fig1ab.txt: 

Two example consecutive frames of a rotating filament. Each line represents 64 horizontal pixels from the frame, each column represents vertical 80 pixels. Frame #1 and frame #2 are in the same file and contain the data for Fig. 1(a) and Fig. 1(b).


fig3abcd.txt: 

Pre-processing steps for the analysis. Same representation as for Fig. 1. Fig 3 (a,b,c,d) are in the same file in respective order:
GPI_SLICED_FULL data (Fig. 3 (a)): Data for Figure 3 (a)
GPI_NORMALIZED data (Fig. 3 (b)): Data for Figure 3 (b)
GPI_DETREND_VEL data (Fig. 3 (c)): Data for Figure 3 (c)
GPI_GAUSS_DIFF data (Fig. 3 (d)): Data for Figure 3 (d)

fig4a.txt: 

Fourier magnitude spectrum of a pre-processed GPI frame.
x (pix) is the x coordinate in the Fourier magnitude spectrum.
y (pix) is the y coordinate.
FFT linear is the Fourier magnitude spectrum. Each line represents a horizontal row of pixels in the x axis while each column represents a vertical column of pixels in the y axis.

fig4b.txt:

Same representation as Fig. 4 (a), but for the consecutive frame.

 
fig4c.txt:

Log-polar transformed Fourier magnitude spectrum of the spectrum in Fig. 4 (a).
phi (deg) is the polar coordinate in the transformed spectrum.
r (pix) is the radial coordinate in the transformed spectrum.
FFT logpol: Log-polar Fourier magnitude spectrum. Each line represents a horizontal row of pixels in the polar axis while each column represents a vertical column of pixels in the radial axis.

fig4d.txt:

Same representation as Fig. 4 (c), but for the consecutive frame.


fig5.txt:

Cross-correlation coefficient function calculated between the log-polar transformed Fourier magnitude spectra for the full polar and the [-31,31] radial range. 
r lag (pix): radial lag coordinate in the CCCF.
phi lag (deg): polar lag coordinate in the CCCF
2D CCCF data: 2D spatial cross-correlation coefficient function. Each line represents a horizontal row of pixels in the phi lag axis while each column represents a vertical column of pixels in the r lag axis.


fig6.txt:

Results of the cross-correlation threshold assessment for NSTX GPI measurement for shot 141319.
Each row represents a point in the time series.
The columns represent the time, and the angular velocity estimates for different correlation thresholds from 0.9 to 0.4 with 0.1 increments, respectively.


fig7.txt:

Synthetic frames of a normal (Frame #1, Fig. 7 (a)) and a rotated (Frame #2, Fig. 7 (b)) Gaussian-structure. Same representation as in Figure 1.


fig8.txt:

Relative uncertainty of the angular rotation estimation vs. angle of rotation and the Gaussian structure size.
Angle rotation: rotation in degrees.
Size vector: FWHM of the Gaussian structures in pixels.
Relative uncertainty: Each row corresponds to a Gaussian size setting. Each column corresponds to an angle of rotation setting.


fig9.txt:

Relative uncertainty of the angular rotation estimation vs. angle and elongation.
Angle rotation: rotation in degrees
Elongation: elongation of the Gaussian-structure.
Relative uncertainty: Each row corresponds to a rotation angle setting, each column corresponds to an elongation setting.


fig10.txt:

Mean and standard deviation of the relative uncertainty vs. angle of rotation and relative noise level.
Angle of rotation: angle of rotation in degrees.
Relative noise level: relative noise level settings.
Mean relative uncertainty of the angular rotation estimation: Each row corresponds to a rotation angle setting, each column corresponds to a relative noise level setting. Data for Fig. 10 (a).
Standard deviation of the relative uncertainty of the angular rotation estimation: Each row corresponds to a rotation angle setting, each column corresponds to a relative noise level setting. Data for Fig. 10 (b).


fig11_[484.1**ms].txt:

GPI frames of a blob event for shot 141307. Same representation as for Fig. 1. The time stamp of the frame is in the file name.


fig12.txt:

Angular velocity estimation results for blobs in shot 141307.
Time (ms): Time vector in milliseconds.
Angular velocity (rad/s): Angular velocity in rad/s.


fig13_[***us].txt:

GPI frames of the ELM crash for the corresponding times around the ELM time, t_ELM for shot 141319. Same representation of the frames as for Fig. 1. The time with respect to the ELM time is in the file name.


fig14.txt:

Angular velocity estimation results for an ELM filament in shot 141319.
Time (ms): Time vector in milliseconds.
Angular velocity (rad/s): Angular velocity in rad/s.


fig15a.txt:

Comparison of the angular velocity estimation techniques for shot 141307.
Time CCCF (ms): Time vector in milliseconds for the CCCF technique.
Angular velocity CCCF (krad/s): Angular velocity calculated with the CCCF based technique.
Time contour (ms): Time vector in milliseconds for the contour technique.
Angular velocity contour (krad/s): Angular velocity estimate with the contour based technique.
Time watershed (ms): Time vector in milliseconds for the watershed technique.
Angular velocity watershed (krad/s):Angular velocity estimate with the watershed based technique.


fig15b.txt:

Comparison of the angular velocity estimation techniques for shot 141319.
Same representation as for Fig. 15 (a).