Aerial handwriting character recognition method based on inertia sensor

Abstract

The invention discloses an aerial handwriting character recognition method based on an inertia sensor. The method comprises steps of carrying out data preprocessing on aerial handwriting character action sensing signals acquired by the inertia sensor and then acquiring a training sample set, a verification sample set and a test sample; carrying out parameter initialization on an LSTM-RNN model; training the LSTM-RNN model subjected to the parameter initialization through each training sample; in the training process, inputting verification samples in the verification sample set into an iteration process to train the obtained LSTM-RNN model; according to error rate of the recognition of the verification sample set, controlling the iteration frequency so as to obtain a final LSTM-RNN classifier; and finally, inputting the test samples into the final LSTM-RNN classifier and recognizing corresponding characters of the test samples through the final LSTM-RNN classifier. The method is advantaged by quite high recognition precision and accuracy of aerial handwriting characters.

Description

Technical field [0001] The invention relates to the technical field of pattern recognition and artificial intelligence, in particular to an air handwritten character recognition method based on inertial sensors. Background technique [0002] Air handwriting recognition based on inertial sensors (accelerometers and gyroscopes) is one of the emerging research frontiers in the computer field in recent years. It uses the inertial sensors built into the wearable device or smart handheld device to collect the user’s air writing process Acceleration signals and angular velocity signals are used to identify the user's writing content through machine learning and deep learning methods, which are one of the important research contents of wearable computing and ubiquitous computing. [0003] At present, aerial handwriting recognition based on inertial sensors mainly includes written content recognition and written person recognition. Writing content recognition is mainly to identify written ...

AI Technical Summary

Problems solved by technology

Writing content recognition is mainly to identify written characters, words, phrases, sentences and other specific content; writer recognition is mainly to realize the identity authentication of the writer, which can be used in fields such as handwritten signature identification; since the signal waveform of the inertial sensor is not intuitive , it is difficult to distinguish the writing content simply by observing the waveform with the naked eye, and due to the difference in writing habits of different people, the signal waveform of the same character is also quite different, so the gesture recognition cannot be judged only based on the change of the signal value, but It is necessary to mine the potential changes of the signal

[0004] Inertial handwriting based on inertial sensors is different from traditional two-dimensional plane writing, without any tactile and visual feedback during the writing process

Images