CN105608432B - A kind of gesture identification method based on instantaneous myoelectricity image - Google Patents
A kind of gesture identification method based on instantaneous myoelectricity image Download PDFInfo
- Publication number
- CN105608432B CN105608432B CN201510973702.2A CN201510973702A CN105608432B CN 105608432 B CN105608432 B CN 105608432B CN 201510973702 A CN201510973702 A CN 201510973702A CN 105608432 B CN105608432 B CN 105608432B
- Authority
- CN
- China
- Prior art keywords
- instantaneous
- image
- electromyography signal
- myoelectricity
- gesture
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of gesture identification methods based on instantaneous electromyography signal.In the training stage, the instantaneous electromyography signal of array electrode acquisition is pre-processed first, and it is arranged into instantaneous myoelectricity image according to electrode position；Then instantaneous myoelectricity image and its corresponding gesture label training image classifier, such as depth convolutional neural networks are used, network model parameter is obtained；In test phase, the instantaneous electromyography signal to be identified of array electrode acquisition is pre-processed first, and it is arranged into instantaneous myoelectricity image according to electrode position；Then trained model parameter is brought into classifier and identifies the corresponding gesture label of instantaneous electromyography signal.The present invention is based on instantaneous myoelectricity image and image classification methods, can rapidly and accurately identify gesture.Document there is no to carry out gesture identification using instantaneous electromyography signal both at home and abroad.
Description
Technical field
The invention belongs to computers, and field is combined with bio signal, and specially use is using depth convolutional neural networks as generation
The Image Classifier of table identifies gesture corresponding to the instantaneous myoelectricity image extracted from instantaneous electromyography signal.
Background technique
With the fast development of the new technologies such as computer vision, touch interaction, perceptual computing, user interface is perceived
(perceptual user interface, PUI) becomes one of research emphasis of field of human-computer interaction.Perceive user interface
It is the user interface that a kind of interacting activity between person to person and people and real world is the high interaction of prototype, multichannel, it
Target be to become human-computer interaction and people unanimously with the interaction of real world to reach intuitive, naturally interact boundary.In order to
Enable a computer to preferably judge and understand that the interaction of the mankind is intended to, " raw, mechanical, electrical integration " is the following man-machine interactive development
One of important trend, i.e., by specific sensing equipment by the cognition of organism or perceptual signal (such as electromyography signal) number
Change, and carry out integrated fusion with the signal of other perception or cognition channel, nature synergistically completes various human-computer interaction tasks.
Previous study carries out the gesture identification based on electromyography signal usually using machine learning method, that is, gives one section of myoelectricity
Signal is classified using trained classifier, obtains gesture label；The classifier is using electromyography signal gathered in advance
What training obtained.Correlative study for a long time is generally acknowledged that instantaneous electromyography signal is full of random noise, thus is not suitable for straight
Connect gesture for identification.Conventional method generally as unit of electromyography signal of the segment length more than or equal to 150 milliseconds, is extracted a variety of
Feature forms feature vector, reuses the classifiers such as support vector machines and classifies to feature vector, to identify one section of flesh
The corresponding gesture of electric signal.
With the appearance of array electromyographic electrode, the electric potential field that muscle activity is formed in skin surface is in the space at each moment
Distribution can be recorded.The present invention acquires instantaneous electromyography signal using array electrode, according to the spatial arrangement of electrode by wink
When electromyography signal be converted into myoelectricity image, and classified using Image Classifier to myoelectricity image, to identify instantaneous myoelectricity
Gesture represented by signal.Electromyogram seems a kind of spatial sampling of above-mentioned electric potential field, has contained muscle activity at a time
Global characteristics, quickly (zero observation time delay) accurately can identify gesture.Document there is no to use instantaneous electromyography signal both at home and abroad
Carry out gesture identification.
Depth convolutional neural networks are a kind of feedforward neural networks, its artificial neuron can respond a part covering model
Enclose interior surrounding cells.The model is commonly used in image classification.
Summary of the invention
In view of the above-mentioned deficiencies in the prior art, it is an object of the present invention to provide a kind of gesture identification based on instantaneous electromyography signal
Method.
The purpose of the present invention is achieved through the following technical solutions: a kind of gesture identification based on instantaneous electromyography signal
Method, method includes the following steps:
(1) instantaneous electromyography signal is acquired by array electrode；
(2) the instantaneous electromyography signal of array electrode acquisition is pre-processed, generates instantaneous myoelectricity image, it is specific comprising such as
Lower step:
(2.1) in the instantaneous electromyography signal linear transformation to 0 to 255 sections in the section general ± 2.5mV；
(2.2) instantaneous electromyography signal is arranged into instantaneous myoelectricity image according to the spatial position of electrode；
(3) the instantaneous myoelectricity Image Classifier of training, obtains the model parameter of classifier；
(4) gesture is identified using the instantaneous myoelectricity Image Classifier of training in step (3), specifically include following sub-step:
(4.1) pretreatment in step (2) is carried out to the instantaneous electromyography signal to be identified of array electrode acquisition, obtain to
The instantaneous myoelectricity image of processing；
(4.2) model parameter that training obtains in step (3) is brought into instantaneous myoelectricity Image Classifier and is identified wait locate
The corresponding gesture of instantaneous myoelectricity image of reason.
Further, in step (2.2), by the instantaneous electromyography signal value at each moment according to the spatial position of array electrode
It is arranged into matrix, forms instantaneous myoelectricity image；Instantaneous electromyogram seems single pass gray level image, height and width respectively with
The line number and columns of array electrode are equal.
Further, in step (3) using convolution deep neural network (Deep Convolutional Network,
ConvNets it) is used as instantaneous myoelectricity Image Classifier, specifically includes following sub-step:
(3.1) using the VGGNet initialization of pre-training on color image for gesture of classifying from instantaneous myoelectricity image
Network model；
(3.2) using the instantaneous electromyography signal of more people gathered in advance through the pretreatment in step (2), using pre-processing
The instantaneous myoelectricity image and its corresponding gesture label training network model arrived, obtains network model parameter.
Further, natural for classify animal, vehicle etc. using colored natural image pre-training in step (3.1)
Then 16 layers of VGGNet of object is input with instantaneous myoelectricity image using preceding 4 convolutional layers initialization of the network, uses
In the network model of classification gesture；Wherein the initial method of first convolutional layer is: by the first layer of the VGGNet of pre-training
3 channels RGB weight matrix summation, and using result as the weight matrix of the first layer of new network model.
The beneficial effects of the present invention are: in the training stage, the first step carries out the instantaneous electromyography signal of array electrode acquisition
Pretreatment, obtains instantaneous myoelectricity image；Second step is classified using instantaneous myoelectricity image and its corresponding gesture label training image
Device, such as depth convolutional neural networks, obtain network model parameter.In test phase, the first step, to array electrode acquisition to
The instantaneous electromyography signal of identification is pre-processed, and instantaneous myoelectricity image is obtained；Second step, by trained network model parameter band
Enter to identify every corresponding gesture of instantaneous myoelectricity image into network model.The present invention is based on instantaneous myoelectricity image and image classifications
Method can rapidly and accurately identify gesture.Compared to conventional method, the present invention, can be with zero after user makes gesture
Observation time delay rapidly identifies gesture.
Detailed description of the invention
Fig. 1 is the method for the invention flow chart；
Fig. 2 is the gesture collection comprising 8 kinds of hand gestures that present invention experiment uses；
Fig. 3 is electrode patch schematic diagram used in the present invention；
Fig. 4 is depth convolutional neural networks structure used in the present invention.
Specific embodiment
Invention is further described in detail in the following with reference to the drawings and specific embodiments.
As shown in Figure 1, a kind of gesture identification method based on instantaneous electromyography signal provided by the invention, including following step
It is rapid:
(1) instantaneous electromyography signal is acquired using array electrode, a kind of preferred reality for acquiring instantaneous electromyography signal is given below
Apply example:
(1.1) it is tested sitting posture, right arm stretches horizontally forward, and the centre of the palm is towards front-left.
(1.2) it using the skin of alcohol cleaning right forearm electrode to be pasted, is dried to it.
(1.3) electrode (dot expression array electrode) is pasted according to Fig. 3, circle electrode (reference electrode) is attached on the outside of wrist just
It is intermediate.
(1.4) connection acquisition equipment, acquires instantaneous electromyography signal.
(2) the instantaneous electromyography signal of array electrode acquisition is pre-processed, generates instantaneous myoelectricity image, it is specific comprising such as
Lower step:
(2.1) in the section general ± 2.5mV electromyography signal linear transformation to 0 to 255 sections；
(2.2) the instantaneous electromyography signal value at each moment is arranged into matrix according to the spatial position of array electrode, is formed
Instantaneous myoelectricity image.Instantaneous electromyogram seems single pass gray level image, height and width respectively with the line number of array electrode
It is equal with columns.
(3) the instantaneous myoelectricity Image Classifier of training, below by taking convolution deep neural network as an example:
Deep learning structure is a kind of multilayer perceptron containing more hidden layers.Deep learning is formed more by combination low-level feature
Add abstract high-rise expression attribute classification or feature, to find that the distributed nature of data indicates.Deep learning structure of the present invention
It is to be modified from VGGNet network structure, this modification allows VGGNet to identify the corresponding hand of instantaneous myoelectricity image
Gesture, this network structure include following network layer:
A) convolutional layer (Conv): every layer of convolutional layer is made of several convolution units in convolutional neural networks, each convolution list
The parameter of member is optimized by back-propagation algorithm.The purpose of convolution algorithm is to extract the different characteristic of input, the
One layer of convolutional layer may can only extract some rudimentary features such as levels such as edge, lines and angle, and the network of more layers can be from low
The more complicated feature of iterative extraction in grade feature.The present invention uses the convolution mask of 3x3.
B) ReLU activation primitive: ReLU can make the output 0 of a part of neuron, so that network is more sparse, and
And reduce the relation of interdependence of parameter, reduce the over-fitting of network in the training process.
C) full articulamentum (Fc): the full articulamentum in CNN model is the hidden layer of traditional neural network.The last one is connected entirely
Layer, i.e. output layer, related to specific classification task, output dimension is equal to gesture number to be identified.
D) Dropout layers: Dropout is mainly used to solve in deep learning because over-fitting is asked caused by quantity of parameters
Topic.Its main thought is: in the training process, throwing away a certain proportion of neuron (including their connection), in this way meeting at random
Prevent interneuronal excessive total adaptation.Dropout ratio of the invention is 0.5.
E) Softmax layers: Softmax layer is that softmax model is applied to last training result to carry out returning operation
Layer.Softmax model is popularization of the Logic Regression Models on more classification problems (categorical measure i.e. to be sorted is greater than 2).
Specific network structure is as shown in Figure 4.The wherein output dimension of digital representation this layer after Conv and Fc.
The step specifically includes following sub-step:
(3.1) it is used to classify 16 layers of the natural forms such as animal, vehicle using colored natural image pre-training
Then VGGNet network is input with the myoelectricity image of gray scale using preceding 4 convolutional layers initialization of the network, for classifying
The network model of gesture.Wherein the initial method of first convolutional layer is: by the RGB 3 of the first layer of the VGGNet of pre-training
The power in a channel is summed, and using result as the weight matrix of the first layer of new network model.
(3.2) pretreatment in step (2) is carried out using the instantaneous electromyography signal of more people gathered in advance, uses pretreatment
Instantaneous myoelectricity image and its corresponding gesture label training network model are obtained, network model parameter is obtained.
(4) gesture is identified using the instantaneous myoelectricity Image Classifier of training in step (3), specifically include following sub-step:
(4.1) pretreatment in step (2) is carried out to the instantaneous electromyography signal to be identified of array electrode acquisition, obtains wink
When myoelectricity image；
(4.2) the network model parameter that training obtains in step (3) is brought into and identifies wink to be processed in network model
When the corresponding gesture of myoelectricity image.
Embodiment
The present invention is based on instantaneous electromyography signals to identify gesture, mainly includes two parts: off-line training part according to Fig. 1
With online recognition part.
Off-line training part includes:
It a. is that tester pastes electrode according to Fig. 3.
B. 8 kinds of gestures in Fig. 2 are acquired.Every kind of gesture acquires 10 times, persistently has an effect 3 seconds every time, has an effect it twice in succession
Between rest 7 seconds.
C. the instantaneous electromyography signal of array electrode acquisition is pre-processed, the section general ± 2.5mV electromyography signal linearly becomes
It changes in 0 to 255 sections.The instantaneous electromyography signal value at each moment is arranged into matrix according to the spatial position of array electrode,
Form instantaneous myoelectricity image.
D. 16 layers of the VGGNet net using colored natural image pre-training for the natural forms such as animal, vehicle of classifying
Then network is input with the instantaneous myoelectricity image of gray scale using preceding 4 convolutional layers initialization of the network, for gesture of classifying
Network model.It according to the ratio cut partition of 7:3 is training set and verifying by the instantaneous myoelectricity image of step c and its gesture label
Collection.Using training set by the iteration optimization training network model, and tested on verifying collection, the accuracy rate on verifying collection
Stop iteration when no longer rising.
Online recognition part includes:
A. electrode is pasted according to the step a of off-line training.
B. instantaneous electromyography signal is acquired, generates instantaneous myoelectricity image sequence according to the step c of off-line training.
C. every corresponding gesture of instantaneous myoelectricity image is identified using trained network model.
D. (optional) agrees to rule ballot according to majority to the gesture label in nearest 150 milliseconds
(MajorityVote), poll soprano is as the corresponding gesture of this section of electromyography signal, to improve gesture identification accuracy rate.
Claims (4)
1. a kind of gesture identification method based on instantaneous electromyography signal, which comprises the following steps:
(1) instantaneous electromyography signal is acquired by array electrode；
(2) the instantaneous electromyography signal of array electrode acquisition is pre-processed, generates instantaneous myoelectricity image, it is specific to include following step
It is rapid:
(2.1) in the instantaneous electromyography signal linear transformation to 0 to 255 sections in the section general ± 2.5mV；
(2.2) instantaneous electromyography signal is arranged into instantaneous myoelectricity image according to the spatial position of electrode；
(3) the instantaneous myoelectricity Image Classifier of training, obtains the model parameter of classifier；
(4) gesture is identified using the instantaneous myoelectricity Image Classifier of training in step (3), specifically include following sub-step:
(4.1) pretreatment in step (2) is carried out to the instantaneous electromyography signal to be identified of array electrode acquisition, obtained to be processed
Instantaneous myoelectricity image；
(4.2) will in step (3) the obtained model parameter of training be brought into instantaneous myoelectricity Image Classifier identify it is to be processed
The instantaneously corresponding gesture of myoelectricity image.
2. a kind of gesture identification method based on instantaneous electromyography signal according to claim 1, which is characterized in that step
(2.2) in, the instantaneous electromyography signal value at each moment is arranged into matrix according to the spatial position of array electrode, forms instantaneous flesh
Electrical image；Instantaneous electromyogram seems single pass gray level image, height and width respectively with the line number of array electrode and columns
It is equal.
3. a kind of gesture identification method based on instantaneous electromyography signal according to claim 1, which is characterized in that step
(3) specifically include following sub-step using convolution deep neural network as instantaneous myoelectricity Image Classifier in:
(3.1) network using the VGGNet initialization of pre-training on color image for gesture of classifying from instantaneous myoelectricity image
Model；
(3.2) it is obtained through the pretreatment in step (2) using pretreatment using the instantaneous electromyography signal of more people gathered in advance
Instantaneous myoelectricity image and its corresponding gesture label training network model, obtain network model parameter.
4. a kind of gesture identification method based on instantaneous electromyography signal according to claim 3, which is characterized in that step
(3.1) in, using colored natural image pre-training for animal of classifying, 16 layers of VGGNet of vehicle natural forms, then
Being initialized using preceding 4 convolutional layers of the network with instantaneous myoelectricity image is input, the network model for gesture of classifying；Its
In the initial method of first convolutional layer be: by the weight matrix in 3 channels RGB of the first layer of the VGGNet of pre-training
Summation, and using result as the weight matrix of the first layer of new network model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510973702.2A CN105608432B (en) | 2015-12-21 | 2015-12-21 | A kind of gesture identification method based on instantaneous myoelectricity image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510973702.2A CN105608432B (en) | 2015-12-21 | 2015-12-21 | A kind of gesture identification method based on instantaneous myoelectricity image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105608432A CN105608432A (en) | 2016-05-25 |
| CN105608432B true CN105608432B (en) | 2019-02-22 |
Family
ID=55988358
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510973702.2A Active CN105608432B (en) | 2015-12-21 | 2015-12-21 | A kind of gesture identification method based on instantaneous myoelectricity image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105608432B (en) |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106228116A (en) * | 2016-07-13 | 2016-12-14 | 华东理工大学 | A kind of grinding trembling recognition methods based on electric current instantaneous nonlinearity index |
| 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 |
| CN106236336A (en) * | 2016-08-15 | 2016-12-21 | 中国科学院重庆绿色智能技术研究院 | A kind of myoelectric limb gesture and dynamics control method |
| CN106980367B (en) * | 2017-02-27 | 2020-08-18 | 浙江工业大学 | Gesture recognition method based on electromyogram |
| CN107480697B (en) * | 2017-07-12 | 2020-04-03 | 中国科学院计算技术研究所 | Myoelectric gesture recognition method and system |
| CN107657233A (en) * | 2017-09-28 | 2018-02-02 | 东华大学 | Static sign language real-time identification method based on modified single multi-target detection device |
| CN107885324B (en) * | 2017-09-28 | 2020-07-28 | 江南大学 | Human-computer interaction method based on convolutional neural network |
| WO2019165972A1 (en) | 2018-02-28 | 2019-09-06 | Chan Russell Wade | Wearable gesture recognition device and associated operation method and system |
| CN108491077B (en) * | 2018-03-19 | 2020-06-16 | 浙江大学 | Surface electromyographic signal gesture recognition method based on multi-stream divide-and-conquer convolutional neural network |
| CN108453736A (en) * | 2018-03-22 | 2018-08-28 | 哈尔滨工业大学 | A kind of multiple degrees of freedom synchronization myoelectricity control method based on deep learning |
| CN108345873A (en) * | 2018-03-22 | 2018-07-31 | 哈尔滨工业大学 | A kind of multiple degrees of freedom body motion information analytic method based on multilayer convolutional neural networks |
| CN109662710A (en) * | 2018-12-06 | 2019-04-23 | 杭州电子科技大学 | A kind of EMG Feature Extraction based on convolutional neural networks |
| CN110333783B (en) * | 2019-07-10 | 2020-08-28 | 中国科学技术大学 | Irrelevant gesture processing method and system for robust electromyography control |
| CN111103976B (en) * | 2019-12-05 | 2023-05-02 | 深圳职业技术学院 | Gesture recognition method and device and electronic equipment |
| CN111898526B (en) * | 2020-07-29 | 2022-07-22 | 南京邮电大学 | Myoelectric gesture recognition method based on multi-stream convolution neural network |
| CN111870242A (en) * | 2020-08-03 | 2020-11-03 | 南京邮电大学 | Intelligent gesture action generation method based on electromyographic signals |
| CN114515146B (en) * | 2020-11-17 | 2024-03-22 | 北京机械设备研究所 | Intelligent gesture recognition method and system based on electrical measurement |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101371804A (en) * | 2008-07-16 | 2009-02-25 | 哈尔滨工业大学 | On-line recognizing method of hand gesture mode established based on sEMG |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8447704B2 (en) * | 2008-06-26 | 2013-05-21 | Microsoft Corporation | Recognizing gestures from forearm EMG signals |
-
2015
- 2015-12-21 CN CN201510973702.2A patent/CN105608432B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101371804A (en) * | 2008-07-16 | 2009-02-25 | 哈尔滨工业大学 | On-line recognizing method of hand gesture mode established based on sEMG |
Non-Patent Citations (3)
| Title |
|---|
| Advancing Muscle-Computer Interfaces with High-Density Electromyography;Christoph Amma 等;《Acm Conference on Human Factors in Computing Systems》;20150423;第929-938页 |
| High-density surface EMG maps from upper-arm and forearm muscles;Monica Rojas-Martínez 等;《Journal of NeuroEngineering and Rehabilitation》;20120930;正文第1-17页 |
| 基于高密度肌电电势图的正常吞咽过程可视化研究;朱明星 等;《集成技术》;20130731;第2卷(第4期);第39-43页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105608432A (en) | 2016-05-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105608432B (en) | A kind of gesture identification method based on instantaneous myoelectricity image | |
| CN105654037B (en) | A kind of electromyography signal gesture identification method based on deep learning and characteristic image | |
| CN106980367B (en) | Gesture recognition method based on electromyogram | |
| CN108491077B (en) | Surface electromyographic signal gesture recognition method based on multi-stream divide-and-conquer convolutional neural network | |
| Esfahani et al. | Classification of primitive shapes using brain–computer interfaces | |
| CN113288183B (en) | Silent voice recognition method based on facial neck surface myoelectricity | |
| CN109199414A (en) | A kind of audiovisual induction Emotion identification method and system based on EEG signals | |
| CN106909220A (en) | A kind of sight line exchange method suitable for touch-control | |
| Wang et al. | Improved multi-stream convolutional block attention module for sEMG-based gesture recognition | |
| Flagg et al. | Conductive fur sensing for a gesture-aware furry robot | |
| CN109766845A (en) | A kind of Method of EEG signals classification, device, equipment and medium | |
| CN109685071A (en) | Brain electricity classification method based on the study of common space pattern feature width | |
| Zhang et al. | Intelligent Facial Action and emotion recognition for humanoid robots | |
| CN106774851A (en) | Sense of touch finger motion rehabilitation system and method based on brain-computer interface | |
| Wei et al. | A novel multi-dimensional features fusion algorithm for the EEG signal recognition of brain's sensorimotor region activated tasks | |
| CN108345873A (en) | A kind of multiple degrees of freedom body motion information analytic method based on multilayer convolutional neural networks | |
| Jo et al. | Real-time hand gesture classification using crnn with scale average wavelet transform | |
| CN109662710A (en) | A kind of EMG Feature Extraction based on convolutional neural networks | |
| CN109498362A (en) | A kind of hemiplegic patient's hand movement function device for healing and training and model training method | |
| Leelakittisin et al. | Compact CNN for rapid inter-day hand gesture recognition and person identification from sEMG | |
| Derr et al. | Signer-independent classification of American sign language word signs using surface EMG | |
| Jia | Neural network in the application of EEG signal classification method | |
| Suganya et al. | Design Of a Communication aid for physically challenged | |
| Shapiro et al. | Translating affective touch into text | |
| Gupta et al. | Hand Gesture recognition and classification by discriminant and principal component analysis using machine learning techniques |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |