Set of functions for dealing with data from the electron microscopes at Utrecht University
The full API documentation can be found here
- Created by: Maarten Bransen
- Email: m.bransen@uu.nl
- Version: 3.3.0
This package can be installed directly from GitHub using pip:
pip install git+https://github.com/UU-SCMB/scm_electron_microscopes
When using the Anaconda distribution, it is safer to run the conda version of pip as follows:
conda install pip
conda install git
pip install git+https://github.com/UU-SCMB/scm_electron_microscopes
To update to the most recent version, use pip install with the --upgrade flag set:
pip install --upgrade git+https://github.com/UU-SCMB/scm_electron_microscopes
Note that since version 2.0.0 Pytesseract is no longer required as dependency
For these microscopes use the tia class (aliasses tecnai and talos are just provided for convenience, they are identical). In TIA, data must be exported as a .tif file. To import data in python, create a class instance using the filename of the TEM image. This automatically loads the image, which is available as numpy.array as the image attribute:
from scm_electron_microscopes import tia
import matplotlib.pyplot as plt
em_data = tia('myimage.tif')
image = em_data.image
plt.figure()
plt.imshow(image,cmap='Greys_r')
plt.axis('off')
plt.show()
Note that this is only the image data, with the scalebar stripped off. The imagedata for the scale bar is available through the scalebar attribute, but more likely you are interested in the pixel size which can be determined semi or fully automatically using tia.get_pixelsize():
pixelsize,unit = em_data.get_pixelsize()
Other information about the microscope is read from the file and can be printed with tia.print_metadata(). Files can be exported with a nicer and customizable scalebar using the tecnai.export_with_scalebar() function.
The native .emd file format used by Velox can be opened using the velox class, and images / videos contained within can be extracted using the get_image method. By default, the actual image data is not loaded into memory (as this would be ineffecient for large datasets). Instead, a get_frame method is available to extract specific video frames as well as a get_data method for explicitely loading the full image data:
from scm_electron_microscopes import velox
em_data = velox('myimage.emd').get_image(0)
myarray = em_data.load_data()
For the now long defunct Tecnai 10, the analySIS program was used, which was the predecessor to TIA and looks similar but inconveniently stored metadata differently. For this the SIS class is available.
For the Helios SEM use the helios class, which unlike the tecnai and talos classes does not load the image into memory by default, such that it is possible to quickly read out metadata of e.g. a slice-and-view series without having to load all the image data. Image data is available through the load_image() function:
from scm_electron_microscopes import helios
em_data = helios('myimage.tif')
image = em_data.get_image()
Pixel sizes written in the metadata and available as a shortcut through a function:
(pixelsize_x,pixelsize_y),unit = em_data.get_pixelsize()
Other information about the microscope is read from the file and can be printed in a human-readable format with helios.print_metadata() or parsed directly through helios.load_metadata(). Files can be exported with a nicer and customizable scalebar using the helios.export_with_scalebar() function.
For any of the Phenom SEMs use the phenom class, which works similar as described above for the helios class.
This microscope is no longer around, but for older data you can use the xl30sfeg class
Some utility functions for e.g. plotting a histogram are included in the util class
tecnaiandtalosclass names are deprecated as they are renamed to thetiaclass to prevent confusion with data recorded with the TIA vs the Velox software- a new velox class is available that reads the native .emd file format of the more modern Velox software installed on the Talos microscopes
Note that this version has some backwards incompatible changes:
load_imagefunctions have been renamed toget_imageload_metadatafunctions have been renamed toget_metadatatecnai.get_pixelsizeno longer uses text recognition to read the scale bar. This is faster and removes Pytesseract and the Tesseract OCR as dependencies, but yields slightly different (and more accurate) values. The old scaling method is available throughget_pixelsize_legacy.- all
get_metadatafunctions now return an xml.etree.Elementtree object by default, rather than a dictionary which was used for some classes.