Odin: Orca image normalization workflow#

This notebook shows how to use the workflow to compute normalized images recorded by the Orca detector on the ODIN instrument.

[1]:

import ess.odin.data  # noqa: F401
from ess import odin
from ess.imaging.types import *
import scipp as sc
import plopp as pp

%matplotlib widget

Workflow setup#

[2]:

wf = odin.OdinOrcaWorkflow()

wf[Filename[AllRuns]] = odin.data.odin_lego_images()

wf[NeXusDetectorName] = odin.nexus_name("orca")

wf[MaskingRules] = {}  # No masks to begin with
wf[UncertaintyBroadcastMode] = UncertaintyBroadcastMode.upper_bound

Downloading file 'ymir_lego_odin.hdf' from 'https://public.esss.dk/groups/scipp/ess/odin/4/ymir_lego_odin.hdf' to '/home/runner/.cache/ess/odin'.

[3]:

wf.visualize(NormalizedImage, compact=True, graph_attr={"rankdir": "LR"})

[3]:

../_images/odin_odin-orca-image-normalization_4_0.svg

Adding masks#

If we want to mask some part of the image, we update the masking rules. For example, here we mask the upper part of the image:

[6]:

wf[MaskingRules] = {'y_pixel_offset': lambda x: x > sc.scalar(0.082, unit='m')}

pp.slicer(wf.compute(NormalizedImage), autoscale=False)

[6]:

Intermediate results#

We can also inspect intermediate results, which is useful for debugging:

[7]:

results = wf.compute(
    [
        FluxNormalizedDetector[SampleRun],
        FluxNormalizedDetector[OpenBeamRun],
        BackgroundSubtractedDetector[SampleRun],
    ]
)

fig = pp.tiled(2, 2, hspace=0.3, wspace=0.3)
fig[0, 0] = results[FluxNormalizedDetector[SampleRun]]['time', 0].plot(
    title='Sample (proton-charge normalized)'
)
fig[0, 1] = results[FluxNormalizedDetector[OpenBeamRun]]['time', 0].plot(
    title='Open beam (proton-charge normalized)'
)
fig[1, 0] = results[BackgroundSubtractedDetector[SampleRun]]['time', 0].plot(
    title='Background subtracted sample'
)
fig[1, 1] = image['time', 0].plot(title='Final image')
fig

[7]: