Abstract
The paper deals with the problem of recognizing the single hand gestures received from a webcam. The model for processing of stream data and recognition of gestures from video images in the form of a 10-layer convolutional neural network is proposed. As a result of the model quality evaluation, the accuracy obtained on the test set was 96%, the value of the loss function 0.02. The results of the test showed that the model is resistant to relatively wide angles of hand rotation and is independent of light, due to the use of contours. Figs.: 3. Refs.: 10 titles. Keywords: model, data stream, recognition, sign language, convolutional neural network, contour.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
