Import and inspect data

Creating an azimuthal integrator

The txs software makes use of pyFAI to perform the azimuthal integration on the images.

Hence, the first step to process your data is to create a so-called AzimuthalIntegrator object.

The azimuthal integrator takes information about the detector and the geometry of the setup to compute the position of the pixels corresponding to the same scattering angle.

In txs, the utils.get_ai() helper function is provided to ease the creation of this object:

import txs

ai = txs.get_ai(
    15e3,  # the incoming X-ray energy (in eV)
    0.15,  # the sample-to-detector distance (in m)
    center=(968, 969),  # the beam center on the detector (in pixels)
    detector='rayonix',  # the detector being used
    binning=(2, 2),  # the pixel binning being used
)

Note

The creation of the AzimuthalIntegrator object is expected to be automatized in the future on ID09 at the ESRF by using the metadata that are recorded during the experiment.

Using .edf files (old setup)

Prior to the installation of BLISS software at the ESRF, the detector images were recorded in the .edf file format.

The raw images can be visualized using the FabIO viewer.

To perform the azimuthal integration for these experiments, the azav.integrate1d_dataset() function is to be used by providing the data folder as well as the file extension as such:

azav = txs.int1d_dset(
    "<data_folder>",  # the scan folder containing the images
    ai,  # the previously created AzimuthalIntegrator object
    extension="edf",
    method='csr_ocl',  # faster, multi-threaded integration
    sample_thickness=1.2e-3,  # sample thickness in m (for transmission)
    sample_material='H2O'
)

Using BLISS dataset

The folder hierarchy in BLISS and associated HDF5 files makes possible to access the data in multiple ways. The recommanded method is to import the ‘dataset’ HDF5 file using the datasets.BlissDataset class:

dset = txs.BlissDataset(
    "path_to/<sample_name>_<dataset_number>.h5"
)

The resulting dset object is an iterator over the scans and the images which contains also metadata related to the experiment:

dset.scans  # -> ['1.1', '1.2', '2.1', '3.1',...]

# to obtain the metadata for scans '1.2'
dset.metdata(1, 2)

Hence, for a given scan ‘2.1’, the images can be accessed and plotted as follows:

import matplotlib.pyplot as plt

plt.imshow(dset["2.1"][0])  # plot the first image
plt.imshow(dset["2.1"][4])  # plot the fifth's image

The azimuthal integration is done similarly as above by using the azav.integrate1d_dataset() function:

azav = txs.azav.integrate1d_dataset(
    dset["2.1"].folder,  # the folder of scan 2.1 containing the images
    extension='h5',  # default value, so it can be ommited
    ai,  # the previously created AzimuthalIntegrator object
    method='csr_ocl',  # faster, multi-threaded integration
    sample_thickness=1.2e-3,  # sample thickness in m (for transmission)
    sample_material='H2O'
)

The resulting azav object

The azav object obtained above is a Python dictionary that contains the 1D scattering patterns with their associated errors as well as other information about the data and the methods used to process them.

The scattering vector magnitude q, the intensities and errors are contained in the q, i and e keys of the azav dictionary, respectively.

Plotting possibilities

Some plotting functions are available to quickly plot useful information. For the azimuthal integration, the function plot.plot_azim_regroup() allows to view the projection map used by the azimuthal integrator:

projection map of the azimuthal integrator

Additional options

Using a mask