Randomization of experimental settings

pykitPIV allows for randomization of various experimental settings. The general philosophy is that the experimental setting which can be randomized across images can be passed as a two-element tuple, where the first element of the tuple denotes the admissible minimum value and the second element of the tuple denotes the admissible maximum value. But it can also be forced to a fixed value across images by passing an int or a float instead of a tuple. Below, we explain the usage of both modes.

Using randomized values

For example, if we wanted to generate a spread of seeding particles’ diameters to be any random value between \(1 \text{px}\) and \(4 \text{px}\), across some number of images, we could specify:

diameters = (1, 4)

which denotes that the minimum particle diameter should be \(1 \text{px}\) and the maximum particle diameter should be \(4 \text{px}\).

During image generation, the actual particle diameters for each of the \(N\) images will be drawn from a uniform distribution like so:

import numpy as np

# Number of images to be generated:
n_images = 20

# Generate a spread of diameters for each of the 20 images:
diameters_per_image = np.random.rand(n_images) * (diameters[1] - diameters[0]) + diameters[0]

diameters_per_image is a numpy array that provides a random uniform spread of particle diameters between \(1 \text{px}\) and \(4 \text{px}\). If we were to print diameters_per_image, we will see something like this:

array([3.9681588 , 2.67486177, 3.71224428, 3.67575301, 3.67462814,
       3.81316186, 1.26272335, 3.09337045, 2.75332142, 1.69280529,
       1.41234225, 1.05471252, 1.02885118, 2.17232387, 2.89499995,
       1.04699985, 1.1061524 , 3.15070309, 1.20268356, 3.06636067])

In this case, the first image will have all particles with diameter \(3.97 \text{px}\), the second image will have all particles with diameter \(2.67 \text{px}\), and so on…

Forcing a fixed value

The same experimental setting, which by default is randomized between some minimum and maximum value, can also be set to a fixed value for all \(N\) images. This can be accomplished by setting that parameter to an int or a float instead of a tuple.

For example, when:

diameters = 2.0

particle diameters will be exactly \(2 \text{px}\) for all \(N\) images.

Of course, this will accomplish the same thing as setting:

diameters = (2, 2)

but using a single int or a float is a more elegant way and makes it clearer in the code that the value is fixed.