基于多脑区注意力机制胶囊融合网络的EEG-fNIRS情感识别

刘悦,张雪英,陈桂军,黄丽霞,孙颖

EEG-fNIRS emotion recognition based on multi-brain attention mechanism capsule fusion network

Yue LIU,Xueying ZHANG,Guijun CHEN,Lixia HUANG,Ying SUN

表 6 与其他情感识别模型对比

Tab.6 Contrast with other emotion that recognition models

模型 Acc/% 平均值（标准差） Sad Happy Calm Fear SVM 89.60(6.59) 90.98 89.13 88.96 89.31 2DCNN 90.56(4.96) 90.48 90.20 89.63 91.93 gcForest 81.91(8.63) 82.76 80.27 82.43 82.17 Transformer 86.84(9.25) 85.62 89.35 83.84 88.54 GCN 90.16(5.10) 90.15 90.71 88.45 91.33 MFM-CapsNet[10] 92.74(3.14) 92.1 93.52 91.61 93.73 MLF-CapsNet[11] 94.65(3.80) 94.48 94.79 93.46 95.86 ST-CapsNet[12] 94.01(2.95) 93.57 95.02 93.04 94.42 MBA-CF-cCapsNet 96.67(2.68) 96.57 97.31 95.02 97.76