A Portable Motion Pattern Real-time Recognition System Based on Multi-source Signals

Abstract

A portable motion pattern real-time recognition system based on multi-source signals, comprising a portable multi-source signal acquisition device, and a host computer carrying out data exchange with it through Wi-Fi; the host computer sets the working mode of the portable multi-source signal acquisition device, And control its startup and shutdown; preprocess the collected multi-source signals, detect the starting point of motion, extract real-time features, train classification models, and recognize motion patterns in real time; archive and manage subject information and experimental data; Develop personalized experimental paradigms for users, complete paradigm design and generation, and provide a control method for experimental paradigm presentation and signal acquisition synchronization; the present invention can realize real-time transmission, processing, display, storage and real-time automatic movement of multi-source signals At the same time, it also provides a way to design and generate experimental paradigms; it provides an ideal signal acquisition front-end and implementation platform for the research and application of real-time recognition of motion patterns.

Description

technical field [0001] The invention relates to the technical field of rehabilitation aids and man-machine interface, in particular to a portable motion pattern real-time recognition system based on multi-source signals. Background technique [0002] Motion pattern recognition technology can realize the recognition and prediction of the activity pattern of the whole or part of the human body in different scenarios, and can be widely used in many fields. For example, in the field of prosthetic rehabilitation, motion pattern recognition can be used to help physically disabled people realize prosthetic control; in the field of human-computer interaction, through the identification of human motion intentions, the motion intentions can be transformed into control instructions to control and interact with machines . [0003] According to the differences in the types of signals used, human motion pattern recognition can be roughly divided into two categories, one is an image-based...

AI Technical Summary

Problems solved by technology

When training a classification model, it is often necessary to collect signals of a specific training paradigm. Generally, experimental stimuli or experimental prompts are presented through other external software or tools, which makes it difficult to achieve synchronous acquisition of the training task stimulus sequence and the signal, which is useful for subsequent EMG signal analysis. Difficulty in positioning and selection of signal fragments in

In summary, the existing motion pattern recognition system has defects such as lack of synchronous acquisition of multi-source signals, poor portability, motion pattern recognition cannot run in real time, and lack of experimental paradigm design and presentation function, which greatly restricts the actual use of the system. cause a lot of inconvenience

Images