Gesture recognition method fusing electromyographic signal and visual image

Abstract

The invention discloses a gesture recognition method fusing an electromyographic signal and a visual image, and the method comprises the steps: collecting an electromyographic signal through a high-density electromyographic signal collection arm band, collecting a visual image through a camera, and fusing an electromyographic image converted from the electromyographic signal with the visual imageto obtain an image; and inputting the image obtained by fusing the surface electromyographic image and the visual image into an image classification module for identification and classification. According to the method, two different network structures do not need to be used for identifying and classifying the surface electromyographic signals and the visual images, the surface electromyographic signals and the visual images only need to be transmitted into the network architecture at the same time, identification and classification can be carried out by using the same network, the network complexity can be effectively reduced, and limited resources are saved. According to the invention, the form of using a single input mode to carry out gesture recognition in each set of traditional network can be improved, and the respective limitations that the surface electromyographic signal and the visual image are respectively used for gesture recognition are improved.

Description

technical field [0001] The invention relates to the field of gesture recognition, in particular to a gesture recognition technology that fuses myoelectric signals and visual images. Background technique [0002] The invention comes from the problem of gesture recognition of electromyographic signals and visual images. In daily communication between people, gesture is a widely used and frequent way of interaction. At the same time, sign language is an important and convenient way for deaf people to communicate with each other. Over the years, researchers have been exploring gesture recognition technology more and more deeply, and gesture recognition has already become one of the hot spots in the field of human-computer interaction. [0003] There are many technologies to realize gesture recognition, among which vision-based gesture recognition technology is the most widely researched and applied, and the most convenient. This non-contact method of collecting gestures does n...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problems of limited application scenarios of visual image gesture recognition and large differences between different individuals in electromyographic signals, and to reduce the amount of calculation and storage space, the present invention provides a method based on electromyographic signals and visual images and uses a nerve Method for Gesture Recognition in Network

Images