CN109171124A - A kind of muscle signals wireless collection bracelet for Sign Language Recognition - Google Patents
A kind of muscle signals wireless collection bracelet for Sign Language Recognition Download PDFInfo
- Publication number
- CN109171124A CN109171124A CN201811056943.0A CN201811056943A CN109171124A CN 109171124 A CN109171124 A CN 109171124A CN 201811056943 A CN201811056943 A CN 201811056943A CN 109171124 A CN109171124 A CN 109171124A
- Authority
- CN
- China
- Prior art keywords
- signal
- bracelet
- sign language
- muscle signals
- wireless collection
- 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
Classifications
-
- A—HUMAN NECESSITIES
- A44—HABERDASHERY; JEWELLERY
- A44C—PERSONAL ADORNMENTS, e.g. JEWELLERY; COINS
- A44C5/00—Bracelets; Wrist-watch straps; Fastenings for bracelets or wrist-watch straps
- A44C5/0007—Bracelets specially adapted for other functions or with means for attaching other articles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/28—Recognition of hand or arm movements, e.g. recognition of deaf sign language
Abstract
The present invention relates to a kind of muscle signals wireless collection bracelets for Sign Language Recognition, which includes: plastic casing: the signal acquisition circuit plate and provide the rechargeable lithium battery of electric energy for signal acquisition circuit plate that inside setting muscle signals acquire and wifi data are transmitted；Elastic wristband: hinged with plastic casing, its inside is respectively equipped with four acceleration transducers connecting respectively by FPC soft arranging wire beam with signal acquisition circuit plate, and the acceleration transducer is close to setting with general extension flesh, musculus flexor carpi radialis, musculus flexor carpi ulnaris and musculus extensor carpi radialis longus is referred to respectively.Compared with prior art, the present invention has many advantages, such as that muscle signals wireless collection, sensor position are adjustable, easy to wear, and easy to carry, portable high, gesture sign language identifies in real time, reduces volume.
Description
Technical field
The present invention relates to Sign Language Recognition technical fields, wirelessly adopt more particularly, to a kind of muscle signals for Sign Language Recognition
Collect bracelet.
Background technique
As most common human action, gesture motion is using finger, palm, wrist, the position of arm, shape etc. come table
It is intended to or transmits order up to certain behavior.It is limited compared to the schema category of human body head movement, facial action to be not easy to capture
Execute the disadvantages of harder with body action, gesture motion may be implemented by means of position the most flexible in people's limbs
The combination of actions of large amount of complex carrys out the meaning of expressed in abundance, is applied in numerous areas, comprising: passes through when ordinary people's social activity
Simple gesture motion, which is expressed, the meanings such as greets, thanks, apologizing；Referee uses gesture on sports ground clearly accurately conveys ratio
The judgement of match；Sound sight people's structuring, standardized gesture motion carry out the communication exchange of mutual thought；Traffic-police's benefit
The traffic of congestion is dredged with specific gesture motion；Physiotherapist is by allowing patient to execute a set of gesture motion pre-established
Come motor behavior and the disease degree etc. for analyzing patient.Gesture motion all plays extremely important in the every aspect that people live
Effect will greatly convenient people if can find a kind of technology realizes the meaning for automatically understanding and conveying gesture motion
Life.
Currently, comparative maturity is to acquire images of gestures using camera in gesture Sign Language Recognition field, know with image
Other technology realizes the identification of gesture motion, can reach certain standard to the identification of gesture sign language under the good environment of illumination condition
True rate.However under the conditions of dark or direct sunlight, image is collected by camera and is limited, relatively good identification is extremely difficult to
Effect, while the position installation etc. of camera limits this method in the practical application value in Sign Language Recognition field.
It is different from a kind of gesture sign Language Recognition, some documents are referred to using motion sensor and myoelectric sensor
Combination, acquire hand motion information, pass through and collected shape information is realized to the action recognition of gesture sign language.It should
Kind of method has certain advantage in terms of use environment relative to former approach in the adaptability of environment-identification, but flesh
The acquisition of electric signal is influenced vulnerable to some states of human body, for example the perspiration of human body can be such that collected dynamoelectric signal is distorted,
Reduce the accuracy rate of identification.
In flesh sound acquisition equipment, the prior art mostly uses the mode of flesh sound sensor and capture card to realize, this kind of method
Although there is biggish progress in discrimination, influenced by weight of equipment volume, is difficult to be applied directly to practical sign language knowledge
In not, need to improve on the basis of original.
Summary of the invention
It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art and provide one kind to be used for Sign Language Recognition
Muscle signals wireless collection bracelet.
The purpose of the present invention can be achieved through the following technical solutions:
A kind of muscle signals wireless collection bracelet for Sign Language Recognition, the bracelet include:
Plastic casing: the signal acquisition circuit plate and be signal that inside setting muscle signals acquire and wifi data are transmitted
The rechargeable lithium battery of collecting circuit board offer electric energy；
Elastic wristband: hinged with plastic casing, inside is respectively equipped with four and is adopted respectively by FPC soft arranging wire beam with signal
Collector plate connection acceleration transducer, the acceleration transducer respectively with refer to general extension flesh, musculus flexor carpi radialis, ulnar side
Wrist musculus flexor and musculus extensor carpi radialis longus are close to setting.
The signal acquisition circuit plate is mono- for an intelligent acquisition chip based on STM32F405RG design, including STM32
Piece machine minimum system and WIFI chip module, power management module and the SPI being connect respectively with STM32 single-chip minimum system
Communication module, the SPI communication module and acceleration transducer wirelessly communicate, the WIFI chip module with it is external upper
Machine communication.
The acceleration transducer is arranged in elastic wristband by drawing button, and by drawing button adjusting position.
The acceleration transducer uses ADXL355 chip, passes sequentially through ADC analog-to-digital conversion and digital Peripheral Interface
It is communicated with SPI communication module.
The sample frequency of the acceleration transducer is 2000Hz.
The application method of the bracelet is as follows:
1) bracelet is dressed: bracelet being worn on arm forearm, is close to four acceleration transducers using elastic wristband
On arm, by draw button four acceleration transducers of slidable adjustment a position, make sensor be tightly attached to respectively refer to general extension flesh,
On four pieces of musculus flexor carpi radialis, musculus flexor carpi ulnaris and musculus extensor carpi radialis longus muscle；
2) equipment connects: the switch on opening signal collecting circuit board, bracelet is started to work, by the network connection of computer
In the wifi network issued to bracelet, the communication connection of host computer and bracelet is completed；
3) signal transmission and processing: adjustment gesture motion, four acceleration transducers acquire MMG signal respectively, and pass through
Signal acquisition circuit plate transmits a signal to host computer, and host computer is by realizing the knowledge acted to gesture sign language to signal analysis
Not.
The step 3) specifically includes the following steps:
31) Signal Pretreatment: by the original MMG signal of the collected acceleration transducer in each channel removal high-frequency noise and
Low frequency baseline, and be normalized；
32) movement segmentation: to treated, MMG signal is single using the non-uniformly distributed load method extraction based on twice-enveloping line
The segmentation waveform signal of movement；
33) feature extraction: carrying out four layers of wavelet packet decomposition using Coif4 small echo to the segmentation waveform signal in each channel,
Extract wavelet package transforms coefficient energy feature；
34) dimension-reduction treatment dimension-reduction treatment: is carried out to wavelet package transforms coefficient energy feature using Principal Component Analysis；
35) using the feature after dimension-reduction treatment as the input of training SVM model, it is dynamic for sign language svm classifier: to export result
Make, and trained SVM model is subjected to sign language action recognition.
In the step 33), the expression formula of wavelet package transforms coefficient energy feature F is extracted are as follows:
F={ log10(Ei), i=1 ..., 2L}
Wherein,For j-th of coefficient in the signal subspace band of i-th layer of WAVELET PACKET DECOMPOSITION, NiForThe wavelet packet of signal band
Transformation coefficient the average energy value, L are Decomposition order, EiFor intermediate variable.
In the step 35), the kernel function of SVM model used is radial basis function K (z, xc), it may be assumed that
Wherein, z is the point in former lower dimensional space, and xc is function center, and σ is width parameter.
Compared with prior art, the invention has the following advantages that
1, muscle signals wireless collection: MMG signal is transmitted by the way of wifi transmission, power supply system is independent, for not
Same acquisition environmental suitability is strong.
2, sensor position is adjustable: acceleration transducer is fixed on wrist strap by drawing button, it can be achieved that in the circumferential
Sliding, to adapt to the muscle feature of different arm size crowds, acquires the forearm muscle signals of different crowd.
3, easy to wear, easy to carry: using the design of bracelet, bracelet wrist strap is flexible resilient belt, it is convenient to be worn
It is worn in wrist, and total quality is less than 300g, it is lighter, facilitate wearing.
4, portable high: the present invention can be obtained from bracelet using TCP/IP communication modes with different equipment
Muscle signals information realizes the acquisition process to muscle signals.
5, gesture sign language identifies in real time: the MMG signal acquisition frequency of this equipment is real-time up to 2000HZ, the partial data
It obtains, can handle in real time the partial data and obtain the direct result of gesture identification by computer.
6, reduce volume: FPC soft arranging wire beam is that FCCL (flexible copper clad foil) processing is obtained line with the mode of chemical etching
Road walks that type difference single side is two-sided and the flexible circuit board of multilayered structure, adapts to the different field of the forearm wrist sizes of different people
It closes, while greatly reducing the volume of electronic product.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of muscle signals wireless collection bracelet of the present invention.
Fig. 2 is muscle signals wireless collection bracelet diagrammatic cross-section of the present invention.
Description of symbols in figure:
1, plastic casing, 2, signal acquisition circuit plate, 3, lithium battery, 4, FPC soft arranging wire harness, the 5, first acceleration sensing
Device, 6, elastic wristband, 7, plastics cap, the 8, second acceleration transducer, 9, third acceleration transducer, the 10, the 4th acceleration
Sensor.
Specific embodiment
The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.
Embodiment
As shown in Figs. 1-2, the present invention provides a kind of muscle signals wireless collection bracelet for Sign Language Recognition, the bracelet knot
It is mainly made of plastic casing 1 and elastic wristband 6 on structure；Elastic wristband 6 is fixed in articulated manner on plastic casing 1,
Make 4 high-precision acceleration transducers 5,8,9,10 on wrist strap tight by the scalability of elastic wristband 6 in donning process
It is attached on forearm muscle, by drawing button design, can be according to different people the case where, by four 5,8,9,10 points of acceleration transducer
It is not tightly attached to and refers to general extension flesh (EDC), musculus flexor carpi radialis (FCR), musculus flexor carpi ulnaris (FCU) and musculus extensor carpi radialis longus (ECR) four
On block muscle.Include the intelligent signal collection for realizing muscle signals acquisition function and wifi data-transformation facility in shell 1
Circuit board 2 and the rechargeable lithium battery 3 of electric energy is provided for it, collecting circuit board 2 passes through FPC soft arranging wire beam 4 and 4 high-precisions
Acceleration transducer 5,8,9,10 is connected, and realizes the acquisition of acceleration information.
Intelligent signal collection circuit board 2 is a intelligent acquisition chip based on STM32F405RG design.Circuit includes
STM32 single-chip minimum system, WIFI chip module and SPI communication module etc., it may be implemented the acquisition to MMG signal,
The foundation of WIFI network and the wireless transmission of signal.Sample frequency reaches as high as 2000HZ.The size of the collection plate is
40x25x5mm。
Acceleration transducer 5,8,9,10 is the acceleration transducer designed based on ADXL355 chip, and supply voltage is
2.25V-3.6V can pass through digital Peripheral Interface (SPI)/I2C interface interacts, and ADC analog-to-digital conversion is 20, sensitivity
For 256000LSB/g, power consumption is 200uA under measurement pattern.The sensor it is small in size, low in energy consumption, precision is high, meets flesh message
The acquisition requirement of number (MMG), the partial circuit plate size are 15x15x4mm.
FPC soft arranging wire beam 4 is that FCCL (flexible copper clad foil) processing is obtained route with the mode of chemical etching to walk type difference
Single side is two-sided and the flexible circuit board of multilayered structure, adapts to the different occasion of the forearm wrist sizes of different people, while significantly
Reduce the volume of electronic product.
Muscle signals wireless collection bracelet of the present invention is the acquisition equipment applied to sign language or gesture identification research, can also be answered
For other pairs of gesture identification required experiments or application.Its working principles are as follows:
The MMG signal that No. four acceleration transducers 5,8,9,10 acquire every piece of muscle passes to letter with the communication modes of SPI
Number collecting circuit board 2, signal acquisition circuit plate 2, which passes through the wifi network independently established and is sent to this partial information, is connected to this
On host computer (computer) under wifi hotspot, host computer (computer) receives this part signal and carries out simultaneously to the part signal
Identifying processing, to realize the identification of gesture sign language.The electric energy of bracelet all devices is by the lithium being fixed on plastic shell 2,7
Battery 3 provides.Here is the operating procedure using the bracelet:
(1) bracelet is dressed.Bracelet is worn on arm forearm, so that four sensors is tightly attached to hand using elastic wristband 6
On arm, slide the position of four sensors, make sensor be tightly attached to respectively refer to general extension flesh (EDC), musculus flexor carpi radialis (FCR),
On four pieces of muscle of musculus flexor carpi ulnaris (FCU) and musculus extensor carpi radialis longus (ECR).
(2) equipment connects.Switch on opening signal collecting circuit board 2, makes bracelet start to work, by the network of computer
It is connected in the wifi network of bracelet sending, opens host computer and connect equipment.
(3) signal transmission and processing.Gesture motion is adjusted, No. four acceleration transducers 5,8,9,10 collect MMG signal, lead to
Intelligent chip circuit board is crossed, transmits a signal to computer, computer is realized and acted to gesture sign language by analyzing signal
Identification.
Detailed process is as follows for identification:
Signal Pretreatment:
The process of Signal Pretreatment includes to denoise and normalize two processes.Collected original signal is by 2-80HZ's
Kaiser window filter removes high-frequency noise and low frequency baseline.The processing that signal is normalized according to formula (1).
Wherein x is the collected acceleration value in each channel,For same channel acceleration average value, S is the channel
Standard deviation.
Movement segmentation:
Movement segmentation is the basis for obtaining identification signal segment, and this method is using a kind of based on the non-of twice-enveloping line
Even partition method, before and after envelope two o'clock respectively as movement starting point and end point, to reach one action extraction
Purpose.
Feature extraction:
After pre-processing and acting segmentation, the segmentation waveform of each movement of four-way is obtained, then from splitting signal
Extract the features such as wavelet package transforms coefficient energy feature.The extraction of wavelet packet coefficient transform characteristics following formula (2), (3) indicate
F={ log10(Ei), i=1 ..., 2L} (2)
Wherein F represents feature, and L is Decomposition order,It is j-th of coefficient in the signal subspace band of i-th layer of WAVELET PACKET DECOMPOSITION,
NiIt is CiWavelet package transforms coefficient the average energy value of signal band.
MMG signal carries out four layers of wavelet packet decomposition, therefore the wavelet package transforms coefficient energy in each channel using Coif4 small echo
Measure feature is 16 dimensions.Each sensor has three direction signals, and there are four channels altogether, so the energy feature of wavelet package transforms coefficient
There is 16x12=192 dimension.
Dimension-reduction treatment:
Since the number of features obtained in features described above is handled is tieed up for 192, there are redundancies between feature, it is therefore desirable to right
It carries out Feature Dimension Reduction processing, and dimension reduction method of the present invention is Principal Component Analysis (PCA).After calculating normalization
The characteristic value of the covariance matrix of data and corresponding feature vector are selected wherein maximum by the sequence of characteristic value from big to small
K feature vector respectively as Column vector groups at eigenvectors matrix, original data set is then projected into these feature vectors
On, the data set of dimensionality reduction can be obtained.
Svm classifier:
After the pretreatment, segmentation, feature extraction, Feature Dimension Reduction of several steps above, finally treated, feature is empty
Between input SVM training pattern, then in the application of practical Sign Language Recognition, by being acquired to real-time movement, according to obtaining
MMG signal dimensionality reduction feature substitute into trained SVM model, so that it may export the sign language information to induction signal.The present invention
Involved in the kernel function that uses of SVM algorithm for radial basis function (RBF), such as formula (4)
Wherein K (z, xc) is radial basis function, and xc is the center of function, and σ is the width parameter of function, and it is radial to play control
The function of distance, z are the points in former lower dimensional space.
Claims (9)
1. a kind of muscle signals wireless collection bracelet for Sign Language Recognition, which is characterized in that the bracelet includes:
Plastic casing (1): the signal acquisition circuit plate (2) and be letter that inside setting muscle signals acquire and wifi data are transmitted
Number collecting circuit board (2) provides the rechargeable lithium battery (3) of electric energy；
Elastic wristband (6): hingedly with plastic casing (1), inside be respectively equipped with four respectively by FPC soft arranging wire beam (4) with
Signal acquisition circuit plate (2) connection acceleration transducer, the acceleration transducer respectively with refer to general extension flesh, carpi radialis
Musculus flexor, musculus flexor carpi ulnaris and musculus extensor carpi radialis longus are close to setting.
2. a kind of muscle signals wireless collection bracelet for Sign Language Recognition according to claim 1, which is characterized in that institute
The signal acquisition circuit plate (2) stated is an intelligent acquisition chip based on STM32F405RG design, including STM32 single-chip microcontroller is most
Mini system and the WIFI chip module connecting respectively with STM32 single-chip minimum system, power management module and SPI communicate mould
Block, the SPI communication module and acceleration transducer wirelessly communicate, and the WIFI chip module and external host computer are logical
Letter.
3. a kind of muscle signals wireless collection bracelet for Sign Language Recognition according to claim 1, which is characterized in that institute
The acceleration transducer stated is arranged on elastic wristband (6) by drawing button, and by drawing button adjusting position.
4. a kind of muscle signals wireless collection bracelet for Sign Language Recognition according to claim 2, which is characterized in that institute
The acceleration transducer stated uses ADXL355 chip, passes sequentially through ADC analog-to-digital conversion and digital Peripheral Interface and SPI is communicated
Module communication.
5. a kind of muscle signals wireless collection bracelet for Sign Language Recognition according to claim 4, which is characterized in that institute
The sample frequency for the acceleration transducer stated is 2000Hz.
6. a kind of muscle signals wireless collection bracelet for Sign Language Recognition according to claim 1-5, special
Sign is that the application method of the bracelet is as follows:
1) bracelet is dressed: bracelet being worn on arm forearm, so that four acceleration transducers is tightly attached to hand using elastic wristband
On arm, by drawing the position of button four acceleration transducers of slidable adjustment, it is tightly attached to sensor respectively and refers to general extension flesh, oar side
On four pieces of wrist musculus flexor, musculus flexor carpi ulnaris and musculus extensor carpi radialis longus muscle；
2) equipment connects: the switch on opening signal collecting circuit board, bracelet is started to work, and the network connection of computer is in one's hands
In the wifi network that ring issues, the communication connection of host computer and bracelet is completed；
3) signal transmission and processing: adjustment gesture motion, four acceleration transducers acquire MMG signal respectively, and pass through signal
Collecting circuit board transmits a signal to host computer, and host computer is by realizing the identification acted to gesture sign language to signal analysis.
7. a kind of muscle signals wireless collection bracelet for Sign Language Recognition according to claim 6, which is characterized in that institute
The step 3) stated specifically includes the following steps:
31) the original MMG signal of the collected acceleration transducer in each channel Signal Pretreatment: is removed into high-frequency noise and low frequency
Baseline, and be normalized；
32) movement segmentation: to treated, MMG signal uses the non-uniformly distributed load method based on twice-enveloping line to extract one action
Segmentation waveform signal；
33) feature extraction: four layers of wavelet packet decomposition are carried out using Coif4 small echo to the segmentation waveform signal in each channel, are extracted
Wavelet package transforms coefficient energy feature；
34) dimension-reduction treatment dimension-reduction treatment: is carried out to wavelet package transforms coefficient energy feature using Principal Component Analysis；
35) svm classifier: using the feature after dimension-reduction treatment as the input of training SVM model, exporting result as sign language movement, and
Trained SVM model is subjected to sign language action recognition.
8. a kind of muscle signals wireless collection bracelet for Sign Language Recognition according to claim 7, which is characterized in that institute
In the step 33) stated, the expression formula of wavelet package transforms coefficient energy feature F is extracted are as follows:
F={ log10(Ei), i=1 ..., 2L}
Wherein,For j-th of coefficient in the signal subspace band of i-th layer of WAVELET PACKET DECOMPOSITION, NiForThe wavelet package transforms of signal band
Coefficient the average energy value, L are Decomposition order, EiFor intermediate variable.
9. a kind of muscle signals wireless collection bracelet for Sign Language Recognition according to claim 7, which is characterized in that institute
In the step 35) stated, the kernel function of SVM model used is radial basis function K (z, xc), it may be assumed that
Wherein, z is the point in former lower dimensional space, and xc is function center, and σ is width parameter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811056943.0A CN109171124A (en) | 2018-09-11 | 2018-09-11 | A kind of muscle signals wireless collection bracelet for Sign Language Recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811056943.0A CN109171124A (en) | 2018-09-11 | 2018-09-11 | A kind of muscle signals wireless collection bracelet for Sign Language Recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109171124A true CN109171124A (en) | 2019-01-11 |
Family
ID=64910365
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811056943.0A Pending CN109171124A (en) | 2018-09-11 | 2018-09-11 | A kind of muscle signals wireless collection bracelet for Sign Language Recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109171124A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109996371A (en) * | 2019-04-24 | 2019-07-09 | 深圳尚一互联技术有限公司 | The method of gesture control light for Intelligent bracelet |
| CN111110268A (en) * | 2019-11-26 | 2020-05-08 | 中国科学院合肥物质科学研究院 | Human body muscle sound signal prediction method based on random vector function connection network technology |
| CN111466945A (en) * | 2020-04-26 | 2020-07-31 | 重庆大学 | Muscle sound signal detection sensing device capable of automatically adjusting contact pressure |
| CN111881697A (en) * | 2020-08-17 | 2020-11-03 | 华东理工大学 | Real-time sign language translation method and system |
| CN114468986A (en) * | 2020-11-13 | 2022-05-13 | 浙江大学台州研究院 | Wearable device based on artificial intelligence multimode epileptic seizure monitoring |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101561709A (en) * | 2009-05-11 | 2009-10-21 | 杨杰 | Method and device for controlling computer |
| CN102013016A (en) * | 2010-11-23 | 2011-04-13 | 华东理工大学 | Muscle sound signal-based hand motion mode identification method for prosthetic hand control |
| CN103294199A (en) * | 2013-06-09 | 2013-09-11 | 华东理工大学 | Silent information identifying system based on facial muscle sound signals |
| CN103654774A (en) * | 2014-01-02 | 2014-03-26 | 北京思睿博创科技有限公司 | Wearable movable bracelet |
| 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 |
| CN104267807A (en) * | 2014-09-16 | 2015-01-07 | 华东理工大学 | Hand action mechanomyography based man-machine interaction method and interaction system |
| CN106744115A (en) * | 2016-12-29 | 2017-05-31 | 苏州感测通信息科技有限公司 | Elevator health intelligent monitoring device |
| CN107346032A (en) * | 2017-06-21 | 2017-11-14 | 湖北省交通规划设计院股份有限公司 | A kind of wireless controlled passes the Tunnel prediction system and method for acceleration transducer |
| CN107744436A (en) * | 2017-10-16 | 2018-03-02 | 华东理工大学 | A kind of wheelchair control method and control system based on the processing of neck muscle signals |
-
2018
- 2018-09-11 CN CN201811056943.0A patent/CN109171124A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101561709A (en) * | 2009-05-11 | 2009-10-21 | 杨杰 | Method and device for controlling computer |
| CN102013016A (en) * | 2010-11-23 | 2011-04-13 | 华东理工大学 | Muscle sound signal-based hand motion mode identification method for prosthetic hand control |
| CN103294199A (en) * | 2013-06-09 | 2013-09-11 | 华东理工大学 | Silent information identifying system based on facial muscle sound signals |
| 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 |
| CN103654774A (en) * | 2014-01-02 | 2014-03-26 | 北京思睿博创科技有限公司 | Wearable movable bracelet |
| CN104267807A (en) * | 2014-09-16 | 2015-01-07 | 华东理工大学 | Hand action mechanomyography based man-machine interaction method and interaction system |
| CN106744115A (en) * | 2016-12-29 | 2017-05-31 | 苏州感测通信息科技有限公司 | Elevator health intelligent monitoring device |
| CN107346032A (en) * | 2017-06-21 | 2017-11-14 | 湖北省交通规划设计院股份有限公司 | A kind of wireless controlled passes the Tunnel prediction system and method for acceleration transducer |
| CN107744436A (en) * | 2017-10-16 | 2018-03-02 | 华东理工大学 | A kind of wheelchair control method and control system based on the processing of neck muscle signals |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109996371A (en) * | 2019-04-24 | 2019-07-09 | 深圳尚一互联技术有限公司 | The method of gesture control light for Intelligent bracelet |
| CN111110268A (en) * | 2019-11-26 | 2020-05-08 | 中国科学院合肥物质科学研究院 | Human body muscle sound signal prediction method based on random vector function connection network technology |
| CN111466945A (en) * | 2020-04-26 | 2020-07-31 | 重庆大学 | Muscle sound signal detection sensing device capable of automatically adjusting contact pressure |
| CN111881697A (en) * | 2020-08-17 | 2020-11-03 | 华东理工大学 | Real-time sign language translation method and system |
| CN114468986A (en) * | 2020-11-13 | 2022-05-13 | 浙江大学台州研究院 | Wearable device based on artificial intelligence multimode epileptic seizure monitoring |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109171124A (en) | A kind of muscle signals wireless collection bracelet for Sign Language Recognition | |
| US10061389B2 (en) | Gesture recognition system and gesture recognition method | |
| Yen et al. | Human daily activity recognition performed using wearable inertial sensors combined with deep learning algorithms | |
| Vamsikrishna et al. | Computer-vision-assisted palm rehabilitation with supervised learning | |
| He et al. | Activity recognition from acceleration data based on discrete consine transform and SVM | |
| CN100594858C (en) | Electric artificial hand combined controlled by brain electricity and muscle electricity and control method | |
| CN108171278B (en) | Motion pattern recognition method and system based on motion training data | |
| CN106383579A (en) | EMG and FSR-based refined gesture recognition system and method | |
| CN102930284A (en) | Surface electromyogram signal pattern recognition method based on empirical mode decomposition and fractal | |
| Zhang et al. | Recognizing hand gestures with pressure-sensor-based motion sensing | |
| CN105786189B (en) | A kind of self contained function recognition methods of finger portion and system that signal is moved based on flesh | |
| Ding et al. | Energy efficient human activity recognition using wearable sensors | |
| Zheng et al. | L-sign: Large-vocabulary sign gestures recognition system | |
| CN113871028A (en) | Interactive rehabilitation system based on myoelectric intelligent wearing | |
| CN110751060B (en) | Portable motion mode real-time identification system based on multi-source signals | |
| Liu et al. | Hand gesture recognition based on deep neural network and sEMG signal | |
| CN110604578A (en) | Human hand and hand motion recognition method based on SEMG | |
| Tao et al. | Research on communication APP for deaf and mute people based on face emotion recognition technology | |
| CN108391049A (en) | Filming control method and relevant device | |
| Zolfaghari et al. | Speed classification of upper limb movements through EEG signal for BCI application | |
| Li et al. | Identification of finger force and motion from forearm surface electromyography | |
| Li et al. | A single smartwatch-based segmentation approach in human activity recognition | |
| CN107028608B (en) | Brain wave signal acquisition device and preprocessing method | |
| CN109965847A (en) | Server and Signal Analysis System | |
| Matsumura et al. | Recognition of wrist EMG signal patterns using neural networks |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190111 |
|
| RJ01 | Rejection of invention patent application after publication |