We Keep Coding

Morphology with OpenCV on Xarray data array is shifting image

I am having an issue using the morphology functions in OpenCV on xarray data with python.
I have generated a list of binary xarray datarrays (I understand that I probably should make these just another dimension of the dataset, but I haven't gotten there yet).
I am trying to run morphological closing on this data. In the code snippet below, I extract the actual data from the datarray as a numpy array, then try to run the closing on that. My understanding of OpenCV is that when it reads an image in, it essentially translates it into a numpy array anyway, so I thought this might work. I should mention, the data is originally a geotiff, and I am reading it in using rioxarray, if that makes a difference.
Essentially, what seems to be happening is that the data is being shifted by the size of the kernel for each iteration. I have confirmed this by reading the same slice of the data in the array before and after running the closing operation. It also throws an error that "Axis Limits cannot be NaN or Inf". The output numpy array is the same size as the original, but parts of it have been cut off in one corner, and the other corner, it appears that a No Data value has been added(?) (value is -1.798e+308).
I am not sure why this is doing this. When I read an image in using imread, the same process seems to work just fine from what I can tell. I also cannot tell if the closing operation is even doing what it is supposed to be doing either. First glance, it only seems like it is shifting it. Code and image below.
import rioxarray as rxr
import xarray as xr
import cv2 as cv
import numpy as np
kSize = 15 #Kernel size
iters = 2 #number of iterations
#Binary list is a list of several binary images generated using xarray.where function
binaryCopy = binaryList[0].copy() #Copy one datarray from list of datarrays...prob should just put this all into one xr dataset, but oh well
inAttrs = binaryCopy.attrs.copy() #Copy attributes to read back in at the end
inDims = binaryCopy.dims
inCoords = binaryCopy.coords
kern = cv.getStructuringElement(cv.MORPH_RECT,(kSize,kSize)) #Create square kernel
print(binaryCopy.data.shape) #Print shape of data array (appears to just be a numpy array)
#I had tried this v first, but it did not work, so I tried the steps individually...same issue
#closed = cv.morphologyEx(binaryCopy.data, cv.MORPH_CLOSE, kern)
dilated = cv.dilate(binaryCopy.data, kern, iters)
closed = cv.erode(dilated, kern, iters)
newBinaryArray= xr.DataArray(closed,
coords=inCoords,
dims=inDims,
attrs=inAttrs)
fig, ax = plt.subplots(nrows=1, ncols=2, sharey=True)
#Plot the original first
binaryList[0].plot(ax=ax[0])
#Plot the closed data
newBinaryArray.plot(ax=ax[1])
plt.rcParams['figure.figsize'] = (15,8)
Before (left) and after (right) I run morphological closing. Notice the blue bar on the bottom and left of the image to right. This appears to be a no data value (-1.798e+308)

Ok, so it looks like my issue was the anchor. If you set argument anchor=(0,0), that seems to prevent the shift in the data. (I thought this should be (-1,-1) since other places on the internet seem to indicate that puts the anchor in the middle, but the (0,0) appears to work better for me.
Also, when I added borderType=cv.BORDER_REPLICATE (see here) as an argument to the morphologEx function (I just did it with that one instead of doing dilation and erosion), that prevented the extra strip that was inserted when the data was shifted from being an enormous "No Data" value and uses the data value at the border.

Voxelization of STL-file; writing into np.array

I would like to voxelise a .stl file and write it into an np.array. The resolution of the voxels should be adjustable.
Here is my code for this:
component_path = r"C:\Users\User\documents\components\Test_1.stl"
mesh = o3d.io.read_triangle_mesh(component_path)
voxel_grid = o3d.geometry.VoxelGrid.create_from_triangle_mesh(mesh, voxel_size = 3)
ply_path = "voxel.ply"
o3d.io.write_voxel_grid(ply_path, voxel_grid, True,True,True)
pcd = o3d.io.read_point_cloud(ply_path)
list_path = "list.xyz"
o3d.io.write_point_cloud(list_path, pcd)
Then I read the coordinate points from the list, write them into a 3D array and plot them. When plotting, the border is not displayed for certain voxel sizes, as can be seen in the image (although it is present in the original). Is there a solution for this so that it doesn't happen no matter what voxel size?
voxelized picture with missing border
In addition, the voxel size changes the maximum dimension. So the component originally has three times the length as it is shown here. How can this be adjusted? (If I just multiply a factor, the voxels stay small but pull the distances apart).
Is there perhaps a more reasonable way to write a voxelisation of a .stl file and put the centers of voxels into an np.array?

If anyone ever has the same problem and is looking for a solution:
This project worked for me: GitHub: stl-to-voxel
The model is then also filled. If the maximum dimension is known, you can determine the exact voxel size via the resolution.
Here is some code:
import stl_reader
import stltovoxel
import numpy as np
import copy
import os
import sys
input=r"C:\Users\user\Example.stl"
output=r"C:\Users\user\Test.xyz"
resolution = 50 #Resolution, into how many layers the model should be divided
stltovoxel.doExport(input, output, resolution)
Afterwards, you can read the coordinates from the list, write them into an array and process them further (quite normally).

Preserving the WCS information of a FITS file when rebinned

Aim : Rebin an existing image (FITS file) and write the new entries into a new rebinned image (also a FITS file).
Issue : Rebinned FITS file and the original FITS file seem to have mismatched co-ordinates (figure shown later in the question).
Process : I will briefly describe my process to shed more light. The first step is to read the existing fits file and define numpy arrays
from math import *
import numpy as np
import matplotlib.pyplot as plt
from astropy.visualization import astropy_mpl_style
from astropy.io import fits
import matplotlib.pyplot as plt
%matplotlib notebook
import aplpy
from aplpy import FITSFigure
file = 'F0621_HA_POL_0700471_HAWDHWPD_PMP_070-199Augy20.fits'
hawc = fits.open(file)
stokes_i = np.array(hawc[0].data)
stokes_i_rebinned = congrid(stokes_i,newdim,method="neighbour", centre=False, minusone=False)
Here "congrid" is a function I have used for near-neigbhour rebinning that rebins the original array to a new dimension given by "newdim". Now the goal is to write this rebinned array back into the FITS file format as a new file. I have several more such arrays but for brevity, I just include one array as an example. To keep the header information same, I read the header information of that array from the existing FITS file and use that to write the new array into a new FITS file. After writing, the rebinned file can be read just like the original :-
header_0= hawc[0].header
fits.writeto("CasA_HAWC+_rebinned_congrid.fits", rebinned_stokes_i, header_0, overwrite=True)
rebinned_file = 'CasA_HAWC+_rebinned_congrid.fits'
hawc_rebinned= fits.open(rebinned_file)
To check how the rebinned image looks now I plot them
cmap = 'rainbow'
stokes_i = hawc[0]
stokes_i_rebinned = hawc_rebinned[0]
axi = FITSFigure(stokes_i, subplot=(1,2,1)) # generate FITSFigure as subplot to have two axes together
axi.show_colorscale(cmap=cmap) # show I
axi_rebinned = FITSFigure(stokes_i_rebinned, subplot=(1,2,2),figure=plt.gcf())
axi_rebinned.show_colorscale(cmap=cmap) # show I rebinned
# FORMATTING
axi.set_title('Stokes I (146 x 146)')
axi_rebinned.set_title('Rebinned Stokes I (50 x 50)')
axi_rebinned.axis_labels.set_yposition('right')
axi_rebinned.tick_labels.set_yposition('right')
axi.tick_labels.set_font(size='small')
axi.axis_labels.set_font(size='small')
axi_rebinned.tick_labels.set_font(size='small')
axi_rebinned.axis_labels.set_font(size='small')
As you see for the original and rebinned image, the X,Y co-ordinates seem mismatched and my best guess was that WCS (world co-ordinate system) for the original FITS file wasn't properly copied for the new FITS file, thus causing any mismatch. So how do I align these co-ordinates ?
Any help will be deeply appreciated ! Thanks

I'm posting my answer in an astropy slack channel here should this be useful for others.
congrid will not work because it doesn't include information about the WCS. For example, your CD matrix is tied to the original image, not the re-binned set. There are a number of way to re-bin data with proper WCS. You might consider reproject although this often requires another WCS header to re-bin to.
Montage (though not a Python tool but has Python wrappers) is potentially another way.

As #astrochun already said, your re-binning function does not adjust the WCS of the re-binned image. In addition to reproject and Montage, astropy.wcs.WCSobject has slice() method. You could try using it to "re-bin" the WCS like this:
from astropy.wcs import WCS
import numpy as np
wcs = WCS(hawc[0].header, hawc)
wcs_rebinned = wcs.slice((np.s_[::2], np.s_[::2]))
wcs_hdr = wcs_rebinned.to_header()
header_0.update(wcs_hdr) # but watch out for CD->PC conversion
You should also make a "real" copy of hawc[0].header in header_0= hawc[0].header, for example as header_0= hawc[0].header.copy() or else header_0.update(wcs_hdr) will modify hawc[0].header as well.

Plotting a large number of stacked spheres generated from python-- mayavi? paraview and pyevtk? How to transmute .npy to .vtk?

I want to produce plots like this, except with many more particles. Matplotlib is woefully inadequate.
Right now I am using mayavi in python 3.5 running through a jupyter notebook. As I need to plot 5x10^5 spheres it will not be practical, since time is a limiting factor already at 2x10^4 spheres.
Here is my python code to produce the mayavi plot. I have a numpy array of values [a,r,x,y,z]. It's not relevant what the first quantity is for this problem.
"""VISUALIZATION WITH MAYAVI"""
#I think this is too slow to be practical.
#view particles with mayavi
import mayavi
from mayavi import mlab
%gui qt
def plot_sphere(p): #feed it p and get back a sphere \n",
t1,R,a,b,c = p
[phi,theta] = np.mgrid[0:2*np.pi:12j,0:np.pi:12j] #increase the numbers before j for better resolution but more time
x = R*np.cos(phi)*np.sin(theta) + a
y = R*np.sin(phi)*np.sin(theta) + b
z = R*np.cos(theta) + c
return mlab.mesh(x, y, z)
#run s over all particles and plot it
def view(particles):
for p in particles:
plot_sphere(p)
view(spheres)
This code produces plots like this:
I have been told I should look into writing my numpy arrays to .vtk files using evtk, then visualizing these in paraview. I downloaded paraview and read this, but perhaps my version of python is limiting me? First, install pyevtk-- okay:
I tried conda install -c opengeostat pyevtk=1.0.0, but it fails due to incompatibility with my python version. I looked for details but could not find any.
Next I downloaded the repository [here][https://pypi.python.org/pypi/PyEVTK/1.0.0], then used pip to install it successfully.
Next I put evtk.py, vtk.py, hl.py, and xml.py, and tried some examples in the repository-- none of them work. Seemingly there is some problem with
from .vtk import *
type commands. I tried replacing all of these in the four .py files with
from evtk import vtk
from vtk import *
and such, but no luck. Long story short, I can't get pyevtk working to export my numpy arrays as .vtk files. I could use some help in this regard, or better yet I would love a different option to get my numpy arrays rendered by paraview. Any help is appreciated !

Ok, I solved my own problem. This image is made using paraview, after converting numpy arrays to a .vtu object using pyevtk.
Out of the box, the repository did not work, there was some problem with importing the modules inside the four .py files, so I modified them all. Instead of from .vtk import *, I changed it to from vtk import *, and so on, in every module in the library. evtk.py was not able to import a class from xml.py, so I just copied it and pasted, then deleted xml.py. After some tinkering and clueless modifying to make the errors go away, eventually it worked.

How do I adjust figtext line spacing for when I have special characters in the text?

I'm trying to add an subscript to some text for my figure caption. When I do this, the line with the subscript (line 2) moves up into the line above it (line 1). Is there easy any way around this issue? Using plt.text(...) seems like it could be tedious and time consuming.
If I don't use any special characters (i.e. $_{Sun}$) with figtext it works perfectly for me.
I also had something similar happen when I was doing the same thing with legend labels, so I'm guessing that any solution to this problem will solve that as well.
Below is the relevant code which I used:
import numpy as np
import matplotlib.pyplot as plt
from pylab import *
plt.figure(1)
figtext(0.02,0.02,"Standard composition model, Spherical geometry\nT = 5000 K, log(g) = 3.0, Microturbulance = 2, Mass = 2M$_{Sun}$\nThe x-axis covers the range of the K-bandpass\n")
plt.show()
Thank you in advanced!

This should have be a comment, if picture can be embedded in a comment. Anyway, this appears to be a bug in the interactive backend (in the following case, it is the MacOSX backend), when you save the plot into a .png or .pdf, etc, it will be rendered correctly.
Interactive:
Save it as .png or use plt.savefig():