 is an IMU sensor for measuring the degree of user's movement, which is located at the forearm of the user, and includes an accelerometer, a gyroscope A gyroscope, and a magnetometer.

In step S130, the control module 130 performs a preprocessing process for removing unnecessary noise including the dynamic noise included in the EMG signal based on the motion signal, and outputs the preprocessed EMG signal and motion And transmits the signal to the computer device unit 140. Here, the control module 130 may further include a Bluetooth module 131 for transmitting an EMG signal and a motion signal, which have undergone a preprocessing process, to the computer device unit 140. At this time, the control module 130 may be implemented as a Lily Pad Simblee BLE board integrated with the low power Bluetooth 4.0 for full power communication with the computer unit 140.

In step S140, the computer device 140 receives the electromyogram signal and the motion signal from which the dynamic noise has been removed from the control module 130 through the wireless local area communication through step S130, and performs a feature extraction, classification, And recognizes the operation of the electromyogram signal and performs control corresponding thereto. In this step S140, the computer device 140 extracts and classifies the gesture characteristics according to the user's muscle movement using the EMG signal received through the Bluetooth communication from the control module 130, ) Signal to perform data selection and mapping with reference to a preset data sheet in order to distinguish gestures taken by the user, and performs data selection and mapping processing with reference to the data classified by the feature extraction of the gesture and the preset data sheet The gesture of the electromyogram signal is judged and recognized based on the result obtained through the gesture detection process, and the gesture corresponding to the recognized gesture is judged and executed on the control application.

In addition, the computer device 140 extracts gestures according to a user's muscle movement using an EMG signal. The computer device 140 applies wavelet transform of time-frequency analysis, As a classification of gesture feature extraction based on motion, it can be classified using K-nearest neighbor (KNN) algorithm.

As described above, the wearable device and the motion recognition control method for motion recognition and control according to an embodiment of the present invention include an EMG sensor module for measuring an EMG signal, an IMU sensor module for measuring a motion signal, And a computer unit for recognizing and controlling the operation of the gesture based on the EMG signal, thereby making it possible to eliminate the noise included in the EMG signal used for motion recognition and control, In addition, the EMG sensor module, the IMU sensor module and the control module are integrated into a single case, and a single-channel EMG sensor is positioned on the user's arm band and an IMU sensor is positioned on the forearm , The EMG-based control interface is wearable and can be conveniently used for attaching and detaching, A call and how the IMU signals are combined using a subtle gesture no motion is possible to ensure that they accurately recognized.

The present invention may be embodied in many other specific forms without departing from the spirit or essential characteristics and scope of the invention.

100: a wearable device according to an embodiment of the present invention
101: Cylindrical case
102: fastening member
110: EMG sensor module
120: IMU sensor module
130: Control (MCU) module
131: Bluetooth module
140:
S110: The control module receives the EMG signal
S120: the control module receives the motion (IMU) signal
S130; A step of transmitting the electromyogram signal and the motion signal subjected to the preprocessing process to the computer device unit
S140: recognizing the operation of the EMG signal through the feature extraction, classification, and data selection mapping processing using the EMG signal and the motion signal from which the computer device has removed the dynamic noise, and performing the corresponding control

Claims (20)

A wearable device (100) for motion recognition and control,
An EMG (electromyogram) sensor module 110 for measuring an EMG signal according to a user's muscle movement, converting the measured EMG signal to digital, and outputting the digital EMG signal;
An IMU (Inertial Motion Unit) sensor module 120 for measuring a user's motion signal, converting the measured motion signal into a digital signal, and outputting the digital signal;
Receives EMG signal of the EMG sensor module 110 and a motion signal of the IMU sensor module 120 and performs filtering to remove dynamic artifacts included in the EMG signal based on the motion signal A control (MCU) module 130; And
The control unit 130 receives the EMG signal and the motion signal from which the dynamic noise has been removed through the WLAN and recognizes the EMG signal by the feature extraction, classification, and data selection mapping process, And a computer device (140) for performing the operation of the wearable device.
The apparatus of claim 1, wherein the EMG sensor module (110)
And a single-channel EMG sensor positioned in an arm band of the wearer's body.
The apparatus of claim 1, wherein the EMG sensor module (110)
Raw EMG signal measured according to the user's muscle movement is attached to the user's arm and then the rectified signal is amplified and the filtered electromyogram signal is converted into a digital signal using an A / To the wearable apparatus for operation recognition and control.
The apparatus of claim 3, wherein the EMG sensor module (110)
Characterized in that a differential amplifier is used as an amplifier for amplifying the rectified signal.
The system of claim 1, wherein the IMU sensor module (120)
An IMU sensor for measuring the degree of user's movement, which is positioned at the forearm of a user and is constituted by an accelerometer, a gyroscope, and a magnetometer. Wearable device.
6. The method of claim 5,
Is used as a reference noise signal for correcting dynamic artifact included in an EMG signal measured by the EMG sensor module (110).
The apparatus of claim 1, wherein the control module (130)
A Bluetooth module 131 for transmitting an electromyogram signal and a motion signal to the computer unit 140, which has undergone a preprocessing process for removing unnecessary noise including dynamic noise included in the EMG signal, Wherein the wearable device is a wearable device for operation recognition and control.
8. The apparatus of claim 7, wherein the control module (130)
And a Lily Pad Simblee BLE board integrated with the low power Bluetooth 4.0 for full power communication with the computer device unit 140. The wearable device of claim 1,
9. The method according to any one of claims 1 to 8,
The EMG sensor module 110, the IMU sensor module 120, and the MCU module 130 are housed in a cylindrical case 101 and are wearable to be fastened to a user's forearm Characterized in that the wearable device for motion recognition and control.
10. The apparatus according to claim 9, wherein the cylindrical case (101)
A wearable device for operation recognition and control, characterized in that the device is made of a 3D printer.
10. The apparatus according to claim 9, wherein the cylindrical case (101)
The wearable device for operation recognition and control according to claim 1, further comprising a fastening member (102) capable of being detachably attached to the forearm of the user.
10. The apparatus according to claim 9, wherein the computer device (140)
Extracts and classifies gesture characteristics according to a user's muscle movement using an EMG signal input through the Bluetooth communication from the control module 130 and outputs the gesture taken by the user using the motion (IMU) Wherein the data selection and mapping are performed with reference to a data sheet set in advance for identification.
13. The apparatus according to claim 12, wherein the computer device (140)
Judges and recognizes the gesture of the electromyogram signal based on the data classified through the feature extraction of the gesture and the result obtained through the data selection and mapping processing with reference to the preset data sheet, Wherein the control unit executes the operation process for performing the operation on the control application.
14. The apparatus according to claim 13, wherein the computer device (140)
By combining the EMG signal and the motion (IMG) signal, it is possible to detect isometric muscle activity that does not appear to be an actual motion, classify subtle gestures that do not follow the motion, The wearable device being capable of recognizing and controlling the operation of the wearable device.
14. The apparatus according to claim 13, wherein the computer device (140)
And a wavelet transform of time-frequency analysis is applied as a feature extraction of a gesture according to a user's muscle movement using the electromyogram (EMG) signal.
16. The method of claim 15,
a hand open gesture, a hand open gesture, a hand open gesture, and a hand extention gesture.
16. The apparatus according to claim 15, wherein the computer device (140)
Wherein the classifying process is performed using a K-nearest neighbor (KNN) algorithm as a classification of gesture feature extraction according to a user's muscle movement using the EMG signal.
1. An operation recognition control method using a wearable device (100) including an EMG sensor module (110), an IMU sensor module (120), a control (MCU) module (130), and a computer device part
(1) receiving the EMG signal, which is measured by the control module 130 from the EMG sensor module 110 through movement of the muscles of the forearm and converted into a digital signal;
(2) receiving the motion (IMU) signal in which the degree of motion of the user is measured by the control module 130 from the IMU sensor module 120 and converted into a digital signal;
(3) The control module 130 performs a preprocessing process for removing unnecessary noise including dynamic noise included in the EMG signal based on the motion signal, and outputs the preprocessed EMG signal And a motion signal to the computer device unit 140; And
(4) The computer device 140 receives the dynamic noise-free electromyogram signal and the motion signal from the control module 130 through the wireless local area communication in step (3), and performs feature extraction, classification, Recognizing an operation of the electromyogram signal through a mapping process of data selection, and performing a control corresponding to the operation of the electromyogram signal.
19. The method of claim 18, wherein in step (4)
The computer unit 140 extracts and classifies gesture characteristics according to a user's muscle movement using an EMG signal received through the Bluetooth communication from the control module 130, And performs data selection and mapping with reference to a preset data sheet in order to identify a gesture taken by the user using the signal and performs data selection and mapping processing with reference to data classified by the feature extraction of the gesture and a preset data sheet Judges and recognizes the gesture of the electromyogram signal on the basis of the result obtained through the operation of the wearable device, and executes an operation process corresponding to the judged and recognized gesture on the control application.
The computer system according to claim 19, wherein the computer device (140)
A wavelet transform of time-frequency analysis is applied to extract a gesture according to a user's muscle movement using the electromyogram (EMG) signal, and a feature extraction of a gesture according to a user's muscle movement using the EMG signal Wherein the classification process is performed using a K-nearest neighbor (KNN) algorithm as a classification of the wearable device.

Images