Gesture recognition technology based on multi-modal information fusion

Abstract

The invention discloses a gesture recognition technology based on multi-modal information fusion, and belongs to the technical field of gesture recognition. The method comprises the following steps: step 1, acquiring a gesture action execution acquisition signal of a subject; 2, denoising the electromyographic signals and the three-axis acceleration and angle original data, extracting signal activity sections, and marking labels; 3, extracting time domain features and AR model parameters of the electromyographic signals, and extracting absolute average values and waveform length features of the three-axis acceleration signals and the angle signals; 4, carrying out fusion dimension reduction on the extracted features, and inputting the data subjected to dimension reduction into a classifier; 5, training to obtain an optimal prediction model; and step 6, inputting the acquired gesture signal needing to be predicted into the back propagation neural network optimized based on the genetic algorithm for identification, and outputting a result. According to the invention, information is extracted from multiple angles, the defect that single electromyographic signal information is not extracted completely is overcome, the diversity of gesture information is increased, and improvement of gesture recognition accuracy is facilitated.

Description

technical field [0001] The invention belongs to the technical field of gesture recognition, and in particular relates to a gesture recognition technology based on multimodal information fusion. Background technique [0002] With the development of artificial intelligence, human-computer interaction technology as its core has attracted more and more attention. The essence of human-computer interaction is the information exchange process between human and computer in order to complete certain tasks in a certain interactive way. As an important organ of the human body, the hand is not only used for labor, but also an important medium for information exchange. Gesture is an effective means of transmitting information. Its non-contact and diverse characteristics make it a common control method in the field of human-computer interaction. Hot topics in other fields. [0003] According to the current total population of China and relevant data of the Second National Sample Survey...

AI Technical Summary

Problems solved by technology

[0012] In view of the above-mentioned problems existing in the prior art, the purpose of the present invention is to provide a gesture recognition technology, which acquires human hand electromyographic signals and pose signals through sensors, and recognizes hand movements. The method extracts information from multiple angles. , which makes up for the shortcoming of incomplete extraction of single EMG signal information, increases the diversity of gesture information, and is conducive to the improvement of gesture recognition accuracy

Images