CN111026268A - Gesture recognition device and method - Google Patents
Gesture recognition device and method Download PDFInfo
- Publication number
- CN111026268A CN111026268A CN201911216438.2A CN201911216438A CN111026268A CN 111026268 A CN111026268 A CN 111026268A CN 201911216438 A CN201911216438 A CN 201911216438A CN 111026268 A CN111026268 A CN 111026268A
- Authority
- CN
- China
- Prior art keywords
- module
- gesture
- digital signal
- unit
- signal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/017—Gesture based interaction, e.g. based on a set of recognized hand gestures
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/389—Electromyography [EMG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/6802—Sensor mounted on worn items
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Biophysics (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Signal Processing (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Psychiatry (AREA)
- Human Computer Interaction (AREA)
- General Physics & Mathematics (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The application discloses a gesture recognition device and method, comprising: the signal acquisition module comprises a plurality of electrodes and is used for acquiring surface electromyographic signals and sending the surface electromyographic signals to the analog front-end module; the analog front end module is used for receiving the surface electromyographic signals, preprocessing the surface electromyographic signals to obtain digital signals and sending the digital signals to the interface module; the interface module is used for sending the received digital signal to the processing module; the processing module is used for determining a gesture result corresponding to the digital signal and sending the gesture result to the wireless transmission module; the wireless transmission module is used for sending the gesture result and collecting the surface electromyographic signal by using the electrode. The electrodes are used for collecting surface electromyographic signals, no wound is caused, wearing is convenient, gesture results corresponding to the digital signals are determined by using a gradient lifting decision tree algorithm, other detection equipment is not needed, and the method is suitable for mobile application program scenes.
Description
Technical Field
The present application relates to the field of gesture recognition, and in particular, to a gesture recognition apparatus and method.
Background
Gestures are one of the common ways of communicating between people. For example, sign language is widely used by language handicapped persons. Compared to using traditional mechanical devices (such as a mouse and/or keyboard), automatic gesture recognition may be used as a human-machine interface, providing a more natural, efficient solution for information exchange between a user and a device.
The existing gesture recognition method comprises the following steps: computer vision based methods, mechanical based methods, ultrasonic based methods, and the like. Computer vision based methods have high accuracy performance but are limited by the field of view of the image acquisition device. Furthermore, if three-dimensional recognition is required, multiple camera systems or ultrasound machines are typically employed to detect depth of field. However, for mobile application scenarios, a huge hardware setup is not possible. Mechanical-based approaches, such as smart gloves, integrated accelerometers, and/or gravity sensors, can provide wearable solutions. However, the applications of smart gloves are still limited due to the size limitation of the accelerometer. While the ultrasound-based approach can reduce the cost of other detection devices, it is also not suitable for mobile application scenarios.
In view of the foregoing, it is desirable to provide a gesture recognition apparatus and method that is small in size and suitable for mobile application scenarios.
Disclosure of Invention
In order to solve the above problems, the present application provides a gesture recognition apparatus and method.
In one aspect, the present application provides a gesture recognition apparatus, including: a signal acquisition module, an analog front end module, an interface module, a processing module and a wireless transmission module, wherein,
the signal acquisition module is sequentially connected with the analog front-end module, the interface module and the processing module one by one, and the wireless transmission module is connected with the interface module;
the signal acquisition module comprises a plurality of electrodes and is used for acquiring surface electromyographic signals and sending the surface electromyographic signals to the analog front-end module;
the analog front-end module is used for receiving the surface electromyogram signal, preprocessing the surface electromyogram signal to obtain a digital signal and sending the digital signal to the interface module;
the interface module is used for sending the received digital signals to the processing module;
the processing module is used for determining a gesture result corresponding to the digital signal and sending the gesture result to the wireless transmission module;
and the wireless transmission module is used for sending the gesture result.
Preferably, the analog front end module comprises:
the first amplification unit is used for receiving the surface electromyogram signal sent by the electrode connected with the first amplification unit, inhibiting common-mode noise of the received surface electromyogram signal and amplifying the surface electromyogram signal, and sending the processed surface electromyogram signal to the multiplexing unit;
the multiplexing unit is used for selecting one from the received surface electromyographic signals to output and sending the selected signal to the second amplifying unit;
a second amplification unit for amplifying the received surface electromyogram signal;
the analog-to-digital conversion unit is used for converting the amplified surface electromyographic signals into digital signals and sending the digital signals to the digital logic unit;
and the digital logic unit is used for adding a channel identifier to the received digital signal and sending the digital signal added with the channel identifier to the interface module.
Preferably, the processing module comprises:
the preprocessing unit is used for segmenting the digital signal to obtain a digital signal segment;
the feature extraction unit is used for extracting the gesture features of the digital signal segments;
and the gesture processing unit is used for determining a gesture result corresponding to the gesture characteristics and sending the gesture result to the wireless transmission module.
Preferably, the gesture processing unit is further configured to receive the trained model parameters, and update the gesture processing unit according to the model parameters.
Preferably, the device further comprises a storage interface module, wherein the storage interface module is connected with the interface module and used for accessing the storage module and sending the digital signal sent by the interface module to the storage module.
Preferably, the interface module further reduces the sampling rate of the digital signal and then sends the reduced sampling rate to the wireless transmission module; the interface module includes:
the interface unit is used for receiving the digital signal and sending the digital signal to the filtering unit or the cache unit;
the buffer unit is used for buffering the received digital signals and sending the buffered digital signals to the processing module or the storage interface module;
and the filtering unit is used for reducing the sampling rate of the digital signal and sending the digital signal with the reduced sampling rate to the wireless transmission module.
Preferably, the wireless transmission module is further configured to transmit the digital signal with the reduced sampling rate; and receiving the trained model parameters and sending the trained model parameters to a processing module.
In a second aspect, the present application provides a gesture recognition method, including:
the signal acquisition module acquires surface electromyographic signals and sends the surface electromyographic signals to the analog front-end module;
the analog front-end module preprocesses the surface electromyogram signal to obtain a digital signal and sends the digital signal to the interface module;
the interface module sends the received digital signal to the processing module;
the processing module determines a gesture result corresponding to the digital signal and sends the gesture result to the wireless transmission module;
and the wireless transmission module sends the gesture result.
Preferably, the analog front-end module preprocesses the surface electromyogram signal to obtain a digital signal, and sends the digital signal to the interface module, and the method includes:
the first amplification unit receives a surface electromyogram signal sent by an electrode connected with the first amplification unit, inhibits common-mode noise of the received surface electromyogram signal, amplifies the surface electromyogram signal, and sends the processed surface electromyogram signal to the multiplexing unit;
the multiplexing unit selects one of the received surface electromyographic signals to output, and sends the selected signal to the second amplifying unit;
the second amplifying unit amplifies the received surface electromyographic signals;
the analog-to-digital conversion unit converts the amplified surface electromyographic signals into digital signals and sends the digital signals to the digital logic unit;
the digital logic unit adds a channel identifier to the received digital signal and sends the digital signal added with the channel identifier to the interface module.
Preferably, the processing module determines a gesture result corresponding to the digital signal, and sends the gesture result to the wireless transmission module, including:
the preprocessing unit segments the digital signal to obtain a digital signal segment;
the feature extraction unit extracts the gesture features of each digital signal segment;
the gesture processing unit determines a gesture result corresponding to the gesture feature and sends the gesture result to the wireless transmission module.
The application has the advantages that: the electrodes are used for collecting surface electromyographic signals, the wearing is convenient, corresponding gesture results are determined through digital signals, other detection equipment is not needed, and the method is suitable for mobile application program scenes.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating preferred embodiments and are not to be construed as limiting the application. Also, like reference numerals are used to denote like parts throughout the drawings. In the drawings:
FIG. 1 is a schematic diagram of a gesture recognition apparatus provided herein;
FIG. 2 is a schematic structural diagram of an embodiment of a gesture recognition apparatus provided in the present application;
FIG. 3 is a schematic diagram of an embodiment of a gesture recognition apparatus provided herein;
FIG. 4 is a schematic diagram of 12 gestures of a gesture recognition device provided herein;
FIG. 5 is a schematic diagram of a surface electromyographic signal of a gesture recognition apparatus provided in the present application;
fig. 6 is a schematic step diagram of a gesture recognition method provided in the present application.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
According to an embodiment of the present application, a gesture recognition apparatus is provided, as shown in fig. 1, including:
a signal acquisition module 101, an Analog Front End (AFE) module 102, an interface module 103, a processing module 104, and a wireless transmission module 105, wherein,
the signal acquisition module is sequentially connected with the analog front-end module, the interface module and the processing module one by one, and the wireless transmission module is connected with the interface module;
the signal acquisition module comprises a plurality of electrodes and is used for acquiring surface muscle (SEMG) signals and sending the SEMG signals to the simulation front-end module;
the analog front-end module is used for receiving the surface electromyographic signals, preprocessing the surface electromyographic signals to obtain digital signals and sending the digital signals to the interface module;
the interface module is used for sending the received digital signals to the processing module;
the processing module is used for determining a gesture result corresponding to the digital signal by using a Gradient Boosting Decision Tree (GBDT) algorithm and sending the gesture result to the wireless transmission module;
and the wireless transmission module is used for sending the gesture result.
The analog front end module includes:
the first amplification unit is used for receiving the surface electromyogram signal sent by the electrode connected with the first amplification unit, inhibiting common-mode noise of the received surface electromyogram signal and amplifying the surface electromyogram signal, and sending the processed surface electromyogram signal to the multiplexing unit;
the multiplexing unit is used for selecting one from the received surface electromyographic signals to output and sending the selected signal to the second amplifying unit;
a second amplification unit for amplifying the received surface electromyogram signal;
the analog-to-digital conversion unit is used for converting the amplified surface electromyographic signals into digital signals and sending the digital signals to the digital logic unit;
and the digital logic unit is used for adding a channel identifier to the received digital signal and sending the digital signal added with the channel identifier to the interface module.
The digital logic unit is also used for generating a control signal according to the clock signal sent by the interface unit and controlling the multiplexing unit and the analog-to-digital conversion unit.
The processing module comprises:
the preprocessing unit is used for segmenting the digital signal to obtain a digital signal segment;
the feature extraction unit is used for extracting the gesture features of the digital signal segments;
and the gesture processing unit is used for determining a gesture result corresponding to the gesture characteristics and sending the gesture result to the wireless transmission module.
And the gesture processing unit is also used for receiving the trained model parameters and updating the gesture processing unit according to the model parameters.
In another embodiment of the present application, optionally, the apparatus further includes a storage interface module, and the storage interface module is connected to the interface module and is configured to send the digital signal sent by the interface module to the storage module.
The memory module preferably comprises a memory card. The digital signals stored on the memory card are used to train the model off-line.
The interface module is used for reducing the sampling rate of the digital signal and then sending the digital signal to the wireless transmission module; the interface module includes:
the interface unit is used for receiving the digital signal and sending the digital signal to the filtering unit or the cache unit;
the buffer unit is used for buffering the received digital signals and sending the buffered digital signals to the processing module or the storage interface module;
and the filtering unit is used for reducing the sampling rate of the digital signal and sending the digital signal with the reduced sampling rate to the wireless transmission module.
The interface unit is also used for generating two clock signals and a control signal. The control signal is directly sent to the second amplifying unit through the digital logic unit; a clock signal is sent to the digital logic unit for generating a control signal; the other clock signal is sent directly to the first amplification unit.
The wireless transmission module is also used for sending the digital signal with the reduced sampling rate; and receiving the trained model parameters and sending the trained model parameters to a processing module.
The digital signal with the reduced sampling rate is used for off-line training of the model.
As shown in fig. 2, an embodiment of the present application includes: the device comprises a flexible surface electrode belt (flexible arm belt), an analog front-end chip, a Field-Programmable Gate Array (FPGA) and a radio frequency module.
As shown in fig. 3, the signal acquisition module may be a flexible surface electrode strip (flexible arm strip) integrating 32 contacts.
The intelligent SEMG recorder comprises an analog front-end chip and an FPGA.
An Analog front end module, i.e., an Analog front end chip, includes a plurality of Low Noise Amplifiers (LNAs), a Multiplexer (MUX), a Programmable Gain Amplifier (PGA), an Analog-to-Digital Converter (ADC), and Digital logic.
The low noise amplifier (first amplification unit) may be a 16-channel low noise amplifier, each channel connecting two electrodes as differential inputs of the low noise amplifier. The received surface electromyogram signal raw signal is amplified by 40 dB. Due to the high common mode rejection ratio of the low noise amplifier, the original signal will cancel most of the common mode noise in the first stage (after passing through the low noise amplifier).
The number of the first amplifying units is determined according to the number of channels and the required number of channels, and may be one first amplifying unit or a plurality of first amplifying units.
In order to improve the area efficiency of the analog front end, channel (channel) control is performed by a multiplexer, so that all channels of the noise amplifier share the programmable gain amplifier and the analog-to-digital converter.
The gain of the programmable gain amplifier (second amplifying unit) is adjustable, 4 gears can be selected, 6dB to 24dB can be selected, and the gain can be amplified by 46dB to 64dB in total when the gain is added with a low noise amplifier.
The analog-to-digital converter may preferably use a Successive Approximation Register (SAR) analog-to-digital converter with 12-bit resolution, each channel of which has a maximum sampling rate of 20kHz, and transmits the received surface electromyogram signal to digital logic after performing analog-to-digital conversion.
The digital logic, preferably comprising a 16 bit parallel interface, adds a 4 bit address header (channel identification) to the received 12 bit digital signal and sends it to the interface module.
The FPGA includes: an interface module and a Neural Signal Processing Unit (NSPU) based on FPGA.
The interface module includes: collector interface, filter and buffer.
The neural signal processing unit (processing module) includes: the device comprises a model loading unit, a processing unit and a result generating unit.
The result generation unit sends the generated gesture result to the wireless module.
The wireless module (radio frequency module) is used for wireless configuration and communication, and sending the gesture result to the corresponding equipment.
The corresponding device includes: the mobile phone, the computer, the automatic wheelchair, the unmanned aerial vehicle, the game machine, the earphone, the projector and other terminals and equipment which can be controlled. The software in the terminal device can be controlled by using the gesture result.
Using the armband and analog front-end chip, 16 SEMG signals can be collected around the middle of the lower arm and stored in a memory card or transmitted to the corresponding device by wireless transmission. These data can be processed off-line to obtain information (signals) about the muscle (surface muscle) and to train a suitable recognition model. To save external devices, gesture recognition is implemented on the FPGA near the analog front end.
The gesture processing unit (a processing unit in the Neural signal processing unit) may determine a gesture result corresponding to the gesture feature using a binary and multivariate classification and regression algorithm such as a gradient boosting decision tree, a Support Vector Machine (SVM), a K-neighborhood (K-Nearest Neighbors) algorithm, an Artificial Neural Network (ANN), a Linear Discriminant Analysis (LDA) algorithm, and the like.
The gradient boosting decision tree has good generalization performance and less calculation requirements, and is an enhanced learning algorithm taking a plurality of decision trees as a basic Weak Learner (Base well leaner).
In the following, embodiments of the present application will be further described by taking the use of a gradient boosting decision tree as an example.
The gradient boosting decision tree model is integrated with several sets of Classification And Regression Trees (CART), which are typically binary trees. Each node in the tree, except for the leaf nodes, has a threshold that can be compared to one of the M features. One of the child nodes is selected based on the comparison, and the tree is traversed in this manner until a leaf node is reached. The leaf nodes of a tree have a predicted score for a particular gesture, which represents the probability that the gesture will be a true result. The tree outputs a predicted score and each tree is specific to a gesture.
Taking the gesture set G as an example, preferably 12 gestures are set in the gesture set G, as shown in fig. 4, including: extending thumb, confirm gesture, extending forefinger, victory gesture, making a fist, making a phone call gesture, opening palm, pulling gesture, wrist outward (making a fist), wrist inward, and wrist inward (making a fist).
As shown in fig. 5, there are 16 surface myoelectric signals corresponding to 4 gestures in time domain, which are wrist-out, wrist-in, confirm gesture and victory gesture.
Data from multiple people performing these gestures is collected to yield a raw surface electromyographic signal dataset S, which may be denoted as S { (S)i,k,gi,k)}。
Wherein S isi,kIs a matrix of 16 rows, 16 rows representing signals taken from 16 channels for the ith slice of the kth topic, gi,kE G is the gesture tag for the segment.
Using a sliding window with the length of 100ms, each original data segment S is divided intoi,kDivided into several segments with a 50ms overlap between adjacent segments. Segmented raw surface electromyographic signal dataset S', S { (S)j,k,gj,k)}。
Wherein S isj,kIs the data of the jth segment, and is g if the jth segment is obtained from the ith segmentj,k=gi,k。
Features extracted from surface electromyographic signals can be classified into four categories: time domain related features, autoregressive coefficients (autoregressive coeffients), frequency domain features, and wavelet transforms and other features.
The time domain feature is also less demanding to implement in hardware because of its lower complexity.
The time domain feature is taken as an example for explanation.
Selecting a plurality of time domain related features for each channel and calculating, and generating m-n × 16 features for each segment, wherein n is the number of the selected time domain related features.
The time domain related features include: mean Absolute Value (MAV), Simple Square Integral (SSI), Minimum Value (Minimum Value), Maximum Value (Maximum Value), Standard Deviation (Standard development), Average Amplitude Change (AAC), zero crossing (zerocross), Slope Sign Change (Slope Sign Change), Willison Amplitude (williamond), and the like.
Generating features for the jth segment (portion) of the kth topicVector quantityWherein R represents a real number set. The generated classification algorithm may directly use the data set D, which may be represented as
Assuming a trained gradient boosting decision tree model, a set F of various trees can be obtained. The set F is composed of several subsets FgA tree for gesture g is included. Given a feature vector zj,kUsing the additive composition model, a predicted score for each gesture is obtained as follows:
After all trees for all gestures are traversed and the prediction score for each gesture is calculated, the maximum prediction score will be the final prediction, which can be generated according to the following formula:
wherein the content of the first and second substances,is the predicted gesture for the jth segment of the kth topic. If it is notThe prediction is correct. Conventional tree traversal first accesses the root node. The child node being the rootAccessed based on higher level comparisons. The delay is proportional to the depth of the tree.
In order to reduce the traversal delay, the path from the root node to the leaf node does not need to be explored layer by layer, and the embodiment of the application preferably adopts a parallel traversal method. The comparison of all leaf nodes is performed simultaneously. For a tree of depth n, there is 2nAn optional output prediction score. An n-bit binary address is used to represent the prediction score preset in the training.
According to an embodiment of the present application, a gesture recognition method is further provided, as shown in fig. 6, including the following steps:
s101, a signal acquisition module acquires surface electromyogram signals and sends the surface electromyogram signals to an analog front-end module;
s102, the analog front-end module preprocesses the surface muscle electrical signal to obtain a digital signal and sends the digital signal to the interface module;
s103, the interface module sends the received digital signal to a processing module;
s104, the processing module determines a gesture result corresponding to the digital signal and sends the gesture result to the wireless transmission module;
and S105, the wireless transmission module sends the gesture result.
Wherein, the analog front end module carries out the preliminary treatment to the surface flesh electricity signal, obtains digital signal, sends interface module to, includes:
the first amplification unit receives a surface electromyogram signal sent by an electrode connected with the first amplification unit, inhibits common-mode noise of the received surface electromyogram signal, amplifies the surface electromyogram signal, and sends the processed surface electromyogram signal to the multiplexing unit;
the multiplexing unit selects one of the received surface electromyographic signals to output, and sends the selected signal to the second amplifying unit;
the second amplifying unit amplifies the received surface electromyographic signals;
the analog-to-digital conversion unit converts the amplified surface electromyographic signals into digital signals and sends the digital signals to the digital logic unit;
the digital logic unit adds a channel identifier to the received digital signal and sends the digital signal added with the channel identifier to the interface module.
Wherein, processing module confirms the gesture result that digital signal corresponds, sends the gesture result to wireless transmission module, includes:
the preprocessing unit segments the digital signal to obtain a digital signal segment;
the feature extraction unit extracts the gesture features of each digital signal segment;
the gesture processing unit determines a gesture result corresponding to the gesture feature and sends the gesture result to the wireless transmission module.
In the embodiment of the application, the electrodes are used for collecting the surface electromyographic signals, no wound is caused, wearing is convenient, gesture results corresponding to the digital signals are determined, other detection equipment is not needed, and the method is suitable for mobile application scenes. And an integrated analog front-end chip is used, so that the size of the equipment can be reduced, and the equipment is convenient to wear.
The above description is only for the preferred embodiment of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A gesture recognition apparatus, comprising: a signal acquisition module, an analog front end module, an interface module, a processing module and a wireless transmission module, wherein,
the signal acquisition module is sequentially connected with the analog front-end module, the interface module and the processing module one by one, and the wireless transmission module is connected with the interface module;
the signal acquisition module comprises a plurality of electrodes and is used for acquiring surface electromyographic signals and sending the surface electromyographic signals to the analog front-end module;
the analog front-end module is used for receiving the surface electromyogram signal, preprocessing the surface electromyogram signal to obtain a digital signal and sending the digital signal to the interface module;
the interface module is used for sending the received digital signals to the processing module;
the processing module is used for determining a gesture result corresponding to the digital signal and sending the gesture result to the wireless transmission module;
and the wireless transmission module is used for sending the gesture result.
2. The gesture recognition device of claim 1, wherein the analog front end module comprises:
the first amplification unit is used for receiving the surface electromyogram signal sent by the electrode connected with the first amplification unit, inhibiting common-mode noise of the received surface electromyogram signal and amplifying the surface electromyogram signal, and sending the processed surface electromyogram signal to the multiplexing unit;
the multiplexing unit is used for selecting one from the received surface electromyographic signals to output and sending the selected signal to the second amplifying unit;
a second amplification unit for amplifying the received surface electromyogram signal;
the analog-to-digital conversion unit is used for converting the amplified surface electromyographic signals into digital signals and sending the digital signals to the digital logic unit;
and the digital logic unit is used for adding a channel identifier to the received digital signal and sending the digital signal added with the channel identifier to the interface module.
3. The gesture recognition apparatus of claim 1, wherein the processing module comprises:
the preprocessing unit is used for segmenting the digital signal to obtain a digital signal segment;
the feature extraction unit is used for extracting the gesture features of the digital signal segments;
and the gesture processing unit is used for determining a gesture result corresponding to the gesture characteristics and sending the gesture result to the wireless transmission module.
4. The gesture recognition device of claim 3, wherein the gesture processing unit is further configured to receive trained model parameters, and update the gesture processing unit according to the model parameters.
5. The gesture recognition device of claim 1, further comprising a storage interface module, connected to the interface module, for accessing the storage module and transmitting the digital signal transmitted by the interface module to the storage module.
6. The gesture recognition device of claim 5, wherein the interface module further reduces a sampling rate of the digital signal and sends the reduced digital signal to the wireless transmission module; the interface module includes:
the interface unit is used for receiving the digital signal and sending the digital signal to the filtering unit or the cache unit;
the buffer unit is used for buffering the received digital signals and sending the buffered digital signals to the processing module or the storage interface module;
and the filtering unit is used for reducing the sampling rate of the digital signal and sending the digital signal with the reduced sampling rate to the wireless transmission module.
7. The gesture recognition device of claim 1, wherein the wireless transmission module is further configured to transmit a digital signal with a reduced sampling rate; and receiving the trained model parameters and sending the trained model parameters to a processing module.
8. A gesture recognition method, comprising:
the signal acquisition module acquires surface electromyographic signals and sends the surface electromyographic signals to the analog front-end module;
the analog front-end module preprocesses the surface electromyogram signal to obtain a digital signal and sends the digital signal to the interface module;
the interface module sends the received digital signal to the processing module;
the processing module determines a gesture result corresponding to the digital signal and sends the gesture result to the wireless transmission module;
and the wireless transmission module sends the gesture result.
9. The gesture recognition method of claim 8, wherein the pre-processing of the surface electromyography signal by the analog front end module to obtain a digital signal, and sending the digital signal to the interface module comprises:
the first amplification unit receives a surface electromyogram signal sent by an electrode connected with the first amplification unit, inhibits common-mode noise of the received surface electromyogram signal, amplifies the surface electromyogram signal, and sends the processed surface electromyogram signal to the multiplexing unit;
the multiplexing unit selects one of the received surface electromyographic signals to output, and sends the selected signal to the second amplifying unit;
the second amplifying unit amplifies the received surface electromyographic signals;
the analog-to-digital conversion unit converts the amplified surface electromyographic signals into digital signals and sends the digital signals to the digital logic unit;
the digital logic unit adds a channel identifier to the received digital signal and sends the digital signal added with the channel identifier to the interface module.
10. The gesture recognition method of claim 8, wherein the processing module determines a gesture result corresponding to the digital signal and sends the gesture result to the wireless transmission module, and the gesture recognition method comprises the following steps:
the preprocessing unit segments the digital signal to obtain a digital signal segment;
the feature extraction unit extracts the gesture features of each digital signal segment;
the gesture processing unit determines a gesture result corresponding to the gesture feature and sends the gesture result to the wireless transmission module.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911216438.2A CN111026268A (en) | 2019-12-02 | 2019-12-02 | Gesture recognition device and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911216438.2A CN111026268A (en) | 2019-12-02 | 2019-12-02 | Gesture recognition device and method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111026268A true CN111026268A (en) | 2020-04-17 |
Family
ID=70203972
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911216438.2A Pending CN111026268A (en) | 2019-12-02 | 2019-12-02 | Gesture recognition device and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111026268A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113589920A (en) * | 2020-04-30 | 2021-11-02 | 北京海益同展信息科技有限公司 | Gesture recognition method, man-machine interaction method, device, equipment and storage medium |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090327171A1 (en) * | 2008-06-26 | 2009-12-31 | Microsoft Corporation | Recognizing gestures from forearm emg signals |
CN101741952A (en) * | 2009-12-10 | 2010-06-16 | 中国科学技术大学 | Mobile phone interactive system for blind and device thereof |
CN101777250A (en) * | 2010-01-25 | 2010-07-14 | 中国科学技术大学 | General remote control device and method for household appliances |
CN103279734A (en) * | 2013-03-26 | 2013-09-04 | 上海交通大学 | Novel intelligent sign language translation and man-machine interaction system and use method thereof |
CN103777752A (en) * | 2013-11-02 | 2014-05-07 | 上海威璞电子科技有限公司 | Gesture recognition device based on arm muscle current detection and motion sensor |
CN104834907A (en) * | 2015-05-06 | 2015-08-12 | 江苏惠通集团有限责任公司 | Gesture recognition method, apparatus, device and operation method based on gesture recognition |
CN204765638U (en) * | 2015-05-25 | 2015-11-18 | 叶可涵 | Surface electromyography signal data acquisition system |
CN105138133A (en) * | 2015-09-14 | 2015-12-09 | 李玮琛 | Biological signal gesture recognition device and method |
CN109062401A (en) * | 2018-07-11 | 2018-12-21 | 北京理工大学 | A kind of real-time gesture identifying system based on electromyography signal |
-
2019
- 2019-12-02 CN CN201911216438.2A patent/CN111026268A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090327171A1 (en) * | 2008-06-26 | 2009-12-31 | Microsoft Corporation | Recognizing gestures from forearm emg signals |
CN101741952A (en) * | 2009-12-10 | 2010-06-16 | 中国科学技术大学 | Mobile phone interactive system for blind and device thereof |
CN101777250A (en) * | 2010-01-25 | 2010-07-14 | 中国科学技术大学 | General remote control device and method for household appliances |
CN103279734A (en) * | 2013-03-26 | 2013-09-04 | 上海交通大学 | Novel intelligent sign language translation and man-machine interaction system and use method thereof |
CN103777752A (en) * | 2013-11-02 | 2014-05-07 | 上海威璞电子科技有限公司 | Gesture recognition device based on arm muscle current detection and motion sensor |
CN104834907A (en) * | 2015-05-06 | 2015-08-12 | 江苏惠通集团有限责任公司 | Gesture recognition method, apparatus, device and operation method based on gesture recognition |
CN204765638U (en) * | 2015-05-25 | 2015-11-18 | 叶可涵 | Surface electromyography signal data acquisition system |
CN105138133A (en) * | 2015-09-14 | 2015-12-09 | 李玮琛 | Biological signal gesture recognition device and method |
CN109062401A (en) * | 2018-07-11 | 2018-12-21 | 北京理工大学 | A kind of real-time gesture identifying system based on electromyography signal |
Non-Patent Citations (1)
Title |
---|
宋伟 等: "Design of a Flexible Wearable Smart sEMG Recorder Integrated Gradient Boosting Decision Tree Based Hand Gesture Recognition", 《DOI:10.1109/TBCAS.2019.2953998》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113589920A (en) * | 2020-04-30 | 2021-11-02 | 北京海益同展信息科技有限公司 | Gesture recognition method, man-machine interaction method, device, equipment and storage medium |
WO2021219039A1 (en) * | 2020-04-30 | 2021-11-04 | 京东数科海益信息科技有限公司 | Gesture recognition method and apparatus, man-machine interaction method and apparatus, and device and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10061389B2 (en) | Gesture recognition system and gesture recognition method | |
US10878818B2 (en) | Methods and apparatus for silent speech interface | |
US8738122B2 (en) | Systems and methods for reproducing body motions via networks | |
WO2017152531A1 (en) | Ultrasonic wave-based air gesture recognition method and system | |
CN111103976B (en) | Gesture recognition method and device and electronic equipment | |
CN110610172B (en) | Myoelectric gesture recognition method based on RNN-CNN architecture | |
CN109271901A (en) | A kind of sign Language Recognition Method based on Multi-source Information Fusion | |
CN103279734A (en) | Novel intelligent sign language translation and man-machine interaction system and use method thereof | |
CN116189681B (en) | Intelligent voice interaction system and method | |
CN104970820A (en) | Intelligent auscultation system and detecting method | |
CN104352248A (en) | Heart sound, lung sound and fetal heart sound acquisition and health management system and method | |
CN105138133A (en) | Biological signal gesture recognition device and method | |
CN111142668B (en) | Interaction method based on Wi-Fi fingerprint positioning and activity gesture joint recognition | |
Yu et al. | A novel hand gesture recognition method based on 2-channel sEMG | |
CN114190953A (en) | Training method and system of electroencephalogram signal noise reduction model for electroencephalogram acquisition equipment | |
CN114384999B (en) | User-independent myoelectric gesture recognition system based on self-adaptive learning | |
CN111026268A (en) | Gesture recognition device and method | |
CN104980599A (en) | Sign language-voice call method and sign language-voice call system | |
CN111913575B (en) | Method for recognizing hand-language words | |
CN110163142B (en) | Real-time gesture recognition method and system | |
Mendes et al. | Subvocal speech recognition based on EMG signal using independent component analysis and neural network MLP | |
CN210697629U (en) | Signal acquisition device, mobile terminal and signal analysis system | |
CN205083477U (en) | Real time control device based on body surface flesh | |
CN113705339B (en) | Cross-user human behavior recognition method based on antagonism domain adaptation strategy | |
CN111783669B (en) | Surface electromyogram signal classification and identification method for individual user |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200417 |