Brain-computer information fusion classification method and system for shared subspace learning

Abstract

The invention belongs to the technical field of brain-computer interface technology application, and discloses a brain-computer information fusion classification method and system for shared subspace learning, and the brain-computer information fusion classification method comprises a training stage and a reasoning stage. In the training stage, paired images and brain response data are utilized, shared subspace model parameters of the images and brain responses are optimized through a contrast learning strategy of positive and negative sample sampling, and an image classifier is trained; in the reasoning stage, image features are extracted for classification, and the application target of the whole brain-computer information fusion classification system is achieved. According to the brain-computer information fusion classification system based on shared subspace learning, the shared subspace can be trained in an end-to-end mode, efficient migration of brain cognitive information is achieved, and the performance of an image classification task in a complex open scene is improved; through the application that the brain is not in the loop, the efficiency and the stability in the practical application are improved, and the method has a wide application prospect under a new normal form of brain-computer information cooperative work.

Description

technical field [0001] The invention belongs to the technical field of application of brain-computer interface technology, and in particular relates to a brain-computer information fusion classification method and system for sharing subspace learning. Background technique [0002] In recent years, artificial intelligence methods represented by deep learning have developed rapidly, and their performance on image classification tasks has surpassed that of humans. However, at present, deep learning systems are only applied on a large scale in limited specific and simple scenarios such as face recognition, speech recognition, and optical character recognition. They mainly rely on data-driven, and need to build appropriate models and use sufficient computing power to fully Mining the distribution rules in massive data cannot achieve human-like cognitive ability. Therefore, in the face of complex open scenes such as complex and changeable targets / backgrounds, occlusion, and anti-...

AI Technical Summary

Problems solved by technology

[0007] (1) The existing technology is limited by the application paradigm of "brain in the loop", it is difficult to achieve high-intensity, real-time fully automated processing, and it is difficult to fully utilize the advantages of machine intelligence fully automated processing

[0008] (2) The existing technology is limited by the difficulty of obtaining brain response data. It is difficult to learn high-quality image-EEG shared subspace end-to-end under limited data, and it is difficult to realize the full migration of brain cognitive information.

[0009] (3) The existing self-adaptive information fusion classification method for the application of "brain not in the loop" is through multiple stages of separate learning and processing, the process is cumbersome, and it is troublesome when deploying the model

Images