Multi-scale motor imagery cross-subject identification method combining global and subdomain adaptation

The invention discloses a multi-scale motor imagery cross-subject recognition method and system combining global and subdomain adaptation. The method comprises the steps that electroencephalogram signals of multiple subjects are divided into source domain data and target domain data, and the source...

Full description

Saved in:
Bibliographic Details
Main Authors LYU ZHAO, LI PING, YANG FAN, LI XINHUI, FAN CUNHANG, PEI SHENGBING
Format Patent
LanguageChinese
English
Published 27.09.2024
Subjects
CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
DIAGNOSIS
ELECTRIC DIGITAL DATA PROCESSING
HUMAN NECESSITIES
HYGIENE
IDENTIFICATION
MEDICAL OR VETERINARY SCIENCE
PHYSICS
SURGERY
Online AccessGet full text

Cover

More Information
Summary:The invention discloses a multi-scale motor imagery cross-subject recognition method and system combining global and subdomain adaptation. The method comprises the steps that electroencephalogram signals of multiple subjects are divided into source domain data and target domain data, and the source domain data and the target domain data are preprocessed; a multi-scale feature extraction unit is adopted to obtain spatio-temporal features; introducing a global domain classifier to train domain discrimination, and aligning edge distribution; a sub-domain classifier is introduced to learn local domain migration, and condition distribution is aligned; training a deep classifier by using the spatio-temporal features after two times of alignment; and converging the total loss function value through iterative training to obtain a motor imagery cross-subject recognition model. The method combines a multi-scale convolutional layer to capture space-time structure information, a global domain classifier to reduce cross-d
Bibliography:Application Number: CN202410702331