Measureing intensity on a surface

Contents

Measureing intensity on a surface#

This example shows how to measure the intensity of a surface. The surface is defined by a set of points(vertices) and a set of triangles. The intensity is measured by averaging the intensity along the normal vectors of the vertices.

from napari_stress import sample_data, reconstruction, measurements, vectors, sample_data
import numpy as np
import napari
import vedo
from skimage import transform

Example data#

We first need to create a surface. In this tutorial, we take the 3d image of the 4d droplet sample dataset and quickly reconstruct a surface from it using the droplet reconstruction toolbox.

viewer = napari.Viewer(ndisplay=3)

Assistant skips harvesting pyclesperanto as it's not installed.
Assistant skips harvesting pyclesperanto_prototype as it's not installed.

points = sample_data.get_droplet_point_cloud()[0]
droplet4d = sample_data.get_droplet_4d()[0]
scale = droplet4d[1]['scale'][1:]

droplet = droplet4d[0][0]
droplet = transform.rescale(droplet, scale, anti_aliasing=True)

viewer.layers.clear()
viewer.add_image(droplet, name='droplet')
viewer.add_layer(napari.layers.Layer.create(*points))
napari.utils.nbscreenshot(viewer)

We now create surface from the reconstructed points:

pointcloud_vedo = vedo.Points(points[0][:, 1:])
surface_vedo = pointcloud_vedo.reconstruct_surface().smooth(niter=50).decimate(fraction=0.01)
surface = (surface_vedo.vertices, np.asarray(surface_vedo.cells))
viewer.add_surface(surface, name='surface', blending='opaque')
napari.utils.nbscreenshot(viewer)

We now measure intensity on the surface. The measured values are simply the intensities on the surface. We can add some more metrics (mean, std, min, max) to results:

measured_intensities = measurements.measure_intensity_on_surface(surface,
                                                                 droplet,
                                                                 measurement_range = 3,
                                                                 sampling_distance=0.5, interpolation_method='linear')

intensity_mean = measured_intensities.mean(axis=1)
intensity_std = measured_intensities.std(axis=1)
intensity_max = measured_intensities.max(axis=1)
intensity_min = measured_intensities.min(axis=1)

measured_intensities['intensity_mean'] = intensity_mean
measured_intensities['intensity_std'] = intensity_std
measured_intensities['intensity_max'] = intensity_max
measured_intensities['intensity_min'] = intensity_min
measured_intensities
step_0 step_1 step_2 step_3 step_4 step_5 intensity_mean intensity_std intensity_max intensity_min
0 0.331049 0.414606 0.497417 0.580326 0.648891 0.667630 0.523320 0.133642 0.667630 0.331049
1 0.382389 0.455524 0.528822 0.602113 0.627379 0.645631 0.540310 0.104645 0.645631 0.382389
2 0.348877 0.422235 0.481675 0.538292 0.592035 0.627899 0.501835 0.105353 0.627899 0.348877
3 0.347693 0.412790 0.477218 0.540746 0.584013 0.600044 0.493751 0.099815 0.600044 0.347693
4 0.332954 0.408572 0.473742 0.536397 0.596677 0.637521 0.497644 0.115328 0.637521 0.332954
... ... ... ... ... ... ... ... ... ... ...
382 0.058715 0.077971 0.139337 0.204018 0.292200 0.385238 0.192913 0.127327 0.385238 0.058715
383 0.052107 0.077134 0.121162 0.196388 0.281357 0.383745 0.185316 0.128284 0.383745 0.052107
384 0.059532 0.086790 0.139312 0.214852 0.304453 0.407926 0.202144 0.134586 0.407926 0.059532
385 0.046520 0.066307 0.117432 0.184905 0.270279 0.365379 0.175137 0.124142 0.365379 0.046520
386 0.060281 0.088385 0.131813 0.209437 0.295303 0.400105 0.197554 0.131001 0.400105 0.060281

387 rows × 10 columns

viewer.layers.clear()
viewer.add_surface((surface[0], surface[1], measured_intensities['intensity_max'].values),
                   name='surface_intensity_max', colormap='inferno', blending='opaque')
napari.utils.nbscreenshot(viewer)
Contents