EEG 数据集整理

news/2024/5/19 18:32:16 标签: EEG, 数据集, Dataset, BCI, 脑电

MI

BCI_II_dataset_Ia_2">BCI II dataset Ia:

任务:确定受试者是试图产生皮质消极性还是皮质积极性

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N26NY268 train; 293 test3.5s
  • Birbaumer, N., Flor, H., Ghanayim, N., Hinterberger, T., Iverson, I., Taub, E., Kotchoubey, B., Kübler, A., & Perelmouter, J, A Brain-Controlled Spelling Device for the Completely Paralyzed, Nature, 398, 297-298.

BCI_II_dataset_Ib_14">BCI II dataset Ib:

任务:确定受试者是试图产生皮质消极性还是皮质积极性

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N27NY200 train; 180 test4.5s
  • Birbaumer, N., Flor, H., Ghanayim, N., Hinterberger, T., Iverson, I., Taub, E., Kotchoubey, B., Kübler, A., & Perelmouter, J, A Brain-Controlled Spelling Device for the Completely Paralyzed, Nature, 398, 297-298.

BCI_II_dataset_IIa_26">BCI II dataset IIa:

任务:提供反馈测试试验的预期目标

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N464NY336 train; 324 test5.6s
  • Wolpaw, J.R., Birbaumer, N., Heetderks, W.J., McFarland, D.J., Peckham, P.H., Schalk, G., Donchin, E., Quatrano, L.A., Robinson, C.J. and Vaughan, T.M.Brain-Computer Interface Technology: A Review of the First International Meeting, IEEE Trans Rehab Eng, 2000, 8(2): 164-173.

BCI_II_dataset_III_38">BCI II dataset III:

任务:对左右运动想象进行分类

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N264NY1409s;静息3s
  • Schlögl A., Neuper C. Pfurtscheller G. (2002) Estimating the mutual information of an EEG-based Brain-Computer-Interface, Biomedizinische Technik 47(1-2): 3-8.

BCI_II_dataset_IV_50">BCI II dataset IV:

任务:在按键前 130 毫秒预测即将到来的手指运动的侧向性(左手与右手)

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N228NY316 train;100 test500ms
  • Benjamin Blankertz, Gabriel Curio and Klaus-Robert Müller, Classifying Single Trial EEG: Towards Brain Computer Interfacing, In: T. G. Diettrich and S. Becker and Z. Ghahramani (eds.), Advances in Neural Inf. Proc. Systems 14 (NIPS 01), 2002.

BCI_III_dataset_I_62">BCI III dataset I:

任务:对来自一个对象的提示运动意象(左小指,舌头)进行分类; 训练和测试数据是来自两个不同会话的 ECoG 记录,间隔大约一周

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N264 ECoGNY278 train;100 test500ms
  • Thomas Lal, Thilo Hinterberger, Guido Widman, Michael Schröder, Jeremy Hill, Wolfgang Rosenstiel, Christian Elger, Bernhard Schölkopf, Niels Birbaumer. Methods Towards Invasive Human Brain Computer Interfaces. Advances in Neural Information Processing Systems (NIPS), 2004 (to appear).

BCI_III_dataset_IIIa_74">BCI III dataset IIIa:

任务:预测左手、右手、脚、舌头的运动

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N460NY2407s,3s
  • A. Schlogl, O. Filz, H. Ramoser, G. Pfurtscheller, GDF-A general dataformat for biosignals, Technical Report, 2004

BCI_III_dataset_IIIb_86">BCI III dataset IIIb:

任务:预测左手、右手的运动想象

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N22 bipolar channelsNY1208s,4s
  • A. Schlogl, O. Filz, H. Ramoser, G. Pfurtscheller, GDF-A general dataformat for biosignals, Technical Report, 2004

BCI_III_dataset_IVa_98">BCI III dataset IVa:

任务:预测脚、右手的运动想象

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N2118NY280*51.75 to 2.25 s
  • Guido Dornhege, Benjamin Blankertz, Gabriel Curio, and Klaus-Robert Müller. Boosting bit rates in non-invasive EEG single-trial classifications by feature combination and multi-class paradigms. IEEE Trans. Biomed. Eng., 51(6):993-1002, June 2004.

BCI_III_dataset_IVb_110">BCI III dataset IVb:

任务:预测脚、左手的运动想象

备注:给出了 118 个 EEG 通道的连续信号,对于训练数据,给出了指示 210 个线索和相应目标类别的时间点的标记。 只为比赛提供了左脚和脚的提示(因为在测试阶段没有进行舌头图像)。其他的动作没有时间线索

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N2118NY210 cues and the corresponding target classes12 min
  • Guido Dornhege, Benjamin Blankertz, Gabriel Curio, and Klaus-Robert Müller. Boosting bit rates in non-invasive EEG single-trial classifications by feature combination and multi-class paradigms. IEEE Trans. Biomed. Eng., 51(6):993-1002, June 2004.

BCI_III_dataset_IVc_124">BCI III dataset IVc:

任务:预测脚、左手的运动想象

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N2118NY210 train;400 test1.75 to 2.25 s
  • Guido Dornhege, Benjamin Blankertz, Gabriel Curio, and Klaus-Robert Müller. Boosting bit rates in non-invasive EEG single-trial classifications by feature combination and multi-class paradigms. IEEE Trans. Biomed. Eng., 51(6):993-1002, June 2004.

BCI_III_dataset_V_136">BCI III dataset V:

任务:对来自 3 个科目的 3 个类(左手、右手、单词联想)进行分类; 除了原始信号,还提供了预先计算的特征

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N332NY3 train ;1 test每次试验4分钟,每个任务15s左右
  • Millán, J. del R.. On the need for on-line learning in brain-computer interfaces Proc. Int. Joint Conf. on Neural Networks., 2004.

BCI_IV_dataset_I_148">BCI IV dataset I:

任务:脑电图、运动想象(左手、右手、脚2类); 评估数据是连续的脑电图,其中也包含空闲状态的时期

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N2(3,但每个subject只选两类)64NY7 subjectcontinuous EEG which contains also periods of idle state
  • Benjamin Blankertz, Guido Dornhege, Matthias Krauledat, Klaus-Robert Müller, and Gabriel Curio. The non-invasive Berlin Brain-Computer Interface: Fast acquisition of effective performance in untrained subjects. NeuroImage, 37(2):539-550, 2007.

BCI_IV_dataset_IIa_160">BCI IV dataset IIa:

任务:脑电图,提示运动想象(左手、右手、脚、舌头)

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N422 EEG channels (0.5-100Hz; notch filtered), 3 EOG channelsYY9 subject*6*4*12=9*2887.5s,3s
  • Brunner C, Leeb R, Müller-Putz G, et al. BCI Competition 2008–Graz data set A[J]. Institute for Knowledge Discovery (Laboratory of Brain-Computer Interfaces), Graz University of Technology, 2008, 16: 1-6.

BCI_IV_dataset_IIb_172">BCI IV dataset IIb:

任务:脑电图,提示运动(左手、右手)

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N43 bipolar EEG channels (0.5-100Hz; notch filtered), 3 EOG channelsYY9 subject*1209s
  • Leeb R, Brunner C, Müller-Putz G, et al. BCI Competition 2008–Graz data set B[J]. Graz University of Technology, Austria, 2008: 1-6.

Dataset_184">High Gamma Dataset

任务:四类动作分别是左手、右手、双脚和休息

下载地址:https://github.com/robintibor/high-gamma-dataset

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N4128NN14 subject*10004s
  • 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.

EEG_Motor_MovementImagery_Dataset_198">EEG Motor Movement/Imagery Dataset

任务:四类动作任务

下载地址:https://www.physionet.org/content/eegmmidb/1.0.0/

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N464NYmore than 15001~2min
  • Schalk, G., McFarland, D.J., Hinterberger, T., Birbaumer, N., Wolpaw, J.R. BCI2000: A General-Purpose Brain-Computer Interface (BCI) System. IEEE Transactions on Biomedical Engineering 51(6):1034-1043, 2004.
  • Goldberger, A., Amaral, L., Glass, L., Hausdorff, J., Ivanov, P. C., Mark, R., … & Stanley, H. E. (2000). PhysioBank, PhysioToolkit, and PhysioNet: Components of a new research resource for complex physiologic signals. Circulation [Online]. 101 (23), pp. e215–e220.

Left/Right Hand MI

任务:四类动作分别是左手、右手、双脚和休息

下载地址:http://gigadb.org/dataset/100295

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N264 EEG;4 EMGYN52 subject*100 or 1203s
  • Cho H, Ahn M, Ahn S, et al. EEG datasets for motor imagery brain–computer interface[J]. GigaScience, 2017, 6(7): gix034.

EEG_Detection_227">Grasp-and-Lift EEG Detection

任务:四类动作分别是左手、右手、双脚和休息

下载地址:https://www.kaggle.com/c/grasp-and-lift-eeg-detection/data

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N632NN12sub*10trail*30test-
  • Luciw M D, Jarocka E, Edin B B. Multi-channel EEG recordings during 3,936 grasp and lift trials with varying weight and friction[J]. Scientific data, 2014, 1(1): 1-11.

The largest SCP data of Motor-Imagery

任务: 4 个 BCI 交互范式,以及多个记录会话和同一个人的范式。 考虑了涉及多达 6 种运动想象状态的 BCI 交互。

下载地址:https://www.kaggle.com/c/grasp-and-lift-eeg-detection/data

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N438NNmore than 60000-
  • Kaya, Murat; Binli, Mustafa Kemal; Ozbay, Erkan; Yanar, Hilmi; Mishchenko, Yuriy (2018): A large electroencephalographic motor imagery dataset for electroencephalographic brain computer interfaces. figshare. Collection. https://doi.org/10.6084/m9.figshare.c.3917698.v1

EEG_motor_activity_data_set_255">EEG motor activity data set

任务:包含两个数据集,一名受试者的 19 电极数据,具有 1D 和 2D 手部动作的各种组合

下载地址:https://sites.google.com/site/projectbci/

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N219NY--
  • Brain Computer Interface research at NUST Pakistan
    If you publish your research making any use of this data, please send us an email so that the usage reference can be added to this page. Thanks

EEG_motor_activity_data_set_270">EEG motor activity data set

任务:包含两个数据集,一名受试者的 19 电极数据,具有 1D 和 2D 手部动作的各种组合

下载地址:https://sites.google.com/site/projectbci/

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N219NY--
  • Brain Computer Interface research at NUST Pakistan
    If you publish your research making any use of this data, please send us an email so that the usage reference can be added to this page. Thanks

Imagination of Right-hand Thumb Movement

任务:想象抬起右手拇指

下载地址:https://archive.ics.uci.edu/ml/datasets/Planning+Relax

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N28NY1825s
  • Rajen Bhatt, ‘Planning-Relax Dataset for Automatic Classification of EEG Signals’, UCI Machine Learning Repository

ERP(含P300)

BCI_II_dataset_IIb_307">BCI II dataset IIb:

任务:屏幕中有6*6的字符,预测用户所关注的数字,P300

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N3664NY12*15-
  • Donchin, E., Spencer, K.M., Wijensinghe, R. The mental prosthesis: Assessing the speed of a P300-based brain-computer interface. IEEE Trans. Rehab. Eng. 8:174-179, \2000.

BCI_III_dataset_II_319">BCI III dataset II

任务:预测测试集中的字符序列,P300

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N3664NY278 train;100 test-
  • Thomas Lal, Thilo Hinterberger, Guido Widman, Michael Schröder, Jeremy Hill, Wolfgang Rosenstiel, Christian Elger, Bernhard Schölkopf, Niels Birbaumer. Methods Towards Invasive Human Brain Computer Interfaces. Advances in Neural Information Processing Systems (NIPS), 2004 (to appear).

AFDP

任务:?P300

特点:

多标记标签数维度/通道数多源小样本数据规模时长
  • Onishi A, Zhang Y, Zhao Q, et al. Fast and reliable P300-based BCI with facial images, 5th Int[C]//BCI Conf. 2011: 191-195.

BCI_Challenge__NER_2015_343">BCI Challenge @ NER 2015

任务:P300 拼写器任务的 26 个受试者、56 个 EEG 通道,以及 P300 解码正确或不正确字母时引发的响应的标记数据集

下载:https://www.kaggle.com/c/inria-bci-challenge

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N256NY80 train;50 test-

Monitoring error-related potentials

任务:6 名受试者有 64 个脑电图电极,观察光标向目标方格移动,并根据光标向正确或错误方向移动来标记引发的反应,ErrP

下载地址:http://bnci-horizon-2020.eu/database/data-sets

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N264NYabout 6sub*50-
  • R. Chavarriaga and J. d. R. Millan, “Learning From EEG Error-Related Potentials in Noninvasive Brain-Computer Interfaces,” in IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 18, no. 4, pp. 381-388, Aug. 2010, doi: 10.1109/TNSRE.2010.2053387.

HCI-Tagging

任务:受试者会在屏幕底部显示带有标签的图像或电影片段。 在某些情况下,标签正确地描述了一些情况。 但是,在其他情况下,标签实际上并不适用于媒体项目。 在每个项目之后,如果参与者同意标签适用于媒体项目,则要求他们按绿色按钮,否则按红色按钮。ErrP

下载地址:https://mahnob-db.eu/hci-tagging/

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N264NY30 subject-

Face vs. House Discrimination

任务:7 名癫痫患者分别接受了 50 个灰度刺激,用于面部和房屋图片。 对于每个受试者,总共进行 3 次实验运行,产生 300 次刺激

下载地址:https://purl.stanford.edu/xd109qh3109

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N--NY7sub*3trail*100400ms
  • Miller K J, Schalk G, Hermes D, et al. Spontaneous decoding of the timing and content of human object perception from cortical surface recordings reveals complementary information in the event-related potential and broadband spectral change[J]. PLoS computational biology, 2016, 12(1): e1004660.

Target Versus Non-Target

任务:P300

下载地址:https://zenodo.org/record/2649069#.YTyQXH2-uUk

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N-16NY36subject-
  • Van Veen G, Barachant A, Andreev A, et al. Building Brain Invaders: EEG data of an experimental validation[J]. arXiv preprint arXiv:1905.05182, 2019.

SSVEP

Dataset_411">Benchmark Dataset

任务:对40个字符进行分类

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N4060NY35 subject*6*405s,1s
  • Wang Y, Chen X, Gao X, et al. A benchmark dataset for SSVEP-based brain–computer interfaces[J]. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2016, 25(10): 1746-1752.

BETA database

任务:提示拼写任务

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N4064NY70 subject*406s
  • Liu B, Huang X, Wang Y, et al. BETA: A large benchmark database toward SSVEP-BCI application[J]. Frontiers in neuroscience, 2020, 14: 627.

FBCCA-DW dataset

任务:?

备注:论文不明http://bci.med.tsinghua.edu.cn/download.html

特点:

多标记标签数维度/通道数多源小样本数据规模时长
?64 electorde + 1 trigger?Y
  • ?同团队最新论文A novel training-free recognition method for SSVEP-based BCIs using dynamic window strategy看不了。。。

Dataset_for_RSVP_BCIs_449">Benchmark Dataset for RSVP BCIs

任务:?

备注:不知道论文http://bci.med.tsinghua.edu.cn/download.html

特点:

多标记标签数维度/通道数多源小样本数据规模时长
?64NY

Dataset_for_Collaborative_BCIs_463">Cross-Session Dataset for Collaborative BCIs

任务:?

备注:论文不明http://bci.med.tsinghua.edu.cn/download.html

特点:

多标记标签数维度/通道数多源小样本数据规模时长
??62 + 1 triggerNY?var
  • ?

BCI_Dataset_477">Wearable SSVEP BCI Dataset

任务:执行基于 12 目标 SSVEP 的 BCI 任务

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N128NY102subject2.84s
  • Zhu F, Jiang L, Dong G, et al. An Open Dataset for Wearable SSVEP-Based Brain-Computer Interfaces[J]. Sensors, 2021, 21(4): 1256.

BCI_Dataset_489">160 Targets SSVEP BCI Dataset

任务:160目标的ssvep任务

备注:论文不明http://bci.med.tsinghua.edu.cn/download.html

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N8/129;160 triggerNY102subject2.84s

dataset-ssvep-exoskeleton

任务:该数据集收集了 12 名受试者在共享控制任务期间操作上肢外骨骼的基于 SSVEP 的 BCI 记录

下载地址;https://github.com/sylvchev/dataset-ssvep-exoskeleton

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N48NY12subject-
  • Vinay Jayaram and Alexandre Barachant. “MOABB: trustworthy algorithm benchmarking for BCIs.” Journal of neural engineering 15.6 (2018): 066011.https://github.com/NeuroTechX/moabb/
  • Kalunga E K, Chevallier S, Barthélemy Q, et al. Online SSVEP-based BCI using Riemannian geometry[J]. Neurocomputing, 2016, 191: 55-68.

A multi-day and multi-band dataset

任务:4个 LED灯注意力

下载地址http://gigadb.org/dataset/100660

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N433Y,有多个频段数据集,多天数据集,呼吸、心电图、肌电图和头部运动等Y30subject*40trail-
  • Choi G Y, Han C H, Jung Y J, et al. A multi-day and multi-band dataset for a steady-state visual-evoked potential–based brain-computer interface[J]. GigaScience, 2019, 8(11): giz133.

EEG_SteadyState_Visual_Evoked_Potential_Signals_Data_Set_530">EEG Steady-State Visual Evoked Potential Signals Data Set

任务:包括 3 种不同的测试,(i) 五盒视觉测试:有参与和无人参与的基于圆盘和正方形的刺激,(ii) 自然图像中的视觉搜索:在黑白自然图像中搜索黄点刺激,(iii) 手抖测试:显示 左/右手关闭/打开图像。 30 个受试者,14 个电极。

下载地址https://archive.ics.uci.edu/ml/datasets/EEG+Steady-State+Visual+Evoked+Potential+Signals#

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N-14NY9200-
  • Fernandez-Fraga, S. M., Aceves-Fernandez, M. A., Pedraza-Ortega, J. C. (2018). Feature Extraction of EEG Signal upon BCI Systems Based on Steady-State Visual Evoked Potentials Using the Ant Colony Optimization Algorithm. Discrete Dynamics in Nature and Society, 2018.
    S. M. Fernandez-Fraga, M. A. Aceves-Fernande, J. C. Pedraza-Ortega & J. M. Ramos-Arreguín (2018). Screen Task Experiments for EEG Signals Based on SSVEP Brain Computer Interface. International Journal of Advanced Research, 2018.

VEP

BCI_with_dry_electrodes_549">c-VEP BCI with dry electrodes

任务:9 名受试者、15 个干脑电图通道用于 VEP BCI 拼写器(32 个字符)任务,以及为与拼写器相关联的标签引出的响应的标记数据集

下载地址:https://figshare.com/articles/dataset/A_High-Speed_Brain-Computer_Interface_BCI_Using_Dry_EEG_Electrodes/4640686

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N3215NY12subject-
  • Spüler M. A high-speed brain-computer interface (BCI) using dry EEG electrodes[J]. PloS one, 2017, 12(2): e0172400.

Emotion Recognition

DEAP

任务:包括 32 个subject,每个观看 1 分钟长的音乐视频摘录,由用户根据唤醒/效价/喜欢-不喜欢/支配性/熟悉度进行评分,以及 22/32 主题的正面记录。

下载地址:http://www.eecs.qmul.ac.uk/mmv/datasets/deap/readme.html

特点:

多标记标签数维度/通道数多源小样本数据规模时长
4维情感32eeg and 8 otherYN32subject*40trail60s
  • Koelstra S, Muhl C, Soleymani M, et al. Deap: A database for emotion analysis; using physiological signals[J]. IEEE transactions on affective computing, 2011, 3(1): 18-31.

SEED、SEED-IV、SEED-VIG

任务:向 15 名受试者展示了引发积极/消极/中性情绪的视频剪辑,并通过 62 个通道记录了 EEG。。

下载地址:https://bcmi.sjtu.edu.cn/~seed/seed.html

特点:

多标记标签数维度/通道数多源小样本数据规模时长
Y62NN15sub*15trailabout 4min
  • Ruo-Nan Duan, Jia-Yi Zhu and Bao-Liang Lu, Differential Entropy Feature for EEG-based Emotion Classification, Proc. of the 6th International IEEE EMBS Conference on Neural Engineering (NER). 2013: 81-84.
  • Wei-Long Zheng, and Bao-Liang Lu, Investigating Critical Frequency Bands and Channels for EEG-based Emotion Recognition with Deep Neural Networks, accepted by IEEE Transactions on Autonomous Mental Development (IEEE TAMD) 7(3): 162-175, 2015.

Imagined Emotion

任务:31 名受试者,受试者会听取暗示情绪感受的录音,并要求受试者想象一个情绪场景或回忆他们以前感受到这种情绪的经历。

下载地址:https://headit.ucsd.edu/studies/3316f70e-35ff-11e3-a2a9-0050563f2612

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N2256NN31sub*15trail-
  • Onton J A, Makeig S. High-frequency broadband modulation of electroencephalographic spectra[J]. Frontiers in human neuroscience, 2009, 3: 61.

Neuro Marketing

任务:25 个subject,14 个电极,喜欢/不喜欢商业电子商务产品,超过 14 个类别,每个类别 3 个图像。 数据集文章:EEG 信号分析及其在神经营销中的应用。

下载地址:https://drive.google.com/open?id=0B2T1rQUvyyWcSGVVaHZBZzRtTms

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N214NY25sub*42-
  • Yadava M, Kumar P, Saini R, et al. Analysis of EEG signals and its application to neuromarketing[J]. Multimedia Tools and Applications, 2017, 76(18): 19087-19111.

AMIGOS

任务:一个数据库,通过神经生理信号研究性格、人格特征和情绪。与其他数据库不同,在两种配置中同时使用短视频和长视频,一种是单个观看者,另一种是观看者组

下载地址: http://www.eecs.qmul.ac.uk/mmv/datasets/amigos/

备注:引文里面还整理了 很多情绪识别的数据集

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N多维评估14Y脑电图(EEG)、心电图(ECG)和电镀皮肤反应(GSR)N40sub-
  • AMIGOS: A Dataset for Affect, Personality and Mood Research on Individuals and Groups , J.A. Miranda-Correa, M.K. Abadi, N. Sebe, and I. Patras, IEEE Transactions on Affective Computing, 2018.

DECAF

任务:一个数据库,通过神经生理信号研究性格、人格特征和情绪。与其他数据库不同,在两种配置中同时使用短视频和长视频,一种是单个观看者,另一种是观看者组

下载地址:http://mhug.disi.unitn.it/wp-content/DECAF/DECAF.html

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N多维评估306Y脑磁图 (MEG)、水平眼电图 (hEOG)、心电图 (ECG)、斜方肌肌电图 (tEMG)、近红外面部视频N30sub*(40+36)-
  • Abadi M K, Subramanian R, Kia S M, et al. DECAF: MEG-based multimodal database for decoding affective physiological responses[J]. IEEE Transactions on Affective Computing, 2015, 6(3): 209-222.

杂项

eeg dataset

任务:一个二元类 MI 系统、一个 36 符号 ERP 拼写器和一个四目标频率 SSVEP 系统。

下载地址http://gigadb.org/dataset/100542

特点:

多标记标签数维度/通道数多源小样本数据规模时长
N2、36、462NN44*(100train+100test);44*(100train+100test);44*(1980train+2160test)4s,4s,13s
  • Lee M H, Kwon O Y, Kim Y J, et al. EEG dataset and OpenBMI toolbox for three BCI paradigms: an investigation into BCI illiteracy[J]. GigaScience, 2019, 8(5): giz002.

physionet dataset

备注:里面有非常非常非常多个开源数据集

地址:https://www.physionet.org/about/database/

BNCI Horizon 2020

备注:里面有非常非常多个开源数据集

地址:http://bnci-horizon-2020.eu/database/data-sets

EEG__ERP_data_686">EEG / ERP data

备注:里面有多个开源数据集

地址:https://sccn.ucsd.edu/~arno/fam2data/publicly_available_EEG_data.html


http://www.niftyadmin.cn/n/770329.html

相关文章

EEG Feature Selection via Global Redundancy Minimization for Emotion Recognition 论文笔记

概述 提出了一种基于最小化全局冗余的特征提取算法(global redundancy minimization in orthogonal regression)。 研究动机 基于EEG的情绪识别研究大多使用低密度(小于128个通道)的EEG数据,如SEED [71]和DEAP[72]数据库。低密度脑电图的缺点是其空间分辨率差。随…

信号数据shannon entropy计算

import math import numpy as np def shannon_entropy(single,k):single:1-D信号k: 分多少个binnumofx single.shape[0]maxV np.max(single)minV np.min(single)bin np.linspace(minV,maxV,k1)bin_numx [0]*k # 落在每个bin的数据点数目for x in single: # 计算落在…

信号数据C0 Complexity计算

参考论文:The study of c0 complexity on epileptic absence seizure def C0_complexity(single):计算C0复杂度 x fft(single)N x.shape[0]avgx np.average(x) # 计算幅度谱的平均值new_x []for i in x:if abs(i) > avgx:new_x.append(i)else:new_x.appe…

多条eeg数据对比图绘制

画图,输入多条数据,进行可视化对比。建议10以内。import plotly.graph_objs as go import numpy as np from plotly.subplots import make_subplots def plot_various_chns(data,names []):画图,输入多条数据,进行可视化对比。建…

mne计算CSP

# get CSP feature from mne.decoding import CSP csp CSP(n_components10, regNone, logTrue, norm_traceTrue) csp_train csp.fit_transform(train_data, train_label) 其中 train_data 的格式为[trial,channel,time];train_label的为[trial,] 详情可以参考mne…

pip 配置下载源

pip官方镜像源 官方: https://pypi.python.org/simple 国内镜像源 清华:https://pypi.tuna.tsinghua.edu.cn/simple 阿里云:http://mirrors.aliyun.com/pypi/simple/ 中国科技大学 https://pypi.mirrors.ustc.edu.cn/simple/ 华中理工大…

Nan 数据的检测与处理

print(is exist nan: , end) print(np.isnan(x).any())# 查找位置 nan_inx np.where(np.isnan(x)) print(nan_inx)# 替换 x[np.isnan(x)]0.00001print(is exist nan: , end) print(np.isnan(x).any())

EEG数据预处理的一些坑,模型loss=nan的原因与解决方法

在使用深度学习模型对数据进行分类时,出现lossnan的现象。 原因:数据中有接近0的坏数据或者Nan 例如: 解决方法:去掉就好 nan的检测与去除可以参考此处。