Research on Sign Language Translation System Based on Data Glove
Targeting at the problem that the deaf-mute and normal people can't communicate effectively,a gesture recognition and sign lan-guage translation system based on multi-sensor is designed.Firstly,the gesture translation glove is researched and used based on the flexi-ble sensor and inertial measurement unit(IMU).Secondly,the gesture output by the deaf-mute is collected.The continuous gesture data are segmented by using template matching method,and the intercepted data are preprocessed and feature template is extracted.By using the feature template similarity algorithm,the gesture of the deaf-mute is compared and analyzed with the gestures in the template database to recognize the sign language information,and the sign language information and its corresponding voice entries are output based on the voice module.Finally,the experiment on the success rate of gesture segmentation and recognition is carried out.The results show that the success rates of single gesture segmentation and recognition are 97% and 98%,and the success rate of continuous sentence recognition is 84%,demonstrating that the proposed system can be used to help the deaf mute communicate with the normal people.
data glovefeature templategesture segmentationgesture recognitionsimilarity algorithm
国家科技期刊平台
NSTL
万方数据
