《计算机应用》唯一官方网站 ›› 2024, Vol. 44 ›› Issue (11): 3354-3363.DOI: 10.11772/j.issn.1001-9081.2023111593
        
                    
            杨思琪1,2, 罗天健1,2( ), 严宣辉1,2, 杨光局1,2
), 严宣辉1,2, 杨光局1,2
                  
        
        
        
        
    
收稿日期:2023-11-20
									
				
											修回日期:2024-03-19
									
				
											接受日期:2024-03-21
									
				
											发布日期:2024-03-22
									
				
											出版日期:2024-11-10
									
				
			通讯作者:
					罗天健
							作者简介:杨思琪(2000—),女,湖南衡阳人,硕士研究生,CCF会员,主要研究方向:脑机接口、模式识别基金资助:
        
                                                                                                                            Siqi YANG1,2, Tianjian LUO1,2( ), Xuanhui YAN1,2, Guangju YANG1,2
), Xuanhui YAN1,2, Guangju YANG1,2
			  
			
			
			
                
        
    
Received:2023-11-20
									
				
											Revised:2024-03-19
									
				
											Accepted:2024-03-21
									
				
											Online:2024-03-22
									
				
											Published:2024-11-10
									
			Contact:
					Tianjian LUO   
							About author:YANG Siqi, born in 2000, M. S. candidate. Her research interests include brain computer interface, pattern recognition.Supported by:摘要:
运动想象脑电(MI-EEG)信号在构建临床辅助康复的无创脑机接口(BCI)中获得了广泛关注。受限于不同被试者的MI-EEG信号样本分布存在差异,跨被试MI-EEG信号的特征学习成为研究重点。然而,现有的相关方法存在域不变特征表达能力弱、时间复杂度较高等问题,无法直接应用于在线BCI。为解决该问题,提出黎曼切空间特征迁移核学习(TKRTS)方法,并基于此构建了高效的跨被试MI-EEG信号分类算法。TKRTS方法首先将MI-EEG信号协方差矩阵投影至黎曼空间,并在黎曼空间上对齐不同被试者的协方差矩阵,同时提取黎曼切空间(RTS)特征;随后,学习RTS特征集上的域不变核矩阵,从而获得完备的跨被试MI-EEG特征表达,并通过该矩阵训练核支持向量机(KSVM)进行分类。为验证TKRTS方法的可行性与有效性,在3个公开数据集上分别进行多源域-单目标域以及单源域-单目标域的实验,平均分类准确率分别提升了0.81个百分点和0.13个百分点。实验结果表明,与主流方法对比,TKRTS方法提升了平均分类准确率并保持相似的时间复杂度。此外,消融实验结果验证了TKRTS方法对跨被试特征表达的完备性和参数不敏感性,适合构建在线脑接机口。
中图分类号:
杨思琪, 罗天健, 严宣辉, 杨光局. 运动想象脑电图的空域特征迁移核学习方法[J]. 计算机应用, 2024, 44(11): 3354-3363.
Siqi YANG, Tianjian LUO, Xuanhui YAN, Guangju YANG. Transfer kernel learning method based on spatial features for motor imagery EEG[J]. Journal of Computer Applications, 2024, 44(11): 3354-3363.
| 数据集 | 被试者数 | MI任务数 | 采样时间点数 | 通道数 | 样本数 | 
|---|---|---|---|---|---|
| MI2a | 9 | 4 | 750 | 22 | 288 | 
| MI2b | 9 | 2 | 750 | 3 | 400 | 
| MI4a | 5 | 2 | 300 | 118 | 280 | 
表1 3个实验数据集的统计信息
Tab. 1 Statistics for three experimental datasets
| 数据集 | 被试者数 | MI任务数 | 采样时间点数 | 通道数 | 样本数 | 
|---|---|---|---|---|---|
| MI2a | 9 | 4 | 750 | 22 | 288 | 
| MI2b | 9 | 2 | 750 | 3 | 400 | 
| MI4a | 5 | 2 | 300 | 118 | 280 | 
| 算法 | 平均分类准确率 | 均值 | |||||||
|---|---|---|---|---|---|---|---|---|---|
| MI2a-1 | MI2a-2 | MI2a-3 | MI2a-4 | MI2a-5 | MI2a-6 | MI2b | MI4a | ||
| CSP-TJM | 57.18 | 57.64 | 61.81 | 57.87 | 60.88 | 56.79 | 67.00 | 57.36 | 59.57 | 
| CSP-JDA | 61.37 | 62.35 | 54.58 | 57.41 | 57.53 | 51.25 | 58.28 | 60.80 | 57.95 | 
| CSP-LDA | 68.90 | 67.90 | 66.90 | 65.05 | 67.67 | 57.79 | 66.73 | 64.00 | 65.62 | 
| EA-CSP-LDA | 73.69 | 70.99 | 80.25 | 70.99 | 77.16 | 68.52 | 68.34 | 78.29 | 73.53 | 
| MDM | 59.62 | 59.83 | 55.24 | 56.08 | 53.88 | 52.58 | 58.14 | 60.69 | 57.01 | 
| RA-MDM | 72.07 | 72.99 | 79.48 | 69.21 | 77.01 | 66.28 | 69.69 | 77.07 | 72.98 | 
| MEKT | 76.31 | 73.46 | 81.10 | 80.86 | 69.98 | 69.47 | 76.37 | ||
| METL | 76.00 | — | — | — | — | — | — | — | — | 
| SB-TA-CSP | 75.15 | — | — | — | — | — | — | — | — | 
| TKCSP | — | — | — | — | — | — | — | 81.14 | — | 
| FWR-JPDA | 75.69 | 80.56 | 74.07 | 78.47 | 70.06 | — | — | — | |
| MMDA | 77.93 | — | — | — | — | — | — | 83.00 | — | 
| EA-CSP-JDA | 76.70 | 70.68 | 79.17 | 69.75 | 76.16 | 66.36 | 69.56 | 78.14 | 73.31 | 
| TKRTS-R | 75.93 | 76.54 | 82.25 | 75.39 | 79.94 | 71.99 | 85.43 | 77.18 | |
| TKRTS-L | 75.77 | 74.77 | 71.30 | 70.03 | 84.29 | ||||
| TKRTS-E | 75.31 | 73.77 | 82.10 | 70.91 | 79.17 | 68.03 | 83.64 | 75.53 | |
表2 MTS策略下的平均分类准确率对比 ( %)
Tab. 2 Comparison of average classification accuracy under MTS strategy
| 算法 | 平均分类准确率 | 均值 | |||||||
|---|---|---|---|---|---|---|---|---|---|
| MI2a-1 | MI2a-2 | MI2a-3 | MI2a-4 | MI2a-5 | MI2a-6 | MI2b | MI4a | ||
| CSP-TJM | 57.18 | 57.64 | 61.81 | 57.87 | 60.88 | 56.79 | 67.00 | 57.36 | 59.57 | 
| CSP-JDA | 61.37 | 62.35 | 54.58 | 57.41 | 57.53 | 51.25 | 58.28 | 60.80 | 57.95 | 
| CSP-LDA | 68.90 | 67.90 | 66.90 | 65.05 | 67.67 | 57.79 | 66.73 | 64.00 | 65.62 | 
| EA-CSP-LDA | 73.69 | 70.99 | 80.25 | 70.99 | 77.16 | 68.52 | 68.34 | 78.29 | 73.53 | 
| MDM | 59.62 | 59.83 | 55.24 | 56.08 | 53.88 | 52.58 | 58.14 | 60.69 | 57.01 | 
| RA-MDM | 72.07 | 72.99 | 79.48 | 69.21 | 77.01 | 66.28 | 69.69 | 77.07 | 72.98 | 
| MEKT | 76.31 | 73.46 | 81.10 | 80.86 | 69.98 | 69.47 | 76.37 | ||
| METL | 76.00 | — | — | — | — | — | — | — | — | 
| SB-TA-CSP | 75.15 | — | — | — | — | — | — | — | — | 
| TKCSP | — | — | — | — | — | — | — | 81.14 | — | 
| FWR-JPDA | 75.69 | 80.56 | 74.07 | 78.47 | 70.06 | — | — | — | |
| MMDA | 77.93 | — | — | — | — | — | — | 83.00 | — | 
| EA-CSP-JDA | 76.70 | 70.68 | 79.17 | 69.75 | 76.16 | 66.36 | 69.56 | 78.14 | 73.31 | 
| TKRTS-R | 75.93 | 76.54 | 82.25 | 75.39 | 79.94 | 71.99 | 85.43 | 77.18 | |
| TKRTS-L | 75.77 | 74.77 | 71.30 | 70.03 | 84.29 | ||||
| TKRTS-E | 75.31 | 73.77 | 82.10 | 70.91 | 79.17 | 68.03 | 83.64 | 75.53 | |
| 算法 | 平均分类准确率 | 均值 | |||||||
|---|---|---|---|---|---|---|---|---|---|
| MI2a-1 | MI2a-2 | MI2a-3 | MI2a-4 | MI2a-5 | MI2a-6 | MI2b | MI4a | ||
| CSP-TJM | 55.23 | 55.40 | 56.59 | 55.30 | 54.93 | 51.90 | 59.95 | 57.52 | 55.85 | 
| CSP-JDA | 58.35 | 56.48 | 58.96 | 54.21 | 56.71 | 53.29 | 58.89 | 60.73 | 57.20 | 
| CSP-LDA | 59.28 | 55.64 | 59.89 | 57.06 | 58.40 | 54.31 | 60.81 | 55.68 | 57.63 | 
| EA-CSP-LDA | 64.57 | 61.27 | 70.18 | 59.34 | 64.91 | 58.73 | 67.39 | 64.29 | |
| MDM | 56.28 | 54.21 | 56.87 | 55.11 | 54.83 | 52.25 | 59.68 | 55.02 | 55.53 | 
| RA-MDM | 66.60 | 69.76 | 76.78 | 65.44 | 71.90 | 59.74 | 67.28 | 75.13 | 69.08 | 
| MEKT | 68.73 | 64.01 | 71.41 | 64.90 | 69.80 | 60.11 | 66.04 | 80.34 | 68.17 | 
| METL | 69.06 | — | — | — | — | — | — | — | — | 
| SB-TA-CSP | 68.76 | — | — | — | — | — | — | — | — | 
| FWR-JPDA | 67.48 | 65.05 | 73.18 | 60.88 | 69.48 | 65.67 | — | — | — | 
| MMDA | 69.17 | — | — | — | — | — | — | 77.21 | — | 
| EA-CSP-JDA | 65.99 | 62.58 | 71.51 | 61.12 | 66.11 | 59.16 | 68.02 | 71.14 | 65.70 | 
| TKRTS-R | 69.73 | 73.49 | 65.18 | 70.96 | 66.78 | ||||
| TKRTS-L | 65.87 | 61.00 | 66.63 | 79.45 | 69.21 | ||||
| TKRTS-E | 68.73 | 65.52 | 73.02 | 63.65 | 70.75 | 60.65 | 66.10 | 78.71 | 68.39 | 
表3 STS策略下的平均分类准确率对比 ( %)
Tab. 3 Comparison of average classification accuracy under STS strategy
| 算法 | 平均分类准确率 | 均值 | |||||||
|---|---|---|---|---|---|---|---|---|---|
| MI2a-1 | MI2a-2 | MI2a-3 | MI2a-4 | MI2a-5 | MI2a-6 | MI2b | MI4a | ||
| CSP-TJM | 55.23 | 55.40 | 56.59 | 55.30 | 54.93 | 51.90 | 59.95 | 57.52 | 55.85 | 
| CSP-JDA | 58.35 | 56.48 | 58.96 | 54.21 | 56.71 | 53.29 | 58.89 | 60.73 | 57.20 | 
| CSP-LDA | 59.28 | 55.64 | 59.89 | 57.06 | 58.40 | 54.31 | 60.81 | 55.68 | 57.63 | 
| EA-CSP-LDA | 64.57 | 61.27 | 70.18 | 59.34 | 64.91 | 58.73 | 67.39 | 64.29 | |
| MDM | 56.28 | 54.21 | 56.87 | 55.11 | 54.83 | 52.25 | 59.68 | 55.02 | 55.53 | 
| RA-MDM | 66.60 | 69.76 | 76.78 | 65.44 | 71.90 | 59.74 | 67.28 | 75.13 | 69.08 | 
| MEKT | 68.73 | 64.01 | 71.41 | 64.90 | 69.80 | 60.11 | 66.04 | 80.34 | 68.17 | 
| METL | 69.06 | — | — | — | — | — | — | — | — | 
| SB-TA-CSP | 68.76 | — | — | — | — | — | — | — | — | 
| FWR-JPDA | 67.48 | 65.05 | 73.18 | 60.88 | 69.48 | 65.67 | — | — | — | 
| MMDA | 69.17 | — | — | — | — | — | — | 77.21 | — | 
| EA-CSP-JDA | 65.99 | 62.58 | 71.51 | 61.12 | 66.11 | 59.16 | 68.02 | 71.14 | 65.70 | 
| TKRTS-R | 69.73 | 73.49 | 65.18 | 70.96 | 66.78 | ||||
| TKRTS-L | 65.87 | 61.00 | 66.63 | 79.45 | 69.21 | ||||
| TKRTS-E | 68.73 | 65.52 | 73.02 | 63.65 | 70.75 | 60.65 | 66.10 | 78.71 | 68.39 | 
| 策略 | 算法 | MI2a-1 | MI2a-2 | MI2a-3 | MI2a-4 | MI2a-5 | MI2a-6 | MI2b | MI4a | 
|---|---|---|---|---|---|---|---|---|---|
| MTS | CSP-LDA | 0.65 | 0.66 | 0.66 | 0.65 | 0.65 | 0.66 | 0.49 | 0.48 | 
| EA-CSP-LDA | 0.77 | 0.71 | 0.79 | 0.77 | 0.76 | 0.68 | 0.50 | 0.50 | |
| MEKT | 0.86 | 0.90 | 0.87 | 0.87 | 0.87 | 0.85 | 1.24 | 1.39 | |
| TKRTS | 0.67 | 0.59 | 0.52 | 0.56 | 0.65 | 0.69 | 1.75 | 1.62 | |
| STS | CSP-LDA | 0.21 | 0.21 | 0.21 | 0.21 | 0.21 | 0.21 | 0.12 | 0.69 | 
| EA-CSP-LDA | 0.23 | 0.23 | 0.23 | 0.23 | 0.23 | 0.23 | 0.15 | 0.96 | |
| MEKT | 0.43 | 0.43 | 0.43 | 0.43 | 0.42 | 0.42 | 0.20 | 1.17 | |
| TKRTS | 0.31 | 0.30 | 0.37 | 0.41 | 0.33 | 0.34 | 0.35 | 1.1 | 
表4 算法执行效率对比 ( s)
Tab. 4 Algorithm execution efficiency comparison
| 策略 | 算法 | MI2a-1 | MI2a-2 | MI2a-3 | MI2a-4 | MI2a-5 | MI2a-6 | MI2b | MI4a | 
|---|---|---|---|---|---|---|---|---|---|
| MTS | CSP-LDA | 0.65 | 0.66 | 0.66 | 0.65 | 0.65 | 0.66 | 0.49 | 0.48 | 
| EA-CSP-LDA | 0.77 | 0.71 | 0.79 | 0.77 | 0.76 | 0.68 | 0.50 | 0.50 | |
| MEKT | 0.86 | 0.90 | 0.87 | 0.87 | 0.87 | 0.85 | 1.24 | 1.39 | |
| TKRTS | 0.67 | 0.59 | 0.52 | 0.56 | 0.65 | 0.69 | 1.75 | 1.62 | |
| STS | CSP-LDA | 0.21 | 0.21 | 0.21 | 0.21 | 0.21 | 0.21 | 0.12 | 0.69 | 
| EA-CSP-LDA | 0.23 | 0.23 | 0.23 | 0.23 | 0.23 | 0.23 | 0.15 | 0.96 | |
| MEKT | 0.43 | 0.43 | 0.43 | 0.43 | 0.42 | 0.42 | 0.20 | 1.17 | |
| TKRTS | 0.31 | 0.30 | 0.37 | 0.41 | 0.33 | 0.34 | 0.35 | 1.1 | 
| 算法 | 准确率 | 算法 | 准确率 | 算法 | 准确率 | 
|---|---|---|---|---|---|
| MI-CNN[ | 60.69 | ConvNet[ | 72.53 | CNN-TKL | 59.19 | 
| C2CM[ | 74.46 | DRDA[ | 74.70 | TKRTS | 76.75 | 
表5 MTS策略下TKRTS与深度学习方法在数据集MI2a上的分类准确率对比 ( %)
Tab. 5 Comparison of classification accuracy between TKRTS and deep learning methods on dataset MI2a under MTS strategy
| 算法 | 准确率 | 算法 | 准确率 | 算法 | 准确率 | 
|---|---|---|---|---|---|
| MI-CNN[ | 60.69 | ConvNet[ | 72.53 | CNN-TKL | 59.19 | 
| C2CM[ | 74.46 | DRDA[ | 74.70 | TKRTS | 76.75 | 
| 策略 | 算法 | 平均准确率 | 均值 | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| MI2a-1 | MI2a-2 | MI2a-3 | MI2a-4 | MI2a-5 | MI2a-6 | MI2b | MI4a | |||
| MTS | RTS-TKL | 66.98 | 59.41 | 62.35 | 66.13 | 65.28 | 57.64 | 65.39 | 66.29 | 63.68 | 
| RA-CSP-TKL | 70.52 | 76.62 | 65.43 | |||||||
| RA-RTS-KSVM | 70.06 | 78.32 | 70.14 | 68.87 | 74.79 | 72.40 | ||||
| RA-RTS-TKL | 75.93 | 76.54 | 82.25 | 75.39 | 79.94 | 71.99 | 69.94 | 85.43 | 77.69 | |
| 策略 | 算法 | 平均准确率 | 均值 | |||||||
| MI2-1 | MI2-2 | MI2-3 | MI2-4 | MI2-5 | MI2-6 | MI3 | MI4 | |||
| STS | RTS-TKL | 64.14 | 56.91 | 56.79 | 59.36 | 57.17 | 53.68 | 62.31 | 68.25 | 59.83 | 
| RA-CSP-TKL | 62.37 | 73.32 | 60.80 | 67.24 | ||||||
| RA-RTS-KSVM | 64.27 | 74.06 | 68.15 | 57.25 | 69.86 | 65.97 | ||||
| RA-RTS-TKL | 69.73 | 65.99 | 65.18 | 70.96 | 61.01 | 66.78 | 80.29 | 69.18 | ||
表6 样本对齐、RTS特征与TKL方法的平均准确率对比 ( %)
Tab. 6 Average accuracy comparison among sample alignment, RTS feature and TKL methods
| 策略 | 算法 | 平均准确率 | 均值 | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| MI2a-1 | MI2a-2 | MI2a-3 | MI2a-4 | MI2a-5 | MI2a-6 | MI2b | MI4a | |||
| MTS | RTS-TKL | 66.98 | 59.41 | 62.35 | 66.13 | 65.28 | 57.64 | 65.39 | 66.29 | 63.68 | 
| RA-CSP-TKL | 70.52 | 76.62 | 65.43 | |||||||
| RA-RTS-KSVM | 70.06 | 78.32 | 70.14 | 68.87 | 74.79 | 72.40 | ||||
| RA-RTS-TKL | 75.93 | 76.54 | 82.25 | 75.39 | 79.94 | 71.99 | 69.94 | 85.43 | 77.69 | |
| 策略 | 算法 | 平均准确率 | 均值 | |||||||
| MI2-1 | MI2-2 | MI2-3 | MI2-4 | MI2-5 | MI2-6 | MI3 | MI4 | |||
| STS | RTS-TKL | 64.14 | 56.91 | 56.79 | 59.36 | 57.17 | 53.68 | 62.31 | 68.25 | 59.83 | 
| RA-CSP-TKL | 62.37 | 73.32 | 60.80 | 67.24 | ||||||
| RA-RTS-KSVM | 64.27 | 74.06 | 68.15 | 57.25 | 69.86 | 65.97 | ||||
| RA-RTS-TKL | 69.73 | 65.99 | 65.18 | 70.96 | 61.01 | 66.78 | 80.29 | 69.18 | ||
| 算法 | 平均准确率 | 均值 | ||
|---|---|---|---|---|
| MI2a | MI2b | MI4a | ||
| CSP-TKL | 63.71 | 65.36 | 64.50 | 64.52 | 
| RA-CSP-TKL | 72.31 | 69.44 | 77.93 | 73.23 | 
| RCSP-TKL | 65.20 | 65.42 | 60.71 | 63.78 | 
| RA-RCSP-TKL | 71.75 | 69.44 | 78.07 | 73.09 | 
| CNN-TKL | 59.19 | — | — | |
| RTS-TKL | 62.97 | 65.39 | 66.29 | 64.88 | 
| RA-RTS-TKL | 77.01 | 69.94 | 85.43 | 77.46 | 
表7 MTS策略下不同特征提取算法的平均准确率对比 ( %)
Tab. 7 Average accuracy comparison of different feature extraction algorithms under MTS strategy
| 算法 | 平均准确率 | 均值 | ||
|---|---|---|---|---|
| MI2a | MI2b | MI4a | ||
| CSP-TKL | 63.71 | 65.36 | 64.50 | 64.52 | 
| RA-CSP-TKL | 72.31 | 69.44 | 77.93 | 73.23 | 
| RCSP-TKL | 65.20 | 65.42 | 60.71 | 63.78 | 
| RA-RCSP-TKL | 71.75 | 69.44 | 78.07 | 73.09 | 
| CNN-TKL | 59.19 | — | — | |
| RTS-TKL | 62.97 | 65.39 | 66.29 | 64.88 | 
| RA-RTS-TKL | 77.01 | 69.94 | 85.43 | 77.46 | 
| 1 | 肖晓琳,辛风然,梅杰,等.自适应脑机接口研究综述[J].电子与信息学报,2023,45(7):2386-2394. | 
| XIAO X L, XIN F R, MEI J, et al. A review of adaptive brain-computer interface research[J]. Journal of Electronics and Information Technology, 2023, 45(7): 2386-2394. | |
| 2 | ALTAHERI H, MUHAMMAD G, ALSULAIMAN M, et al. Deep learning techniques for classification of ElectroEncephaloGram (EEG) Motor Imagery (MI) signals: a review[J]. Neural Computing and Applications, 2023, 35(20): 14681-14722. | 
| 3 | 赵欣,陈志堂,王坤,等.运动想象脑-机接口新进展与发展趋势[J].中国生物医学工程学报,2019,38(1):84-93. | 
| ZHAO X, CHEN Z T, WANG K, et al. New developments and trends of BCI based on motor imagery[J]. Chinese Journal of Biomedical Engineering, 2019, 38(1): 84-93. | |
| 4 | PFURTSCHELLER G. EEG event-related desynchronization (ERD) and synchronization (ERS)[J]. Electroencephalography and Clinical Neurophysiology, 1997, 1(103): No.26. | 
| 5 | 潘林聪,王坤,许敏鹏,等.面向运动意图解码的共空间模式及其扩展算法研究综述[J].中国生物医学工程学报,2022,41(5):577-588. | 
| PAN L C, WANG K, XU M P, et al. Review of researches on common spatial pattern and its extended algorithms for movement intention decoding[J]. Chinese Journal of Biomedical Engineering, 2022, 41(5): 577-588. | |
| 6 | 何秋妍,邓明华.通用域适应综述[J].计算机研究与发展,2024,61(1):120-144. | 
| HE Q Y, DENG M H. Survey of universal domain adaptation[J]. Journal of Computer Research and Development, 2024, 61(1): 120-144. | |
| 7 | QIN X, WANG J, CHEN Y, et al. Domain generalization for activity recognition via adaptive feature fusion[J]. ACM Transactions on Intelligent Systems and Technology, 2023, 14(1): No.9. | 
| 8 | ZANINI P, CONGEDO M, JUTTEN C, et al. Transfer learning: a Riemannian geometry framework with applications to brain-computer interfaces[J]. IEEE Transactions on Biomedical Engineering, 2018, 65(5): 1107-1116. | 
| 9 | HE H, WU D. Transfer learning for brain-computer interfaces: a Euclidean space data alignment approach[J]. IEEE Transactions on Biomedical Engineering, 2020, 67(2): 399-410. | 
| 10 | ZHANG X, SHE Q, CHEN Y, et al. Sub-band target alignment common spatial pattern in brain-computer interface[J]. Computer Methods and Programs in Biomedicine, 2021, 207: No.106150. | 
| 11 | LONG M, WANG J, DING G, et al. Transfer feature learning with joint distribution adaptation[C]// Proceedings of the 2013 IEEE International Conference on Computer Vision. Piscataway: IEEE, 2013: 2200-2207. | 
| 12 | WANG J, CHEN Y, HAO S, et al. Balanced distribution adaptation for transfer learning[C]// Proceedings of the 2017 IEEE International Conference on Data Mining. Piscataway: IEEE, 2017: 1129-1134. | 
| 13 | ZHANG W, WU D. Discriminative Joint Probability Maximum Mean Discrepancy (DJP-MMD) for domain adaptation[C]// Proceedings of the 2020 International Joint Conference on Neural Networks. Piscataway: IEEE, 2020: 1-8. | 
| 14 | LONG M, WANG J, DING G, et al. Transfer joint matching for unsupervised domain adaptation[C]// Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2014: 1410-1417. | 
| 15 | ZHANG W, WU D. Manifold embedded knowledge transfer for brain-computer interfaces[J]. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2020, 28(5): 1117-1127. | 
| 16 | CAI Y, SHE Q, JI J, et al. Motor imagery EEG decoding using manifold embedded transfer learning[J]. Journal of Neuroscience Methods, 2022, 370: No.109489. | 
| 17 | LUO T J. Dual regularized feature extraction and adaptation for cross-subject motor imagery EEG classification[C]// Proceedings of the 2022 IEEE International Conference on Bioinformatics and Biomedicine. Piscataway: IEEE, 2022: 1092-1099. | 
| 18 | GAO Y, LIU Y, SHE Q, et al. Domain adaptive algorithm on multi-manifold embedded distributed alignment for brain-computer interfaces[J]. IEEE Journal of Biomedical and Health Informatics, 2023, 27(1): 296-307. | 
| 19 | ZHAO H, ZHENG Q, MA K, et al. Deep representation-based domain adaptation for nonstationary EEG classification[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021, 32(2): 535-545. | 
| 20 | HONG X, ZHENG Q, LIU L, et al. Dynamic joint domain adaptation network for motor imagery classification[J]. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2021, 29: 556-565. | 
| 21 | SHE Q, CHEN T, FANG F, et al. Improved domain adaptation network based on Wasserstein distance for motor imagery EEG classification[J]. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2023, 31: 1137-1148. | 
| 22 | LONG M, WANG J, SUN J, et al. Domain invariant transfer kernel learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2015, 27(6): 1519-1532. | 
| 23 | DAI M, ZHENG D, LIU S, et al. Transfer kernel common spatial patterns for motor imagery brain-computer interface classification[J]. Computational and Mathematical Methods in Medicine, 2018, 2018: No.9871603. | 
| 24 | WILLIAMS C K I, SEEGER M. Using the Nyström method to speed up kernel machines[C]// Proceedings of the 13th International Conference on Neural Information Processing Systems. Cambridge: MIT Press, 2000: 661-667. | 
| 25 | ZHANG T, ANDO R K. Analysis of spectral kernel design based semi-supervised learning[C]// Proceedings of the 18th International Conference on Neural Information Processing Systems. Cambridge: MIT Press, 2005: 1601-1608. | 
| 26 | JIN R, YANG T, MAHDAVI M, et al. Improved bounds for the Nyström method with application to kernel classification[J]. IEEE Transactions on Information Theory, 2013, 59(10): 6939-6949. | 
| 27 | 韦泓妤,陈黎飞,罗天健. 运动想象脑电信号的跨域特征学习方法[J]. 计算机应用研究, 2022, 39(8):2340-2346, 2351. | 
| WEI H Y, CHEN L F, LUO T J. Cross-domain feature learning method for motor imagery EEG signals[J]. Application Research of Computers, 2022, 39(8): 2340-2346, 2351. | |
| 28 | DOSE H, MØLLER J S, IVERSEN H K, et al. An end-to-end deep learning approach to MI-EEG signal classification for BCIs[J]. Expert Systems with Applications, 2018, 114: 532-542. | 
| 29 | SAKHAVI S, GUAN C, YAN S. Learning temporal information for brain-computer interface using convolutional neural networks[J]. IEEE Transactions on Neural Networks and Learning Systems, 2018; 29(11):5619-5629. | 
| 30 | SCHIRRMEISTER R T, SPRINGENBERG J T, FIEDERER L D J, et al. Deep learning with convolutional neural networks for EEG decoding and visualization[J]. Human Brain Mapping, 2017, 38(11): 5391-5420. | 
| 31 | LOTTE F, GUAN C. Regularizing common spatial patterns to improve BCI designs: unified theory and new algorithms[J]. IEEE Transactions on Biomedical Engineering, 2011, 58(2): 355-362. | 
| 32 | MISHUHINA V, JIANG X. Complex common spatial patterns on time-frequency decomposed EEG for brain-computer interface[J]. Pattern Recognition, 2021, 115: No.107918. | 
| 33 | LUO T J. Parallel genetic algorithm based common spatial patterns selection on time-frequency decomposed EEG signals for motor imagery brain-computer interface[J]. Biomedical Signal Processing and Control, 2023, 80(Pt 1): No.104397. | 
| [1] | 晁浩, 封舒琪, 刘永利. 脑电情感识别中多上下文向量优化的卷积递归神经网络[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2041-2046. | 
| [2] | 时旺军, 王晶, 宁晓军, 林友芳. 小样本场景下的元迁移学习睡眠分期模型[J]. 《计算机应用》唯一官方网站, 2024, 44(5): 1445-1451. | 
| [3] | 曹铉, 罗天健. 运动想象脑电信号的跨被试动态多域对抗学习方法[J]. 《计算机应用》唯一官方网站, 2024, 44(2): 645-653. | 
| [4] | 杨淑莹, 国海铭, 李欣. 基于通道选择和多维特征融合的脑电信号分类[J]. 《计算机应用》唯一官方网站, 2023, 43(11): 3418-3427. | 
| [5] | 张军鹏, 施玉杰, 蒋睿, 董静静, 邱昌建. 基于脑电信号的认知功能障碍识别与分类进展综述[J]. 《计算机应用》唯一官方网站, 2023, 43(10): 3297-3308. | 
| [6] | 秦静, 孙法莉, HUI Fang, 汪祖民, 高兵, 季长清. 可穿戴脑电图设备关键技术及其应用综述[J]. 《计算机应用》唯一官方网站, 2022, 42(4): 1029-1035. | 
| [7] | 彭禹, 宋耀莲, 杨俊. 基于数据增强的运动想象脑电分类[J]. 《计算机应用》唯一官方网站, 2022, 42(11): 3625-3632. | 
| [8] | 霍首君, 郝琰, 石慧宇, 董艳清, 曹锐. 基于深度卷积网络的运动想象脑电信号模式识别[J]. 计算机应用, 2021, 41(4): 1042-1048. | 
| [9] | 罗飞, 刘鹏飞, 罗元, 朱思蒙. 多特征融合的运动想象脑电特征提取方法[J]. 《计算机应用》唯一官方网站, 2020, 40(2): 616-620. | 
| [10] | 胡章芳, 张力, 黄丽嘉, 罗元. 基于时频域的卷积神经网络运动想象脑电信号识别方法[J]. 计算机应用, 2019, 39(8): 2480-2483. | 
| [11] | 胡敏, 李冲, 路荣荣, 黄宏程. 基于三维视觉指导的运动想象训练性能分析[J]. 计算机应用, 2018, 38(3): 836-841. | 
| [12] | 王若凡, 刘静, 王江, 于海涛, 曹亦宾. 基于功率谱及有限穿越可视图的癫痫脑电信号分析算法[J]. 计算机应用, 2017, 37(1): 175-182. | 
| [13] | 韩敏, 孙卓然. 基于小波变换和AdaBoost极限学习机的癫痫脑电信号分类[J]. 计算机应用, 2015, 35(9): 2701-2705. | 
| [14] | 艾玲梅 李营. HHT方法在不同思维作业脑电信号分析中的应用[J]. 计算机应用, 2008, 28(12): 3089-3091. | 
| 阅读次数 | ||||||
| 全文 |  | |||||
| 摘要 |  | |||||