CN112101298A - Gesture recognition system and method based on muscle electrical impedance signals - Google Patents
Gesture recognition system and method based on muscle electrical impedance signals Download PDFInfo
- Publication number
- CN112101298A CN112101298A CN202011103545.7A CN202011103545A CN112101298A CN 112101298 A CN112101298 A CN 112101298A CN 202011103545 A CN202011103545 A CN 202011103545A CN 112101298 A CN112101298 A CN 112101298A
- Authority
- CN
- China
- Prior art keywords
- signal
- module
- electrical impedance
- upper computer
- gesture recognition
- 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
- 210000003205 muscle Anatomy 0.000 title claims abstract description 17
- 238000000034 method Methods 0.000 title claims description 18
- 238000001514 detection method Methods 0.000 claims abstract description 18
- 238000004891 communication Methods 0.000 claims abstract description 15
- 238000012545 processing Methods 0.000 claims abstract description 9
- 230000005540 biological transmission Effects 0.000 claims abstract description 4
- 238000010801 machine learning Methods 0.000 claims description 8
- 210000000245 forearm Anatomy 0.000 claims description 7
- 230000003387 muscular Effects 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000013145 classification model Methods 0.000 claims description 4
- 230000000087 stabilizing effect Effects 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 claims description 3
- 239000013598 vector Substances 0.000 claims description 3
- 238000012706 support-vector machine Methods 0.000 claims description 2
- 238000002759 z-score normalization Methods 0.000 claims 1
- 230000033001 locomotion Effects 0.000 abstract description 8
- 230000035945 sensitivity Effects 0.000 abstract description 5
- 230000004118 muscle contraction Effects 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000003750 conditioning effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 210000003414 extremity Anatomy 0.000 description 1
- 230000007274 generation of a signal involved in cell-cell signaling Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The invention relates to a gesture recognition system based on a muscle electrical impedance signal, which comprises: the device comprises a signal acquisition unit and an upper computer data processing unit; the signal acquisition unit comprises a main control module, a signal driving module, a signal detection module, an AD acquisition module and a wireless communication module; the master control module is respectively connected with the signal driving module, the AD acquisition module and the wireless communication module; the signal detection module is respectively connected with the electrode and the AD acquisition module; the wireless communication module is connected with the upper computer through wireless transmission and transmits the acquired signals to the upper computer for processing. The invention can directly reflect the essential state of the muscle, has better sensitivity to low-speed movement, and has high sensitivity to muscle contraction, good robustness, large signal amplitude, controllable frequency and simple pretreatment.
Description
Technical Field
The invention relates to the field of machine vision, in particular to a gesture recognition system based on a muscle electrical impedance signal.
Background
Machine vision carries out motion recognition by capturing human motion images, and is the most common gesture recognition method, but the problems of strong privacy invasiveness, high susceptibility to the influence of illumination and the like, limited observation range and high susceptibility to shielding limit the application of machine vision in the field of gesture recognition. With the development of sensor technology, some wearable gesture recognition device designs based on sensor technology have been gradually developed in recent years, mainly including motion sensors (such as gyroscopes, accelerometers, etc.) and surface electromyographic signal sensors, etc. Accelerometers, gyroscopes, and the like, are responsive to motion information of the limb, which is less sensitive to low speed motion. The surface electromyogram signals reflect essential information of muscles, but belong to very weak electrical signals, the acquisition is easily interfered by the outside, the subsequent processing process is complex, and the difficulty is increased for the software and hardware design of the wearable equipment.
Disclosure of Invention
In view of the above, the present invention provides a gesture recognition system and method based on muscle electrical impedance signals, which can directly reflect the intrinsic state of muscle and have better sensitivity to low-speed motion.
In order to achieve the purpose, the invention adopts the following technical scheme:
a gesture recognition system based on muscular electrical impedance signals, comprising: the device comprises a signal acquisition unit and an upper computer data processing unit; the signal acquisition unit comprises a main control module, a signal driving module, a signal detection module, an AD acquisition module and a wireless communication module; the master control module is respectively connected with the signal driving module, the AD acquisition module and the wireless communication module; the signal detection module is respectively connected with the signal driving module and the AD acquisition module; the wireless communication module is connected with the upper computer through wireless transmission and transmits the acquired signals to the upper computer for processing.
Furthermore, the signal driving module comprises a human body surface measuring electrode, a signal source circuit, an amplifier circuit and a radio-frequency follower circuit which are connected in sequence.
Furthermore, the signal detection module comprises a filter circuit, a voltage stabilizing circuit, a phase and polarity identification circuit and an amplitude and phase detection circuit which are connected in sequence.
Further, the wireless communication module adopts Bluetooth, Zig-Bee or Wi-Fi.
A gesture recognition method based on a muscle electrical impedance signal comprises the following steps:
step S1, collecting human body forearm EIM signals through a signal collecting unit and transmitting the human body forearm EIM signals to an upper computer;
step S2, the upper computer collects VMAGAnd VPHSV is calculated according to the relation between signal attenuation and phase differenceiAnd VoThe amplitude ratio and the phase difference;
step S3, converting the voltage amplitude ratio and the phase difference into impedance modes | Z | and φ of the muscle by using the reference resistance according to ohm' S law, and carrying out normalization processing;
and step S4, taking the normalized | Z | 'and φ' as two variables, and training a gesture classification model by using a machine learning method to realize final gesture classification.
Further, the normalization process is normalized by Z-score.
Further, the machine learning method adopts a neural network or a support vector machine.
Further, the extreme learning machine classifier of the machine learning method comprises an input layer, a hidden layer and an output layer, and the specific construction process is as follows:
randomly generating a connection weight between the input layer and the hidden layer;
calculating an output matrix H of the hidden layer, and mapping input data to a result of a node of the hidden layer;
minimization of error function L = min | | Hβ-T | |, where T is the target output of the network;
whereinβThe weight vectors for the hidden layer and the output layer,
after introducing the regularization term, the calculation formula becomesβ= (HTH +1/C) -1HTH, resulting in the final neural network model.
Compared with the prior art, the invention has the following beneficial effects:
the invention can directly reflect the essential state of the muscle, has better sensitivity to low-speed movement, and has high sensitivity to muscle contraction, good robustness, large signal amplitude, controllable frequency and simple pretreatment.
Drawings
FIG. 1 is a graph of the values of | Z | and φ at different measurement frequencies (10 kHz-1 MHz) for six different gestures and six different gestures in accordance with an embodiment of the present invention;
FIG. 2 is a system block diagram of the apparatus of the present invention;
FIG. 3 is a schematic diagram of a signal driver module in accordance with an embodiment of the present invention;
FIG. 4 is a schematic diagram of a signal detection module according to an embodiment of the present invention
FIG. 5 is a diagram of an ELM neural network according to an embodiment of the present invention.
Detailed Description
The invention is further explained below with reference to the drawings and the embodiments.
Referring to fig. 2, the present invention provides a gesture recognition system based on a muscular electrical impedance signal, including: the device comprises a signal acquisition unit and an upper computer; the signal acquisition unit comprises a main control module, a signal driving module, a signal detection module, an AD acquisition module and a wireless communication module; the master control module is respectively connected with the signal driving module, the AD acquisition module and the wireless communication module; the signal detection module is respectively connected with the human body surface measuring electrode and the AD acquisition module; the wireless communication module is connected with the upper computer through wireless transmission and transmits the acquired signals to the upper computer for processing.
In this embodiment, the signal driving module includes an electrode, a signal source circuit, an amplifier circuit, and a emitter follower circuit, which are connected in sequence. The electrodes are arranged according to specific scenes; STM32F103RCT6 is selected as the control unit for the main control module to design peripheral circuit according to pin function.
Referring to fig. 3, in the present embodiment, the signal driving module includes a signal source, an amplifier, and a radio frequency follower circuit, the signal source employs an AD9833, a typical output voltage thereof is 600mV, a sinusoidal voltage output at a frequency of 10kHz to 200kHz does not vary with the frequency, and a function of signal generation can be realized without an external component. The AD8011 is a low-power, wide-band, low-distortion, high-speed operational amplifier that amplifies the signal emitted by the AD9833 into a 1V sinusoidal signal. The emitter follower circuit has a voltage stabilizing function, and ensures that an excitation signal injected into a human body is safe and reliable.
Referring to fig. 4, in this embodiment, the signal detection module is composed of a signal conditioning part and a magnitude-phase detection part, and the signal detection end is composed of a passive filter circuit composed of an RC and an AD8646 voltage stabilizing circuit, so as to suppress interference signals well and ensure the stability of signals under the full frequency. The amplitude and phase detection part takes AD8302 as a core, the AD8302 can measure the gain/loss and the phase difference of two paths of input signals, the measurement frequency is high, the dynamic gain range is-30 dB to +30dB, the phase difference is-180 degrees to +180 degrees, and the standard of human body impedance parameter measurement is met. The output of the AD8302 is two paths of direct-current voltage signals, and the attenuation and the phase difference of the two paths of input signals can be calculated through a simple expression. Taking a voltage signal Vi and a response terminal voltage Vo at two sides of a reference resistor as input, and measuring and outputting a direct-current voltage signal VMAGAnd VPHS. The ADC module converts an analog signal into a digital signal by using the ADC function of the main control chip STM32F103RCT 6. The wireless communication module is used for transmitting signals to the upper computer, and methods such as Bluetooth, Zig-Bee and Wi-Fi can be used. And transmitting the two paths of voltage signals after AD conversion to an upper computer.
In this embodiment, a gesture recognition method based on a muscle electrical impedance signal is further provided, which includes the following steps:
step S1, collecting human body forearm EIM signals through a signal collecting unit and transmitting the human body forearm EIM signals to an upper computer;
step S2, the upper computer collects VMAGAnd VPHSV is calculated according to the relation between signal attenuation and phase differenceiAnd VoThe amplitude ratio and the phase difference;
step S3, converting the voltage amplitude ratio and the phase difference into impedance modes | Z | and φ of the muscle by using the reference resistance according to ohm' S law, and carrying out normalization processing;
and step S4, taking the normalized | Z | 'and φ' as two variables, and training a gesture classification model by using a machine learning method to realize final gesture classification.
Preferably, the normalization process is normalized using Z-score.
Preferably, in this embodiment, the extreme learning machine classifier based on the machine learning method includes an input layer, a hidden layer, and an output layer, the number of neurons in the input layer is 2, and normalized | Z | 'and Φ' are used as two input ends. The output layer is set according to the number of gestures, and referring to fig. 1, in this embodiment, 6 kinds of gestures are taken as an example (6 < 23), so the number of neurons in the output layer is set to 3.
The specific construction process is as follows:
randomly generating a connection weight between the input layer and the hidden layer;
calculating an output matrix H of the hidden layer, and mapping input data to a result of a node of the hidden layer;
minimization of error function L = min | | Hβ-T | |, where T is the target output of the network;
whereinβThe weight vectors for the hidden layer and the output layer,
after introducing the regularization term, the calculation formula becomesβ= (HTH +1/C) -1HTH, resulting in the final neural network model.
Example 1:
in this embodiment, during specific work, the device is placed at the correct position of the forearm, the master control module is adjusted, an alternating current signal is injected through the constant voltage source module, the attenuation value and the phase difference between an input voltage signal and a voltage signal at two ends of the reference resistor are measured by the amplitude-phase detection circuit, and data are transmitted to the upper computer through the bluetooth module after the AD module samples the voltage signal. And (4) carrying out data preprocessing and model training on the upper computer. A series of EIM signals of different gestures are collected to be used as a training set, an impedance model and a phase are normalized to be used as two inputs of a network, and a classification model is built in an upper computer for training. During real-time identification, impedance information of different gestures is collected in real time, the gestures are classified by using a trained classifier model, and results are displayed on an upper computer.
The above description is only a preferred embodiment of the present invention, and all equivalent changes and modifications made in accordance with the claims of the present invention should be covered by the present invention.
Claims (8)
1. A gesture recognition system based on a muscular electrical impedance signal, comprising: the device comprises a signal acquisition unit and an upper computer; the signal acquisition unit comprises a main control module, a signal driving module, a signal detection module, an AD acquisition module and a wireless communication module; the master control module is respectively connected with the signal driving module, the AD acquisition module and the wireless communication module; the signal detection module is respectively connected with the signal driving module and the AD acquisition module; the wireless communication module is connected with the upper computer through wireless transmission, transmits the acquired signals to the upper computer, and the upper computer performs gesture recognition according to the acquired signals.
2. The muscular electrical impedance signal-based gesture recognition system according to claim 1, wherein the signal driving module comprises a human body surface measuring electrode, a signal source circuit, an amplifier circuit and a radio-level follower circuit which are connected in sequence.
3. The muscular electrical impedance signal-based gesture recognition system according to claim 1, wherein the signal detection module comprises a filter circuit, a voltage stabilizing circuit, a phase and polarity discrimination circuit and an amplitude and phase detection circuit which are connected in sequence.
4. A muscular electrical impedance signal based gesture recognition system according to claim 1, wherein the wireless communication module employs bluetooth, Zig-zag or Wi-Fi.
5. A gesture recognition method based on muscle electrical impedance signals is characterized by comprising the following steps:
step S1, collecting human body forearm EIM signals through a signal collecting unit and transmitting the human body forearm EIM signals to an upper computer;
step S2, the upper computer collects VMAGAnd VPHSV is calculated according to the relation between signal attenuation and phase differenceiAnd VoThe amplitude ratio and the phase difference;
step S3, converting the voltage amplitude ratio and the phase difference into impedance modes | Z | and φ of the muscle by using the reference resistance according to ohm' S law, and carrying out normalization processing;
and step S4, taking the normalized | Z | 'and φ' as two variables, and training a gesture classification model by using a machine learning method to realize final gesture classification.
6. The method for identifying a muscle electrical impedance signal-based gesture according to claim 5, wherein the normalization process uses Z-score normalization.
7. The method of identifying a muscle electrical impedance signal-based gesture according to claim 5, wherein the machine learning method employs a neural network or a support vector machine.
8. The method for recognizing the gesture based on the muscular electrical impedance signal according to claim 5, wherein the machine learning method extreme learning machine classifier comprises an input layer, a hidden layer and an output layer, and the specific construction process is as follows:
randomly generating a connection weight between the input layer and the hidden layer;
calculating an output matrix H of the hidden layer, and mapping input data to a result of a node of the hidden layer;
minimum error functionNumber L = min | | Hβ-T | |, where T is the target output of the network;
whereinβThe weight vectors for the hidden layer and the output layer,
after introducing the regularization term, the calculation formula becomesβ= (HTH +1/C) -1HTH, resulting in the final neural network model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011103545.7A CN112101298A (en) | 2020-10-15 | 2020-10-15 | Gesture recognition system and method based on muscle electrical impedance signals |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011103545.7A CN112101298A (en) | 2020-10-15 | 2020-10-15 | Gesture recognition system and method based on muscle electrical impedance signals |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112101298A true CN112101298A (en) | 2020-12-18 |
Family
ID=73784273
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011103545.7A Pending CN112101298A (en) | 2020-10-15 | 2020-10-15 | Gesture recognition system and method based on muscle electrical impedance signals |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112101298A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112966662A (en) * | 2021-03-31 | 2021-06-15 | 安徽大学 | Short-range capacitive dynamic gesture recognition system and method |
| CN114879841A (en) * | 2022-05-09 | 2022-08-09 | 南昌航空大学 | Gesture recognition system and measurement method based on D-type plastic optical fiber |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101667346A (en) * | 2009-09-18 | 2010-03-10 | 国家康复辅具研究中心 | Rehabilitation training system of amputation upper limb based on virtual reality |
| CN106293057A (en) * | 2016-07-20 | 2017-01-04 | 西安中科比奇创新科技有限责任公司 | Gesture identification method based on BP neutral net |
| CN106388819A (en) * | 2016-10-28 | 2017-02-15 | 许昌学院 | Human upper limb muscle state monitoring system based on surface electromyogram signals and judging method thereof |
| CN108992066A (en) * | 2018-08-15 | 2018-12-14 | 东北大学 | Portable lower limb behavior pattern real-time identifying system and method based on electromyography signal |
| CN110794961A (en) * | 2019-10-14 | 2020-02-14 | 无锡益碧医疗科技有限公司 | Wearable gesture analysis system |
| CN212647500U (en) * | 2020-10-15 | 2021-03-02 | 福州大学 | Gesture recognition system based on muscle electrical impedance signal |
-
2020
- 2020-10-15 CN CN202011103545.7A patent/CN112101298A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101667346A (en) * | 2009-09-18 | 2010-03-10 | 国家康复辅具研究中心 | Rehabilitation training system of amputation upper limb based on virtual reality |
| CN106293057A (en) * | 2016-07-20 | 2017-01-04 | 西安中科比奇创新科技有限责任公司 | Gesture identification method based on BP neutral net |
| CN106388819A (en) * | 2016-10-28 | 2017-02-15 | 许昌学院 | Human upper limb muscle state monitoring system based on surface electromyogram signals and judging method thereof |
| CN108992066A (en) * | 2018-08-15 | 2018-12-14 | 东北大学 | Portable lower limb behavior pattern real-time identifying system and method based on electromyography signal |
| CN110794961A (en) * | 2019-10-14 | 2020-02-14 | 无锡益碧医疗科技有限公司 | Wearable gesture analysis system |
| CN212647500U (en) * | 2020-10-15 | 2021-03-02 | 福州大学 | Gesture recognition system based on muscle electrical impedance signal |
Non-Patent Citations (1)
| Title |
|---|
| 田岚;姜乃夫;李光林;: "基于微处理器的多功能肌电假肢控制系统", 集成技术, no. 04, 15 July 2013 (2013-07-15) * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112966662A (en) * | 2021-03-31 | 2021-06-15 | 安徽大学 | Short-range capacitive dynamic gesture recognition system and method |
| CN114879841A (en) * | 2022-05-09 | 2022-08-09 | 南昌航空大学 | Gesture recognition system and measurement method based on D-type plastic optical fiber |
| CN114879841B (en) * | 2022-05-09 | 2023-04-07 | 南昌航空大学 | Gesture recognition system and measurement method based on D-type plastic optical fiber |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111553307B (en) | Gesture recognition system fusing bioelectrical impedance information and myoelectric information | |
| Cecotti et al. | Convolutional neural network with embedded Fourier transform for EEG classification | |
| CN105210088B (en) | Using the signal processing of partial discharge method and device of neural network | |
| CN105769173B (en) | A kind of cardioelectric monitor system with electrocardiosignal noise removal function | |
| Zardoshti-Kermani et al. | EMG feature evaluation for movement control of upper extremity prostheses | |
| CN103142225B (en) | Electroencephalograph for determining contact status between electrode and scalp and method for determining same | |
| WO2018214529A1 (en) | Electromyographic signal acquisition method and apparatus | |
| CN112101298A (en) | Gesture recognition system and method based on muscle electrical impedance signals | |
| CN101828919A (en) | Surface electromyographic electrode with amplifying function | |
| WO2020118797A1 (en) | Prosthesis control method, apparatus, system and device, and storage medium | |
| CN107822629A (en) | The detection method of extremity surface myoelectricity axle | |
| CN212647500U (en) | Gesture recognition system based on muscle electrical impedance signal | |
| Zhang et al. | Real-time surface EMG pattern recognition for hand gestures based on support vector machine | |
| CN112732092A (en) | Surface electromyogram signal identification method based on double-view multi-scale convolution neural network | |
| CN110710984B (en) | Ankle moment prediction method of recursion cerebellum model based on surface electromyogram signal | |
| CN105796091B (en) | A kind of intelligent terminal for removing electrocardiosignal vehicle movement noise | |
| CN203829261U (en) | Data collecting and processing system for upper limb rehabilitation robots | |
| Wen et al. | Multimodal sensor motion intention recognition based on three-dimensional convolutional neural network algorithm | |
| CN112405539B (en) | Robot natural control method based on electromyographic signals and electroencephalogram error potentials | |
| CN112057090B (en) | Emotion judgment wearable device and method based on body surface extremely-low-frequency potential difference characteristics | |
| CN111242100B (en) | Action identification method based on GST and VL-MOBPNN | |
| CN210697629U (en) | Signal acquisition device, mobile terminal and signal analysis system | |
| Fu et al. | Finger gesture recognition using sensing and classification of surface electromyography signals with high-precision wireless surface electromyography sensors | |
| CN110794961A (en) | Wearable gesture analysis system | |
| Yang et al. | Design of the human surface electromyogra signal acquisition system and signal analysis |
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 |