Open AccessResearch on Gesture Recognition Method Based on Deep Learning
Author(s) -
YiZhi Li
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1861/1/012049
Subject(s) - gesture , convolutional neural network , computer science , gesture recognition , artificial intelligence , set (abstract data type) , deep learning , artificial neural network , machine learning , human–computer interaction , programming language
Gestures are necessary and indispensable in people’s ordinary life communication. And gestures have become an important part of human-computer interaction. With the gradual maturity of human-computer interaction systems, the role of gestures in human-computer interaction is becoming more and more important. Due to the variety of gestures in practical use, coupled with factors such as environment and light changes, it makes it very difficult for computers to recognize the correct gestures from image information. Deep learning has achieved certain research results through several years of development, which provides new research ideas for gesture recognition. This paper firstly introduces the common methods and processes of gesture recognition, and then proposes a new convolutional neural network structure by studying convolutional neural networks and adding Inception structure to the network to improve the performance of the network. Finally, experimental validation is conducted, and the experimental results show that the new convolutional neural network achieves an average recognition rate of 97.8% on the test set, which is a good result.