Skip to content
Gallery
BCV
Timelines
Thoughts
Experiment
Documentation of used HWs
Measure System Performance
Sloved Issues
DL Models
CS-2 tutorial at Durham HPC Days
Increasing buffer size on ROS
Streaming EEG
Cap electrodes
Reading List
Flashing Jetson Xavier NX
Installing Required PKGs
Visualizing EEG Data
Shill Unix Commands
Create Systemd Services
Paper - check
Nvidia Jetson Developer Kit
Normalizing EEG Channels
Systematic Review
EEG Annotation
Meetings
Signal Processing
Hamilton Supercomputer
VS issues
Adjusting Resolution and Frame Rate
Testing flickering circles
DL Models
Bag Files
LV - 11 JUL
Share
Explore
Meetings
Link to page
⁠
fft_all_channels.png
⁠
⁠
⁠
fft_separate_channels.png
⁠
⁠
⁠
correlation_matrix_all_channels.png
⁠
⁠


Dominant freq befor anf after SSVEP band filter
Label
Channel
Dominant Frequency Before (Hz)
Dominant Frequency After (Hz)
stop
FP1
3.5
4.5
stop
FP2
3.5
4.5
stop
C3
3
4.5
stop
C4
3.5
4.5
stop
T3
3
4.5
stop
T4
3
4.5
stop
O1
3
4.5
stop
O2
4
4.5
stop
F7
3
4.5
stop
F8
3
4.5
stop
F3
3
4.5
stop
F4
3
4.5
stop
T5
3.5
4.5
stop
T6
3
4.5
stop
P3
3
4.5
stop
P4
3
4.5
turn_left
FP1
3.5
9.5
turn_left
FP2
3.5
10
turn_left
C3
3
9.5
turn_left
C4
3.5
9.5
turn_left
T3
3
9.5
turn_left
T4
3
10
turn_left
O1
3
9.5
turn_left
O2
4
9
turn_left
F7
3
9.5
turn_left
F8
3
10.5
turn_left
F3
3
9.5
turn_left
F4
3
10
turn_left
T5
3.5
10
turn_left
T6
3
10
turn_left
P3
3
10.5
turn_left
P4
3
8.5
reverse
FP1
3.5
14
reverse
FP2
3.5
15
reverse
C3
3
14
reverse
C4
3.5
14
reverse
T3
3
13.5
reverse
T4
3
15
reverse
O1
3
13.5
reverse
O2
4
15
reverse
F7
3
16
reverse
F8
3
14.5
reverse
F3
3
15
reverse
F4
3
15
reverse
T5
3.5
14.5
reverse
T6
3
16
reverse
P3
3
13.5
reverse
P4
3
13.5
turn_right
FP1
3.5
20.5
turn_right
FP2
3.5
19
turn_right
C3
3
18.5
turn_right
C4
3.5
20.5
turn_right
T3
3
19.5
turn_right
T4
3
19
turn_right
O1
3
19.5
turn_right
O2
4
19.5
turn_right
F7
3
19
turn_right
F8
3
20
turn_right
F3
3
19.5
turn_right
F4
3
20
turn_right
T5
3.5
19
turn_right
T6
3
21.5
turn_right
P3
3
18.5
turn_right
P4
3
19
forward
FP1
3.5
30
forward
FP2
3.5
30
forward
C3
3
28.5
forward
C4
3.5
30
forward
T3
3
28.5
forward
T4
3
28.5
forward
O1
3
28.5
forward
O2
4
30
forward
F7
3
28.5
forward
F8
3
28.5
forward
F3
3
28.5
forward
F4
3
28.5
forward
T5
3.5
29.5
forward
T6
3
28.5
forward
P3
3
30
forward
P4
3
30
There are no rows in this table
⁠

Statistical Features:

Mean (feature_mean):
Mean of each signal.
Standard Deviation (feature_stddev):
Standard deviation of each signal.
Skewness and Kurtosis (feature_moments):
Skewness: Asymmetry of the signal distribution.
Kurtosis: "Peakedness" of the signal distribution.
Max and Min (feature_max and feature_min):
Maximum and minimum values of each signal.
Covariance Matrix (feature_covariance_matrix):
Variances (diagonal) and covariances (off-diagonal) of signals.
Eigenvalues of Covariance Matrix (feature_eigenvalues):
Eigenvalues of the covariance matrix of signals.
Logarithm of Covariance Matrix (feature_logcov):
Matrix logarithm of the covariance matrix.
⁠

Temporal Features:

Mean Differences (feature_mean_d):
Change in mean values between two half-windows.
Standard Deviation Differences (feature_stddev_d):
Change in standard deviation values between two half-windows.
Max/Min Differences (feature_max_d, feature_min_d):
Change in max/min values between two half-windows.
Mean Differences Across Quarters (feature_mean_q):
Means of each quarter-window and differences between them.
Max/Min Differences Across Quarters (feature_max_q, feature_min_q):
Maximum and minimum values of each quarter-window and their differences.
⁠

Spectral Features:

Fast Fourier Transform (FFT) (feature_fft):
Top N frequencies with the highest power.
Full power spectrum of each signal (optional).
Ability to filter out DC components and mains frequency (e.g., 50 Hz).
⁠

Custom Feature Engineering:

Lag Window Features:
For each time window, features from the previous window (lag) are concatenated.
Redundant lag features can be removed to avoid unnecessary overlap.
Want to print your doc?
This is not the way.
Try clicking the ⋯ next to your doc name or using a keyboard shortcut (
CtrlP
) instead.