Aerial handwriting inertial sensing signal generation method based on deep adversarial learning

Abstract

The invention discloses an aerial handwriting inertia sensing signal generation method based on deep adversarial learning, which comprises the following steps: carrying out filtering, smoothing and noise reduction on an obtained aerial handwriting inertia sensing signal sequence to serve as a training sample set; designing a deep convolution condition confrontation generation network based on timesequence feature map position coding; using the training sample set to train the generation network; and inputting the formulated sample length, the sample category label and the random noise vectorinto the trained adversarial generation network, and taking the input of the generator as a generated air handwriting inertial sensing signal sequence. According to the invention, aerial handwriting inertial sensing signal samples with certain diversity and good quality can be generated, and the effective lengths and types of the generated samples are controllable.

Description

technical field [0001] The invention relates to the technical fields of deep learning and artificial intelligence, in particular to a method for generating inertial sensing signals of handwriting in the air based on deep confrontation learning. Background technique [0002] Air handwriting recognition based on inertial sensors (accelerometers and gyroscopes) is one of the emerging cutting-edge research directions in the computer field in recent years, and has a wide range of applications in smart home, automatic driving, education, medical care, industrial production, and auxiliary communication. . A handwriting recognition model in the air often requires sufficient training samples to obtain better generalization capabilities. However, there are few public air handwriting data sets based on inertial sensors. There are three main reasons for the scarcity of data. First, the process of sensor data collection is time-consuming and laborious. Secondly, the poor readability of...

AI Technical Summary

Problems solved by technology

First, the process of sensor data collection is time-consuming and laborious

Secondly, the poor readability of inertial sensor data leads to difficulties in data cleaning and labeling

Third, handwritten data is related to the personal privacy of the collected person and is not easy to be disclosed

However, most of the current adversarial generative network structures are specially designed for image data. When directly applied to aerial handwritten data, problems such as poor generation effect and mode collapse are prone to occur.

Images