Created
January 12, 2017 15:31
-
-
Save DanHickstein/2342229c9af2aa5223ab0e5e81616f13 to your computer and use it in GitHub Desktop.
cross-peak-DH
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import numpy as np | |
import matplotlib.pyplot as plt | |
import matplotlib.gridspec as gridspec | |
import abel | |
from scipy.ndimage import zoom | |
from scipy import signal | |
import peakutils | |
fn = "O-10N21024.txt" # raw VMI binned to 1024x1024 pixel image | |
im = np.loadtxt(fn) | |
im = zoom(im, 2) # improves angular resolution for small radius | |
im /= im.max() # normalize, consistent between images | |
# centre | |
imc = abel.tools.center.center_image(im, center='convolution') | |
# map to polar coordinates | |
polarim, r, t = abel.tools.polar.reproject_image_into_polar(imc, Jacobian=False) | |
R = r[:, 0] # slice radial grid | |
nslices = 32 # divide image into 32 slices | |
pie = np.array_split(polarim, nslices, axis=1) | |
angle = np.array(np.hsplit(t[0], nslices)).mean(axis=1) | |
n2 = len(angle)//2 # mid point is zero angle | |
# restrict radial range to peak of interest | |
subr = np.logical_and(R >= 0, R <= 1000) | |
previous = pie[0][subr].sum(axis=1) # -pi/2 slice intensity profile | |
xcorrpk = [] | |
profpk = [] | |
for ang, aslice in zip(angle[1:], pie[1:]): | |
# slice profile | |
profile = aslice[subr].sum(axis=1) | |
# cross correlation | |
corr = signal.correlate(previous, profile, "same") | |
plt.plot(corr) | |
# find xcorr position | |
ind = peakutils.indexes(corr, thres=0.3, min_dist=40) | |
# fine the peak location, setting the center of the R-array to zero | |
# I'm not quite sure why the "-0.5" is needed. It might be something to do with the pixels? | |
peaks = peakutils.interpolate( R[subr]-np.max(R[subr])*0.5-0.5, corr, ind=ind) | |
# sort the peaks by smallest absolute value, to find the one in the center | |
peaks = peaks[np.argsort(np.abs(peaks))] | |
print peaks | |
peak = peaks[0] | |
xcorrpk.append(peak) | |
# peak position using raw profile | |
indraw = peakutils.indexes(profile, thres=0.3, min_dist=40) | |
peaks = peakutils.interpolate(R[subr], profile, ind=indraw) | |
profpk.append(peaks[0]) | |
previous = profile | |
plt.figure() | |
xcorrpk = np.array(xcorrpk) | |
# --- plots -------------- | |
gs = gridspec.GridSpec(1, 1) | |
ax0 = plt.subplot(gs[0, 0]) | |
ax0.plot(angle[1:], np.cumsum(-xcorrpk)+np.max(R[subr])*0.5, 'o-', label='xcorr') | |
ax0.plot(angle[1:], profpk, 'o-', label='peak') | |
ax0.axis(xmin=-np.pi, xmax=np.pi) | |
ax0.set_xticks([-np.pi, 0, np.pi]) | |
ax0.set_xticklabels([r"-$\pi$", "0", r"$\pi$"]) | |
ax0.set_xlabel("angle (radians)") | |
ax0.set_ylabel("relative position") | |
plt.legend(frameon=False) | |
plt.savefig("correlation.png", dpi=75) | |
plt.show() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment