Abstract
In order to develop a non-contact and simple gesture recognition technology, a recognition method with a charge induction array of nine electrodes is proposed. Firstly, the principle of signal acquisition based on charge induction is introduced, and the whole system is given. Secondly, the recognition algorithms, including the pre-processing algorithm and back propagation neural network (BPNN) algorithm, are given to recognize three input modes of hand gestures, digital input, direction input and key input, respectively. Finally, experiments of three input modes of hand gestures are carried out, and the recognition accuracy is 97.2%, 94%, and 100% for digital input, direction input, and key input, respectively. The outstanding characteristic of this method is the real-time recognition of three hand gestures in the distance of 2 cm without the need of wearing any device, as well as being low cost and easy to implement.
Funder
National Nature Science Foundation Program of China
Subject
Electrical and Electronic Engineering,Biochemistry,Instrumentation,Atomic and Molecular Physics, and Optics,Analytical Chemistry
Reference25 articles.
1. Said, F.N., Yacoub, A., and Suen, C.Y. (1999, January 20–22). Recognition of English and Arabic numerals using a dynamic number of hidden neurons. Proceedings of the Fifth International Conference on Document Analysis and Recognition, ICDAR’99 (Cat. No.PR00318), Bangalore, India.
2. Fully Convolutional Sequence Recognition Network for Water Meter Number Reading;IEEE Access,2019
3. HDSR-Flor: A Robust End-to-End System to Solve the Handwritten Digit String Recognition Problem in Real Complex Scenarios;IEEE Access,2020
4. A hybrid deep learning CNN-ELM model and its application in handwritten numeral recognition;J. Comput. Inf. Syst.,2015
5. Handwritten numeral recognition using gradient and curvature of gray scale image;Pattern Recognit.,2002