An electroencephalogram-image collaborative contrast learning method based on asymmetric feedback

By constructing an EEG-image collaborative contrast learning method based on asymmetric feedback, and using the feedback signal of the correctly determined modality to optimize the model parameters, the problem of insufficient determination stability and self-optimization ability of EEG and visual models in RSVP tasks in the existing technology is solved, and high-precision target detection in complex scenes is achieved.

CN122398331APending Publication Date: 2026-07-17BEIJING MECHANICAL EQUIP INST
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
BEIJING MECHANICAL EQUIP INST
Filing Date
2026-04-23
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

In existing technologies, RSVP target detection methods based on EEG and visual models suffer from insufficient decision stability and self-optimization capabilities. Especially in complex scenes or under low-quality image conditions, the robustness of single-modal methods is poor, and the complementary relationship between EEG and visual decision results is not fully explored.

Method used

We construct an EEG-image co-contrast learning method based on asymmetric feedback. Through the co-contrast learning model, we utilize EEG decoder, image decoder, mapper and unified embedding space to achieve cross-modal feature embedding and dynamic feedback correction, optimize model parameters, and use feedback signals of correctly determined modalities to correct erroneous modalities, thus forming a correction mechanism of visual guidance and neural guidance.

Benefits of technology

It improves the accuracy and robustness of target detection in RSVP tasks, and achieves self-optimization of human-machine collaborative judgment through dynamic feedback mechanism, thereby improving detection accuracy and stability under complex conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122398331A_ABST
    Figure CN122398331A_ABST
Patent Text Reader

Abstract

The present application relates to a kind of electroencephalogram-image collaborative contrast learning method based on asymmetric feedback, belong to electroencephalogram-image collaborative processing technical field, solve the problem that the stability and self-optimization ability of man-machine collaborative determination system is insufficient due to the difference information of the determination result of electroencephalogram and visual model is not effectively utilized in prior art dynamic feedback learning.The method comprises: the multi-modal data flow of the subject in the process of RSVP experiment is synchronously collected and preprocessed;Each preprocessed electroencephalogram signal, corresponding stimulus image and real label are constructed into aligned electroencephalogram-image data pair;Each group of electroencephalogram-image data is trained using collaborative contrast learning model based on asymmetric feedback electroencephalogram-image collaborative contrast learning, and the parameters of collaborative contrast learning model are optimized;The optimized collaborative contrast learning model is used to determine the target of the preprocessed electroencephalogram signal or stimulus image of new subject.
Need to check novelty before this filing date? Find Prior Art