A Method of Air Gesture Recognition Based on Inertial Sensor

Abstract

The invention discloses an air gesture recognition method based on an inertial sensor. The method extracts an air gesture signal sequence from a sensing signal obtained by an inertial sensor, and then obtains a training sample set, a verification sample set, and a test sample set after data preprocessing. At the same time, initialize the parameters of the LSTM-RNN model, and then train the LSTM-RNN model through the training sample set, and during the training process, input the verification samples in the verification sample set to the LSTM-RNN model trained in the iterative process , the number of iterations is controlled according to the error rate identified by the verification sample set to obtain the final LSTM‑RNN classifier; finally, the test sample is input to the final LSTM‑RNN classifier, and the test sample is identified by the final LSTM‑RNN classifier corresponding gesture. The method of the invention has the advantages of higher air gesture recognition precision and accuracy.

Description

technical field [0001] The invention belongs to the technical field of pattern recognition and artificial intelligence, in particular to an air gesture recognition method based on an inertial sensor. Background technique [0002] Inertial handwriting recognition based on inertial sensors (accelerometers and gyroscopes) is one of the emerging research frontiers in the computer field in recent years. The inertial sensors built into wearable devices or smart handheld devices on the user's body are used to collect the user's handwriting in the air. Acceleration signals and angular velocity signals, through machine learning and deep learning methods to identify the user's writing content, it is one of the important research contents of wearable computing (Wearable Computing) and ubiquitous computing (Ubiquitous Computing). [0003] At present, air gesture recognition based on inertial sensors has a wide range of applications in smart home, automatic driving, education, medical ca...

AI Technical Summary

Problems solved by technology

Because the signal waveform of the inertial sensor is very abstract, it is difficult to distinguish different gestures in the air simply by observing the waveform with the naked eye, and due to the differences in the behavior habits of different users, the signal waveform of the same gesture is also quite different, so the gesture recognition cannot only be based on the change of the signal value To make a judgment, it is necessary to mine the potential change law of the signal

Images