福建电脑2024，Vol.40Issue(1)：39-43,5.DOI:10.16707/j.cnki.fjpc.2024.01.007
脑电信号的稳定扩散样本增强方法
EEG Sample Enhancement Method based on Stable Diffusion Model
摘要
Abstract
In response to the classification performance bottleneck caused by the scarcity of EEG signal samples,this paper proposes a stable diffusion model sample enhancement method for EEG.By converting EEG samples into time-frequency maps and using effective cue words to fine tune the image stabilization diffusion model,the EEG sample set is expanded to improve classification and recognition accuracy.The experimental results show that the classification accuracy of the sample set enhanced by the stable diffusion model has been improved from 72.94%to 76.80%,indicating that the proposed method effectively overcomes the classification performance bottleneck and provides a new approach for small sample EEG sample analysis.关键词
脑电信号/样本分类/稳定扩散模型/模式识别Key words
EEG Signal/Sample Classification/Stable Diffusion Model/Pattern Recognition分类
信息技术与安全科学引用本文复制引用
蔡子堃,罗天健..脑电信号的稳定扩散样本增强方法[J].福建电脑,2024,40(1):39-43,5.基金项目
本文得到福建省自然科学基金面上项目(No.2022J01655)资助. （No.2022J01655）
