Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Model, training method and surface electromyogram signal gesture recognition method

A training method and EMG technology, applied in the field of biometrics, can solve the problems of large amount of calculation and low accuracy of gesture recognition of surface EMG

Pending Publication Date: 2021-11-26
NANTONG UNIVERSITY
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a HDC-BiGRU-Attention model and training method, and a surface electromyography signal gesture recognition method, which are used to improve the overfitting phenomenon that cannot be avoided during model training in the prior art and affect the final recognition of the model. effect, resulting in the problem of low accuracy and large amount of calculation for gesture recognition based on surface electromyography signals

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Model, training method and surface electromyogram signal gesture recognition method
  • Model, training method and surface electromyogram signal gesture recognition method
  • Model, training method and surface electromyogram signal gesture recognition method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0040] Please refer to figure 1 , figure 1 Shown is a schematic structural diagram of an HDC-BiGRU-Attention model provided in the embodiment of the present application. An HDC-BiGRU-Attention model, which includes a mixed dilated convolution module, a Maxpooling pooling layer, a first Fullconnection layer, a BiGRU layer, an Attention layer, a second Fullconnection layer, and a Softmax layer arranged in sequence according to a processing direction. The mixed hole convolution module is used to receive the EMG signal, extract the features of the EMG signal, and transmit the features to the Maxpooling pooling layer. The above hybrid atrous convolution module can not only expand the receptive field without increasing the number of first data in the training set, but also reduce the network depth and reduce overfitting. The Maxpooling pooling layer is used to process the features and then input the features into the first Fullconnection layer. Since the training set belongs to a...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides an HDC-BiGRU-Attention model, a training method and a surface electromyogram signal gesture recognition method, and relates to the technical field of biological characteristics. The HDC-BiGRU-Attention model comprises a mixed cavity convolution module, a Maxpooling layer, a first Fullconnection layer, a BiGRU layer, an Attention layer, a second Fullconnection layer and a Softmax layer, wherein the mixed cavity convolution module, the Maxpooling layer, the first Fullconnection layer, the BiGRU layer, the Attention layer, the second Fullconnection layer and the Softmax layer are sequentially arranged in the processing direction. The HDC-BiGRU-Attention model does not need manual feature extraction, the workload is reduced, the efficiency is improved, the overfitting phenomenon generated during model training can be avoided, the gesture recognition accuracy of the surface electromyogram signals is improved, and the calculation amount is reduced.

Description

technical field [0001] The invention relates to the technical field of biometric features, in particular to a model and a training method, and a surface electromyography signal gesture recognition method. Background technique [0002] In recent years, with the rapid development of science and technology, the way of human-computer interaction has also been greatly changed. Gesture recognition is a very important process in gesture-based human-computer interaction. During gesture recognition, the general process is to first extract the features of the gesture, and then perform gesture recognition according to an effective recognition method based on the extracted features. [0003] There are many traditional gesture recognition methods. For example, the neural network-based recognition method has a strong classification ability for recognition and classification. However, the number of neural network layers used in this method is generally shallow, and it is prone to overfitt...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06K9/62G06K9/00G06N3/04A61B5/389
CPCA61B5/389G06N3/045G06F2218/08G06F18/214Y02T10/40
Inventor 张凯陈峰
Owner NANTONG UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com