Wearable gesture language recognition device and gesture language translation method

Abstract

The invention discloses a wearable gesture language recognition device and a gesture language translation method. The wearable gesture language recognition device comprises a wearable equipment side and an upper computer side; the wearable equipment side is provided with a finger ring and an arm ring, specifically a real-time man-machine interaction system for resolving a motion posture and recognizing a mode of the gesture language behavior based on a multi-shaft motion sensor and a muscle electric signal sensor, thereby translating the gesture language into the voice of text. The muscle electric signal and the multi-shaft motion signal are merged to accurately recognize the gesture language behavior through the adoption of a mode recognition technology, and the recognition cannot be limited by the vocabulary of the gesture language behavior. The above signal acquisition, data handling and classifying processes are integrated on the wearable equipment so as to facilitate the wearing of the user and not influence the daily life of the user.

Description

technical field [0001] The present invention relates to the field of sign language translation systems, in particular to a wearable sign language recognition device and a sign language translation method, in particular to translating sign language into speech or text based on multi-axis motion sensors and muscle electrical signal motion gesture resolution and pattern recognition technology real-time human-computer interaction system. Background technique [0002] Sign language is the main tool for people with speech or hearing impairments to communicate with others, but the proportion of ordinary people who can use sign language proficiently is relatively low. Therefore, there are many inconveniences for people with language or hearing impairments to communicate with ordinary people. Due to the many types of sign language, taking Chinese sign language vocabulary as an example, there are more than 5,000 sign words in total; the dynamic changes of sign language movements and ...

AI Technical Summary

Problems solved by technology

[0004] At present, the main technical solution for gesture recognition is to collect joint angle information based on data gloves, but the wearing of data gloves restricts the daily life of people with language or hearing impairments; the gesture recognition method based on image processing is limited by the environment, and it is not portable in terms of portability. Advantages

In addition, the Chinese patent application number: 201410376205.X "Sign language translation and display sound system based on myoelectric signals and motion sensors" proposes a device worn on the arm to collect arm myoelectric signals and multi-axis motion sensor signals , but according to its principle, it can only detect the posture of fingers and forearm, but not the relative posture of palm and arm, and the relative posture of palm and arm in sign language movements is of great significance for the distinction of sign language movements; according to its stated principle, there is no feedback feature, the hearing impaired cannot be sure that the translation system is working, so this method has technical flaws

Images