From e9f65e2c30a806adfd1bdff3b341b0730c49f49a Mon Sep 17 00:00:00 2001
From: amitjans <amitjans@ethz.ch>
Date: Mon, 31 May 2021 20:47:44 +0200
Subject: [PATCH] Compute 2d distance instead of 1d
---
tools.py | 71 ++++++++++++++++++++++++++++++++------------------------
1 file changed, 41 insertions(+), 30 deletions(-)
diff --git a/tools.py b/tools.py
index 9d79448..fe274d5 100644
--- a/tools.py
+++ b/tools.py
@@ -218,6 +218,14 @@ def ion_positions(image):
Finally return the coordinates together with the maximum ion width.
"""
coordinates, max_width = _max_coord(image, width=True) # measure local max
+ # delete local maxima that are not in the same axis as the brightest spot.
+ try:
+ h_axis = coordinates[0, 0]
+ coordinates = coordinates[np.logical_and(coordinates[:, 0] < h_axis + 15,
+ coordinates[:, 0] > h_axis - 15)]
+ except IndexError:
+ pass
+
coordinates += np.flip(settings.REGION[0]) # change reference
coordinates = coordinates[coordinates[:, 1].argsort()] # order
# add 3rd column indicating the ion's state (bright, dark, wrong isotope)
@@ -287,8 +295,11 @@ def compensate_gradient(image, coord):
slope = _measure_gradient(image, coord)
- for i in range(image.shape[1]):
- image[:, i] = np.clip(image[:, i] - i*slope, 0, None)
+ start = coord[0, 1] - 30
+ end = coord[-1, 1] + 30
+
+ for i in np.r_[start:end]:
+ image[:, i] = np.clip(image[:, i] - (i-start)*slope, 0, None)
return image.astype(np.uint16)
@@ -335,7 +346,7 @@ def dark_ions(image, coord):
"""
# Compute the positions starting from the left/right
x_pos = np.stack((coord[:, 1] - coord[0, 1], np.sort(np.abs(coord[:, 1] - coord[-1, 1]))))
- x_dist = _ion_distance(coord[:, 1]) # compute relative distances
+ x_dist = _ion_distance(coord) # compute relative distances
# Load the masks. TODO: Remove the mask options that end up outside the image!
low_idx = max(0, len(x_pos[0]) - 2)
high_idx = min(28, len(x_pos[0]) * 2)
@@ -364,33 +375,32 @@ def dark_ions(image, coord):
comp[2, :, -1] = 0
loss = np.sum(comp, axis=2)/len(x_pos)
- if loss.min() <= 20:
- idx = np.where(loss == np.min(loss))
-
- if idx[0][0] == 1: # if using "inverted positions", transform back
- new_x_pos = np.abs(mask[idx[1][0], :][mask[idx[1][0], :] >= 0] - coord[-1, 1])
- else:
- new_x_pos = mask[idx[1][0], :][mask[idx[1][0], :] >= 0] + coord[0, 1]
-
- diff_n = len(new_x_pos) - len(x_pos[0])
- for i in range(diff_n):
- mask2 = comparison[idx[0][0], idx[1][0], :] < 100000
- idx2 = np.argsort(comparison[idx[0][0], idx[1][0], :][mask2])[-(diff_n - i)]
- if idx2 != 0:
- # If we are using the distances' loss, then we have to add 1 to the index
- if idx[0][0] == 2:
- idx2 += 1
- coord = np.append(coord, np.array([[np.mean(coord[:, 0]),
- new_x_pos[idx2], 2]]), axis=0)
-
- elif idx2 == 0 and idx[0][0] == 2:
- coord = np.append(coord, np.array([[np.mean(coord[:, 0]), new_x_pos[0] -
- comparison[2, idx[1][0], 0], 2]]), axis=0)
-
- coord = coord[coord[:, 1].argsort()].astype(int)
- coord = coord[coord[:, 1] < image.shape[1]] # remove points outside the image
+ idx = np.where(loss == np.min(loss))
+ if idx[0][0] == 1: # if using "inverted positions", transform back
+ new_x_pos = np.abs(mask[idx[1][0], :][mask[idx[1][0], :] >= 0] - coord[-1, 1])
else:
+ new_x_pos = mask[idx[1][0], :][mask[idx[1][0], :] >= 0] + coord[0, 1]
+
+ diff_n = len(new_x_pos) - len(x_pos[0])
+ for i in range(diff_n):
+ mask2 = comparison[idx[0][0], idx[1][0], :] < 100000
+ idx2 = np.argsort(comparison[idx[0][0], idx[1][0], :][mask2])[-(diff_n - i)]
+ if idx2 != 0:
+ # If we are using the distances' loss, then we have to add 1 to the index
+ if idx[0][0] == 2:
+ idx2 += 1
+ coord = np.append(coord, np.array([[np.mean(coord[:, 0]),
+ new_x_pos[idx2], 2]]), axis=0)
+
+ elif idx2 == 0 and idx[0][0] == 2:
+ coord = np.append(coord, np.array([[np.mean(coord[:, 0]), new_x_pos[0] -
+ comparison[2, idx[1][0], 0], 2]]), axis=0)
+
+ coord = coord[coord[:, 1].argsort()].astype(int)
+ coord = coord[coord[:, 1] < image.shape[1]] # remove points outside the image
+
+ if loss.min() > 20:
warnings.warn("Loss is too high... "
"Try calculating a new factor value using calibrate() function.")
@@ -543,7 +553,7 @@ def _max_coord(image, width=False):
image = image[region[0, 1]:region[1, 1], region[0, 0]:region[1, 0]]
filt_image = _difference_of_gaussians(image, low_sigma=settings.BLOB_SIZE, preserve_range=True)
- coordinates = peak_local_max(filt_image, threshold_abs=0.85, min_distance=25)
+ coordinates = peak_local_max(filt_image, threshold_abs=0.85, min_distance=settings.ION_DIST)
if width:
try:
@@ -648,7 +658,8 @@ def _ion_distance(coord):
"""
dist = []
for i in range(len(coord) - 1):
- dist.append(coord[i + 1] - coord[i])
+ dist.append(np.sqrt((coord[i, 0] - coord[i + 1, 0])**2 +
+ (coord[i, 1] - coord[i + 1, 1])**2).item())
dist = np.array(dist)
return dist
--
GitLab