Unlock instant, AI-driven research and patent intelligence for your innovation.

Wearable device gesture recognition method based on neural network optimization

A wearable device, neural network technology, applied in neural learning methods, biological neural network models, character and pattern recognition, etc., can solve the problems of low recognition efficiency, long recognition cycle, large amount of training data, etc. speed, reduce training time, and improve the efficiency of gesture recognition

Active Publication Date: 2019-10-18
GUANGDONG UNIV OF TECH
View PDF10 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The present invention provides a neural network optimization based Gesture recognition method for wearable devices

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
  • Wearable device gesture recognition method based on neural network optimization
  • Wearable device gesture recognition method based on neural network optimization
  • Wearable device gesture recognition method based on neural network optimization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0052] Such as figure 1 As shown, a gesture recognition method for wearable devices based on neural network optimization includes the following steps:

[0053]S1: Use wearable devices to collect gesture motion data; it should be noted that in this embodiment, data gloves are used as smart wearable devices to collect data; Noise is generated.

[0054] S2: The gesture movement data is normalized and filtered to obtain the training sample gesture data for neural network training; in this embodiment, the gesture movement data is mapped to the [0,1] interval for normalization deal with.

[0055] The gesture motion data [x 0i ,y 0i ] to perform normalization processing, that is, to map the gesture motion data to the [0,1] interval to obtain the normalized data [x i ,y i ],which is:

[0056]

[0057]

[0058] where x 0i,min ,x 0i,max Respectively represent x in the random representative sample data 0i The minimum and maximum values ​​of , where y 0i,min ,y 0i,max It...

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 discloses a wearable device gesture recognition method based on neural network optimization. The method comprises the following steps: collecting gesture motion data; preprocessing to obtain training sample gesture data; determining a gesture template of the static gesture; performing clustering fusion on the training sample gesture data by using a clustering fusion algorithm; establishing and optimizing an RBF neural network inputting the training sample gesture data obtained after clustering processing into the optimized RBF neural network to be learned and trained to obtain corresponding classifiers, judging corresponding static gestures according to output results of the classifiers, and calculating classification errors; constructing a hybrid clustering cluster, and training the optimized RBF neural network by using the hybrid clustering cluster; and performing a gesture recognition test by using the trained RBF neural network. Gesture data are clustered through a clustering fusion algorithm, so that a training data set is reduced, the training time is shortened. The gesture recognition efficiency is improved, a neural network is optimized through a genetic algorithm, and the recognition accuracy is improved.

Description

technical field [0001] The present invention relates to the field of intelligent recognition, and more specifically, to a gesture recognition method for wearable devices based on neural network optimization. Background technique [0002] With the development of today's science and technology, computers have penetrated into all aspects of production and life, and human-computer interaction technology is also improving day by day. Gesture, as the most common way of expression for human beings, is a natural, intuitive, efficient and easy-to-learn means of human-computer interaction. Gesture recognition technology is a key technology in the field of human-computer interaction, which realizes human gesture recognition through mathematical algorithms. The commonly used gesture recognition algorithms are divided into three categories: one is the recognition algorithm based on neural network, which has strong analysis ability for data; the other is the recognition algorithm based o...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/08
CPCG06N3/08G06V40/28G06F18/23213
Inventor 刘治廖佳培章云
Owner GUANGDONG UNIV OF TECH