myimagelib.pivLib.compact_PIV

class myimagelib.pivLib.compact_PIV(data)

Compact PIV data structure. Instead of saving PIV data of each frame pair in separated text files, we can save them in a more compact form, where (x, y, mask) information are only saved once and only velocity informations are kept in 3D arrays. The data will be saved in a Matlab style .mat file, and the internal structure is a Python dictionary, with entries (x, y, labels, u, v, mask). Since here x, y, u, v are no longer the same shape, accessing PIV data from a specific frame becomes less straight forward. For example, when doing a quiver plot, one needs to call quiver(x, y, u[0], v[0], instead of quiver(x, y, u, v]. This class is written to enable straightforward data access and saving. For a more detailed guide of using this class, see compact_PIV tutorial. You can also download the notebook to run the code locally. Note that it requires you to download myimagelib package.

Syntax

from myimagelib import readdata, compact_PIV

# construct compact_PIV object from a folder containing PIV data
folder = "path/to/piv/data" # folder containing .csv files
l = readdata(folder, "csv")
cpiv = compact_PIV(l)

# get the data for one frame
x, y, u, v = cpiv.get_frame(0)

# save the data to a .mat file
cpiv.to_mat("cpiv.mat")

# Update mask
from skimage.io import imread
mask = imread("path/to/mask.tif")
cpiv.update_mask(mask)

Edit

• Jan 13, 2023 – Add update_mask(). The idea is that the original mask may include regions where image quality is bad, e.g. the bottom shadow region of droplet xz images. In this case, we usually realize the problem after performing the PIV. And to refine the PIV data, we want to update the mask to make it more conservative (i.e. mask out the bad quality region). Therefore, a method is needed to update the “mask” entry in a compact_PIV object.

__init__(data)

Initialize compact_PIV object from data.

Parameters:

data – can be dict or pandas.DataFrame. If it’s dict, set the value directly to self.data. If it’s DataFrame, construct a dict from the DataFrame (filelist).

Methods

__init__(data)	Initialize compact_PIV object from data.
get_frame(i[, by])	Get PIV data [x, y, u, v] for a specific frame.
get_labels()	Returns filenames originally used for constructing this data.
to_csv(folder)	Save as .csv files to given folder.
to_mat(fname)	Save the compact_PIV data to a .mat file.
update_mask(mask_img)	Update mask in the compact_PIV object.