《计算机应用》唯一官方网站 ›› 2023, Vol. 43 ›› Issue (11): 3418-3427.DOI: 10.11772/j.issn.1001-9081.2022101590
所属专题: 人工智能
收稿日期:
2022-10-24
修回日期:
2023-01-31
接受日期:
2023-01-31
发布日期:
2023-04-12
出版日期:
2023-11-10
通讯作者:
杨淑莹
作者简介:
杨淑莹(1964—),女,四川成都人,教授,博士,主要研究方向:模式识别、智能机器人 yangshuying@email.tjut.edu.cn基金资助:
Shuying YANG(), Haiming GUO, Xin LI
Received:
2022-10-24
Revised:
2023-01-31
Accepted:
2023-01-31
Online:
2023-04-12
Published:
2023-11-10
Contact:
Shuying YANG
About author:
YANG Shuying, born in 1964, Ph. D., professor. Her research interests include pattern recognition, intelligent robots.Supported by:
摘要:
针对多通道脑电信号(EEG)相互干扰、存在个体差异性导致分类结果不同和单域特征识别率低等问题,提出一种通道选择和特征融合的方法。首先,对获取到的EEG进行预处理,使用梯度提升决策树(GBDT)选出重要通道;其次,采用广义预测控制(GPC)模型构建重要通道的预测信号,辨析多维相关信号之间的细微差别,再使用SE?TCNTA(Squeeze and Excitation block-Temporal Convolutional Network-Temporal Attention)模型提取不同帧之间的时序特征;然后,使用皮尔逊相关系数计算通道间的关系,提取EEG的频域特征和预测信号的控制量作为输入,建立空间图结构,并采用图卷积网络(GCN)提取频域、空域的特征;最后,将上述二者特征输入全连接层进行特征融合,实现EEG的分类。在公共数据集BCICIV_2a上的实验结果表明,在进行通道选择的情况下,与首个用于ERP检测的EEG-Inception模型以及同样采用双分支提取特征的DSCNN (Shallow Double-branch Convolutional Neural Network)模型方法相比,所提方法的分类准确率分别提升了1.47%和1.69%,Kappa值分别提升了1.25%和2.53%。所提方法能够提高EGG的分类精度,同时减少冗余数据对特征提取的影响,因此更适用于脑机接口(BCI)系统。
中图分类号:
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, 2023, 43(11): 3418-3427.
Shuying YANG, Haiming GUO, Xin LI. EEG classification based on channel selection and multi-dimensional feature fusion[J]. Journal of Computer Applications, 2023, 43(11): 3418-3427.
通道数 | 不同受试者的准确率 | 平均值 | ||||||||
---|---|---|---|---|---|---|---|---|---|---|
A01 | A02 | A03 | A04 | A05 | A06 | A07 | A08 | A09 | ||
1 | 80.69 | 75.38 | 80.29 | 75.68 | 76.49 | 83.87 | 82.56 | 81.42 | 85.96 | 80.26 |
2 | 83.62 | 79.84 | 78.88 | 72.28 | 80.26 | 83.26 | 84.54 | 86.35 | 80.72 | 81.08 |
3 | 88.31 | 76.24 | 81.85 | 74.56 | 78.54 | 84.32 | 83.30 | 84.26 | 82.88 | 81.58 |
4 | 85.60 | 73.20 | 81.59 | 80.54 | 84.48 | 83.50 | 81.58 | 84.32 | 85.88 | 82.29 |
5 | 88.27 | 79.59 | 87.70 | 85.68 | 84.39 | 83.16 | 89.49 | 89.56 | 86.79 | 86.07 |
6 | 85.89 | 73.59 | 83.29 | 86.12 | 82.99 | 83.43 | 84.29 | 85.51 | 85.49 | 83.40 |
︙ | ︙ | ︙ | ︙ | ︙ | ︙ | ︙ | ︙ | ︙ | ︙ | ︙ |
22 | 78.26 | 71.28 | 70.94 | 78.26 | 72.40 | 76.34 | 80.76 | 79.84 | 81.46 | 76.61 |
表1 不同通道数的分类结果 ( %)
Tab. 1 Classification results of different channel numbers
通道数 | 不同受试者的准确率 | 平均值 | ||||||||
---|---|---|---|---|---|---|---|---|---|---|
A01 | A02 | A03 | A04 | A05 | A06 | A07 | A08 | A09 | ||
1 | 80.69 | 75.38 | 80.29 | 75.68 | 76.49 | 83.87 | 82.56 | 81.42 | 85.96 | 80.26 |
2 | 83.62 | 79.84 | 78.88 | 72.28 | 80.26 | 83.26 | 84.54 | 86.35 | 80.72 | 81.08 |
3 | 88.31 | 76.24 | 81.85 | 74.56 | 78.54 | 84.32 | 83.30 | 84.26 | 82.88 | 81.58 |
4 | 85.60 | 73.20 | 81.59 | 80.54 | 84.48 | 83.50 | 81.58 | 84.32 | 85.88 | 82.29 |
5 | 88.27 | 79.59 | 87.70 | 85.68 | 84.39 | 83.16 | 89.49 | 89.56 | 86.79 | 86.07 |
6 | 85.89 | 73.59 | 83.29 | 86.12 | 82.99 | 83.43 | 84.29 | 85.51 | 85.49 | 83.40 |
︙ | ︙ | ︙ | ︙ | ︙ | ︙ | ︙ | ︙ | ︙ | ︙ | ︙ |
22 | 78.26 | 71.28 | 70.94 | 78.26 | 72.40 | 76.34 | 80.76 | 79.84 | 81.46 | 76.61 |
模型 | 网络层 | 神经元数 | 输出 | 激活函数 |
---|---|---|---|---|
SE-TCNTA | Input | (None, 11, 25) | ||
SE-Block | (None, 11, 25) | ReLU hard_Sigmoid | ||
TCN1 | 12 | (None, 11, 12) | ||
Dropout | Rate=0.8 | |||
TCN2 | 6 | (None, 11, 6) | ||
Dropout | Rate=0.6 | |||
Permute | (None, 6, 11) | |||
Dense | 11 | (None, 6, 11) | Softmax | |
Permute | (None, 6, 11) | |||
Multiply | (None, 11, 6) | |||
Flatten | (None, 66) | |||
GCN | Input-Data | (None, None, 280) | ||
Input-Edge | (None, None, 5) | |||
GraphConv1 | 16 | (None, 5, 16) | ReLU | |
GraphConv2 | 8 | (None, 5, 8) | ReLU | |
Flatten | (None, 40) | |||
Dense | 32 | (None, 64) | ReLU | |
Dense | 4 | (None, 4) | Softmax |
表2 网络模型结构参数
Tab. 2 Structure parameters of network models
模型 | 网络层 | 神经元数 | 输出 | 激活函数 |
---|---|---|---|---|
SE-TCNTA | Input | (None, 11, 25) | ||
SE-Block | (None, 11, 25) | ReLU hard_Sigmoid | ||
TCN1 | 12 | (None, 11, 12) | ||
Dropout | Rate=0.8 | |||
TCN2 | 6 | (None, 11, 6) | ||
Dropout | Rate=0.6 | |||
Permute | (None, 6, 11) | |||
Dense | 11 | (None, 6, 11) | Softmax | |
Permute | (None, 6, 11) | |||
Multiply | (None, 11, 6) | |||
Flatten | (None, 66) | |||
GCN | Input-Data | (None, None, 280) | ||
Input-Edge | (None, None, 5) | |||
GraphConv1 | 16 | (None, 5, 16) | ReLU | |
GraphConv2 | 8 | (None, 5, 8) | ReLU | |
Flatten | (None, 40) | |||
Dense | 32 | (None, 64) | ReLU | |
Dense | 4 | (None, 4) | Softmax |
算法 | 不同受试者的准确率 | 平均值 | ||||||||
---|---|---|---|---|---|---|---|---|---|---|
A01 | A02 | A03 | A04 | A05 | A06 | A07 | A08 | A09 | ||
GPC-SE-TCNTA | 85.27 | 72.29 | 87.42 | 80.24 | 80.39 | 82.59 | 88.27 | 87.69 | 85.46 | 83.30 |
GCN | 88.59 | 78.83 | 86.28 | 83.64 | 83.72 | 80.89 | 86.75 | 86.49 | 83.19 | 84.26 |
特征融合 | 88.27 | 79.59 | 87.70 | 85.68 | 84.39 | 83.16 | 89.49 | 89.56 | 86.79 | 86.07 |
表3 单一算法与特征融合算法的比较 ( %)
Tab. 3 Comparison of single algorithms and feature fusion algorithm
算法 | 不同受试者的准确率 | 平均值 | ||||||||
---|---|---|---|---|---|---|---|---|---|---|
A01 | A02 | A03 | A04 | A05 | A06 | A07 | A08 | A09 | ||
GPC-SE-TCNTA | 85.27 | 72.29 | 87.42 | 80.24 | 80.39 | 82.59 | 88.27 | 87.69 | 85.46 | 83.30 |
GCN | 88.59 | 78.83 | 86.28 | 83.64 | 83.72 | 80.89 | 86.75 | 86.49 | 83.19 | 84.26 |
特征融合 | 88.27 | 79.59 | 87.70 | 85.68 | 84.39 | 83.16 | 89.49 | 89.56 | 86.79 | 86.07 |
方法 | 不同受试者的准确率 | 平均值 | ||||||||
---|---|---|---|---|---|---|---|---|---|---|
A01 | A02 | A03 | A04 | A05 | A06 | A07 | A08 | A09 | ||
本文方法 | 88.27 | 79.59 | 87.70 | 85.68 | 84.39 | 83.16 | 89.49 | 89.56 | 86.79 | 86.07 |
AMSI-EEGNet[ | 84.03 | 55.21 | 89.58 | 69.79 | 66.32 | 61.46 | 94.10 | 85.07 | 80.90 | 76.27 |
M3DCNN[ | 77.38 | 60.14 | 82.93 | 72.29 | 75.84 | 68.99 | 76.04 | 76.86 | 84.67 | 75.02 |
FBSF-TSCNN[ | 85.80 | 60.10 | 87.80 | 64.20 | 48.60 | 86.90 | 83.00 | 81.60 | 80.20 | 72.00 |
EEG-inception[ | 81.52 | 78.68 | 94.09 | 80.48 | 79.66 | 76.98 | 91.47 | 91.36 | 89.17 | 84.82 |
MCNN[ | 90.21 | 63.40 | 89.35 | 71.16 | 62.82 | 47.86 | 90.86 | 83.72 | 82.32 | 75.72 |
DSCNN[ | 92.36 | 72.22 | 94.79 | 81.25 | 78.82 | 69.79 | 95.14 | 89.24 | 88.19 | 84.64 |
表4 本文方法与其他文献方法的准确率对比 ( %)
Tab. 4 Accuracy comparison of proposed method and other literature methods
方法 | 不同受试者的准确率 | 平均值 | ||||||||
---|---|---|---|---|---|---|---|---|---|---|
A01 | A02 | A03 | A04 | A05 | A06 | A07 | A08 | A09 | ||
本文方法 | 88.27 | 79.59 | 87.70 | 85.68 | 84.39 | 83.16 | 89.49 | 89.56 | 86.79 | 86.07 |
AMSI-EEGNet[ | 84.03 | 55.21 | 89.58 | 69.79 | 66.32 | 61.46 | 94.10 | 85.07 | 80.90 | 76.27 |
M3DCNN[ | 77.38 | 60.14 | 82.93 | 72.29 | 75.84 | 68.99 | 76.04 | 76.86 | 84.67 | 75.02 |
FBSF-TSCNN[ | 85.80 | 60.10 | 87.80 | 64.20 | 48.60 | 86.90 | 83.00 | 81.60 | 80.20 | 72.00 |
EEG-inception[ | 81.52 | 78.68 | 94.09 | 80.48 | 79.66 | 76.98 | 91.47 | 91.36 | 89.17 | 84.82 |
MCNN[ | 90.21 | 63.40 | 89.35 | 71.16 | 62.82 | 47.86 | 90.86 | 83.72 | 82.32 | 75.72 |
DSCNN[ | 92.36 | 72.22 | 94.79 | 81.25 | 78.82 | 69.79 | 95.14 | 89.24 | 88.19 | 84.64 |
方法 | 不同受试者的Kappa值 | 平均值 | ||||||||
---|---|---|---|---|---|---|---|---|---|---|
A01 | A02 | A03 | A04 | A05 | A06 | A07 | A08 | A09 | ||
本文方法 | 0.90 | 0.66 | 0.88 | 0.78 | 0.81 | 0.76 | 0.86 | 0.85 | 0.87 | 0.81 |
DSCNN[ | 0.90 | 0.63 | 0.93 | 0.75 | 0.72 | 0.60 | 0.94 | 0.86 | 0.84 | 0.79 |
M3DCNN[ | 0.70 | 0.46 | 0.79 | 0.60 | 0.54 | 0.65 | 0.54 | 0.70 | 0.71 | 0.64 |
FBSF-TSCNN[ | 0.81 | 0.47 | 0.84 | 0.52 | 0.32 | 0.43 | 0.77 | 0.76 | 0.74 | 0.63 |
EEG-inception[ | 0.85 | 0.54 | 0.87 | 0.78 | 0.77 | 0.66 | 0.95 | 0.83 | 0.90 | 0.80 |
EEGNet [ | 0.79 | 0.34 | 0.84 | 0.51 | 0.54 | 0.47 | 0.82 | 0.73 | 0.77 | 0.65 |
MCNN [ | 0.87 | 0.51 | 0.86 | 0.62 | 0.50 | 0.30 | 0.88 | 0.78 | 0.76 | 0.68 |
表5 本文方法与其他文献方法的Kappa值对比
Tab. 5 Kappa value comparison of proposed method and other literature methods
方法 | 不同受试者的Kappa值 | 平均值 | ||||||||
---|---|---|---|---|---|---|---|---|---|---|
A01 | A02 | A03 | A04 | A05 | A06 | A07 | A08 | A09 | ||
本文方法 | 0.90 | 0.66 | 0.88 | 0.78 | 0.81 | 0.76 | 0.86 | 0.85 | 0.87 | 0.81 |
DSCNN[ | 0.90 | 0.63 | 0.93 | 0.75 | 0.72 | 0.60 | 0.94 | 0.86 | 0.84 | 0.79 |
M3DCNN[ | 0.70 | 0.46 | 0.79 | 0.60 | 0.54 | 0.65 | 0.54 | 0.70 | 0.71 | 0.64 |
FBSF-TSCNN[ | 0.81 | 0.47 | 0.84 | 0.52 | 0.32 | 0.43 | 0.77 | 0.76 | 0.74 | 0.63 |
EEG-inception[ | 0.85 | 0.54 | 0.87 | 0.78 | 0.77 | 0.66 | 0.95 | 0.83 | 0.90 | 0.80 |
EEGNet [ | 0.79 | 0.34 | 0.84 | 0.51 | 0.54 | 0.47 | 0.82 | 0.73 | 0.77 | 0.65 |
MCNN [ | 0.87 | 0.51 | 0.86 | 0.62 | 0.50 | 0.30 | 0.88 | 0.78 | 0.76 | 0.68 |
1 | DOUIBI K, LE BARS S, LEMONTEY A, et al. Toward EEG-based BCI applications for Industry 4.0: challenges and possible applications[J]. Frontiers in Human Neuroscience, 2021, 15: No.705064. 10.3389/fnhum.2021.705064 |
2 | MINI P P, THOMAS T, GOPIKAKUMARI R, et al. EEG based direct speech BCI system using a fusion of SMRT and MFCC/LPCC features with ANN classifier[J]. Biomedical Signal Processing and Control, 2021, 68: No.102625. 10.1016/j.bspc.2021.102625 |
3 | VIDAURRE C, JORAJURÍA T, RAMOS-MURGUIALDAY A, et al. Improving motor imagery classification during induced motor perturbations[J]. Journal of Neural Engineering, 2021, 18(4): No.0460b1. 10.1088/1741-2552/ac123f |
4 | BAVKAR S, IYER B, DEOSARKAR S, et al. Rapid screening of alcoholism: an EEG based optimal channel selection approach[J]. IEEE Access, 2019, 7:99670-99682. 10.1109/access.2019.2927267 |
5 | LV J, LIU M. Common spatial pattern and particle swarm optimization for channel selection in BCI[C]// Proceedings of the 3rd International Conference on Innovative Computing Information and Control. Piscataway: IEEE, 2008: 457-457. 10.1109/icicic.2008.196 |
6 | ANNABY M H, SAID M H, ELDEIB A M, et al. EEG-based motor imagery classification using digraph Fourier transforms and extreme learning machines[J]. Biomedical Signal Processing and Control, 2021, 69: No.102831. 10.1016/j.bspc.2021.102831 |
7 | RAMOSER H, MÜLLER-GERKING J, PFURTSCHELLER G. Optimal spatial filtering of single trial EEG during imagined hand movement[J]. IEEE Transactions on Rehabilitation Engineering, 2000, 8(4): 441-446. 10.1109/86.895946 |
8 | ZAHID S Z, AQIL M, TUFAIL M, et al. Online classification of multiple motor imagery tasks using filter bank based maximum-a-posteriori common spatial pattern filters[J]. IRBM, 2020, 41(3):141-150. 10.1016/j.irbm.2019.11.002 |
9 | ANG K K, CHIN Z Y, ZHANG H, et al. Filter Bank Common Spatial Pattern (FBCSP) in brain-computer interface[C]// Proceedings of the 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence). Piscataway: IEEE, 2008: 2390-2397. 10.1109/ijcnn.2008.4634130 |
10 | HAN Y, WANG B, LUO J, et al. A classification method for EEG motor imagery signals based on parallel convolutional neural network[J]. Biomedical Signal Processing and Control, 2022, 71(Pt B): No.103190. 10.1016/j.bspc.2021.103190 |
11 | VENKATA PHANIKRISHNA B, CHINARA S. Automatic classification methods for detecting drowsiness using wavelet packet transform extracted time-domain features from single-channel EEG signal[J]. Journal of Neuroscience Methods, 2021, 347: No.108927. 10.1016/j.jneumeth.2020.108927 |
12 | SUN B, ZHAO X, ZHANG H, et al. EEG motor imagery classification with sparse spectrotemporal decomposition and deep learning[J]. IEEE Transactions on Automation Science and Engineering, 2021, 18(2): 541-551. 10.1109/tase.2020.3021456 |
13 | XU M, YAO J, NI H. Dual head and dual attention in deep learning for end-to-end EEG motor imagery classification[J]. Applied Sciences, 2021, 11(22): No.10906. 10.3390/app112210906 |
14 | TORTORA S, GHIDONI S, CHISARI C, et al. Deep learning-based BCI for gait decoding from EEG with LSTM recurrent neural network[J]. Journal of Neural Engineering, 2020, 17(4): No.046011. 10.1088/1741-2552/ab9842 |
15 | FENG N, HU F, WANG H, et al. Motor intention decoding from the upper limb by graph convolutional network based on functional connectivity[J]. International Journal of Neural Systems, 2021, 31(12): No.2150047. 10.1142/s0129065721500477 |
16 | CHANG Q, LI C C, TIAN Q, et al. Classification of first-episode schizophrenia, chronic schizophrenia and healthy control based on brain network of mismatch negativity by graph neural network[J]. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2021, 29:1784-1794. 10.1109/tnsre.2021.3105669 |
17 | LI H, DING M, ZHANG R, et al. Motor imagery EEG classification algorithm based on CNN-LSTM feature fusion network[J]. Biomedical Signal Processing and Control, 2022, 72(Pt A): No.103342. 10.1016/j.bspc.2021.103342 |
18 | 李红利,丁满,张荣华,等. 基于特征融合神经网络的运动想象脑电分类算法[J]. 中国医学物理学杂志, 2022, 39(1):69-75. 10.3969/j.issn.1005-202X.2022.01.012 |
LI H L, DING M, ZHANG R H, et al. Motor imagery EEG classification algorithm based on feature fusion neural network[J]. Chinese Journal of Medical Physics, 2022, 39(1): 69-75. 10.3969/j.issn.1005-202X.2022.01.012 | |
19 | Institute for Knowledge Discovery, Graz University of Technology. BCI Competition IV — data sets 2a[DB/OL]. [2022-05-13]. . |
20 | DEVOS L, MEERT W, DAVIS J. Fast gradient boosting decision trees with bit-level data structures[C]// Proceedings of the 2019 Joint European Conference on Machine Learning and Knowledge Discovery in Databases, LNCS 11906. Cham: Springer, 2020: 590-606. |
21 | MENDOZA-SOTO J L, RODRÍGUEZ CORTÉS H. Generalized predictive control for trajectory tracking of a quadcopter vehicle[C]// Proceedings of the 2017 International Conference on Unmanned Aircraft Systems. Piscataway: IEEE, 2017: 206-212. 10.1109/icuas.2017.7991368 |
22 | ZHANG T, LI C X, LI Z L. Generalized predictive control and delay compensation for high-speed EMU network control system[C]// Proceedings of the 6th International Conference on Computer Science and Network Technology. Piscataway: IEEE, 2017: 511-515. 10.1109/iccsnt.2017.8343751 |
23 | 李红利,尹飞超,张荣华,等. 基于通道注意力和稀疏时频分解的运动想象分类[J]. 生物医学工程学杂志, 2022, 39(3):488-497. 10.7507/1001-5515.202111031 |
LI H L, YIN C F, ZHANG R H, et al. Motor imagery electroencephalogram classification based on sparse spatiotemporal decomposition and channel attention[J]. Journal of Biomedical Engineering, 2022, 39(3): 488-497. 10.7507/1001-5515.202111031 | |
24 | ZHOU Y, ZHENG H, HUANG X, et al. Graph neural networks: taxonomy, advances and trends[J]. ACM Transactions on Intelligent Systems and Technology, 2022, 13(1): No.15. 10.1145/3495161 |
25 | SCARSELLI F, GORI M, TSOI A C, et al. The graph neural network model[J]. IEEE Transactions on Neural Networks, 2009, 20(1): 61-80. 10.1109/tnn.2008.2005605 |
26 | SHI X, LV F, SENG D, et al. Visualizing and understanding graph convolutional network[J]. Multimedia Tools and Applications, 2021, 80(6): 8355-8375. 10.1007/s11042-020-09885-4 |
27 | SALIM A, SUMITRA S. Spectral graph convolutional neural networks in the context of regularization theory[J]. IEEE Transactions on Neural Networks and Learning Systems, 2022(Early Access): 1-12. 10.1109/tnnls.2022.3177742 |
28 | MA W, GONG Y, XUE H, et al. A lightweight and accurate double-branch neural network for four-class motor imagery classification[J]. Biomedical Signal Processing and Control, 2022, 75: No.103582. 10.1016/j.bspc.2022.103582 |
29 | ZHANG C, KIM Y K, ESKANDARIAN A. EEG-inception: an accurate and robust end-to-end neural network for EEG-based motor imagery classification[J]. Journal of Neural Engineering, 2021, 18(4): No.046014. 10.1088/1741-2552/abed81 |
30 | ZHAO X, ZHANG H, ZHU G, et al. A multi-branch 3D convolutional neural network for EEG-based motor imagery classification[J]. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2019, 27(10): 2164-2177. 10.1109/tnsre.2019.2938295 |
31 | CHEN J, YU Z, GU Z, et al. Deep temporal-spatial feature learning for motor imagery-based brain-computer interfaces[J]. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2020, 28(11): 2356-2366. 10.1109/tnsre.2020.3023417 |
32 | RIYAD M, KHALIL M, ADIB A. A novel multi-scale convolutional neural network for motor imagery classification[J]. Biomedical Signal Processing and Control, 2021, 68: No.102747. 10.1016/j.bspc.2021.102747 |
33 | AMIN S U, ALSULAIMAN M, MUHAMMAD G, et al. Deep learning for EEG motor imagery classification based on multi-layer CNNs feature fusion[J]. Future Generation Computer Systems, 2019, 101:542-554. 10.1016/j.future.2019.06.027 |
34 | LAWHERN V J, SOLON A J, WAYTOWICH N R, et al. EEGNet: a compact convolutional neural network for EEG-based brain-computer interfaces[J]. Journal of Neural Engineering, 2018, 15(5): No.056013. 10.1088/1741-2552/aace8c |
[1] | 庞川林, 唐睿, 张睿智, 刘川, 刘佳, 岳士博. D2D通信系统中基于图卷积网络的分布式功率控制算法[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2855-2862. |
[2] | 潘烨新, 杨哲. 基于多级特征双向融合的小目标检测优化模型[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2871-2877. |
[3] | 薛桂香, 王辉, 周卫峰, 刘瑜, 李岩. 基于知识图谱和时空扩散图卷积网络的港口交通流量预测[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2952-2957. |
[4] | 刘禹含, 吉根林, 张红苹. 基于骨架图与混合注意力的视频行人异常检测方法[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2551-2557. |
[5] | 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. |
[6] | 刘瑞华, 郝子赫, 邹洋杨. 基于多层级精细特征融合的步态识别算法[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2250-2257. |
[7] | 李欢欢, 黄添强, 丁雪梅, 罗海峰, 黄丽清. 基于多尺度时空图卷积网络的交通出行需求预测[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2065-2072. |
[8] | 吕锡婷, 赵敬华, 荣海迎, 赵嘉乐. 基于Transformer和关系图卷积网络的信息传播预测模型[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1760-1766. |
[9] | 黄梦源, 常侃, 凌铭阳, 韦新杰, 覃团发. 基于层间引导的低光照图像渐进增强算法[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1911-1919. |
[10] | 黎施彬, 龚俊, 汤圣君. 基于Graph Transformer的半监督异配图表示学习模型[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1816-1823. |
[11] | 刘越, 刘芳, 武奥运, 柴秋月, 王天笑. 基于自注意力机制与图卷积的3D目标检测网络[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1972-1977. |
[12] | 李鸿天, 史鑫昊, 潘卫国, 徐成, 徐冰心, 袁家政. 融合多尺度和注意力机制的小样本目标检测[J]. 《计算机应用》唯一官方网站, 2024, 44(5): 1437-1444. |
[13] | 韩贵金, 张馨渊, 张文涛, 黄娅. 基于多特征融合的自监督图像配准算法[J]. 《计算机应用》唯一官方网站, 2024, 44(5): 1597-1604. |
[14] | 周菊香, 刘金生, 甘健侯, 吴迪, 李子杰. 基于多尺度时序感知网络的课堂语音情感识别方法[J]. 《计算机应用》唯一官方网站, 2024, 44(5): 1636-1643. |
[15] | 李鑫, 孟乔, 皇甫俊逸, 孟令辰. 基于分离式标签协同学习的YOLOv5多属性分类[J]. 《计算机应用》唯一官方网站, 2024, 44(5): 1619-1628. |
阅读次数 | ||||||
全文 |
|
|||||
摘要 |
|
|||||