openformat.mims package¶

openformat.mims.apply_mask(data, mask, replacement_values)¶

Replace all values in data by replacement_values where mask is True.

Parameters

data (np.ndarray) – data tensor with shape (n, p, w, h), where n is the number of frames, p is the number of detectors, w is the width, and h is the height
mask (np.ndarray) – binary mask of hot pixels with shape (n, w, h)
replacement_values (np.ndarray) – tensor of replacement values with the same shape as data, a list of the same length as non-zero elements in mask, or a scalar

openformat.mims.apply_translations(data, translations, padding=8)¶

Apply the sequence of translations to the data.

Parameters

data (np.ndarray) – data tensor with shape (n, p, w, h), where n is the number of frames, p is the number of detectors, w is the width, and h is the height
translations (np.ndarray) – sequence of translations that should be applied to align the images

Returns

translated – translated data tensor with shape (n, p, w + 2 * padding, h + 2 * padding)

Return type

np.ndarray

openformat.mims.infer_hot_pixels(data, num_outliers, kernel_size=3, outlier_factor=3, min_count=1, max_repeat_fraction=0.1, num_nonzero=None, return_filtered=False)¶

Identify hot pixels due to correlated noise.

Parameters

data (np.ndarray) – data tensor with shape (n, p, w, h), where n is the number of frames, p is the number of detectors, w is the width, and h is the height
num_outliers (int) – number of times a pixel needs to be labelled as an outlier in different detectors to be considered correlated noise
kernel_size (int) – size of the kernel used for median filtering
outlier_factor (float) – factor multiplying the median-filtered image to define the threshold that serves to flag a pixel as an outlier
min_count (float) – minimum count used to determine the threshold if the median-filtered image is smaller
max_repeat_fraction (float) – maximum fraction of frames in which a pixel is allowed to be a hot pixel. If a pixel is a hot pixel candidate in more than max_repeat_fraction of all frames, it is not considered a hot pixel.
num_nonzeros (int) – number of detectors in which a hot pixel candidate must be non-zero. Default is all detectors.
return_filtered (bool) – whether to return the median-filtered image

Returns

mask – binary mask of hot pixels with shape (n, w, h)

Return type

np.ndarray

openformat.mims.infer_translations(data, padding=8, method='sequential')¶

Infer translations between successive frames to increase sharpness of aggregated nanoSIMS images.

Parameters

data (np.ndarray) – data tensor for a single detector with shape (n, w, h), where n is the number of frames, w is the width, and h is the height
padding (int) – padding applied to the data tensor before computing correlations
method (str) – method used to align different frames. ‘sequential’ aggregates frames in order and aligns successive frames with the current best aggregate.

Returns

translations – sequence of translations that should be applied to align the images

Return type

np.ndarray

openformat.mims.load_mims(filename, byte_order='<', roll_data=True)¶

Load a multi-isotope imaging mass spectrometry (MIMS) file.

Note

This function returns a dictionary with keys listed in the Returns section.

Parameters

filename (str) – path to the MIMS file with .im extension
byte_order (str) – byte order of the encoded binary data
roll_data (bool) – whether to roll image data along the last two dimenions. Rolling the image fixes an issue with Cameca’s encoding of pixel positions.

Returns

def_analysis (Def_analysis) – information about the analysis performed. See [fmt] for details.
def_analysis_bis (Def_analysis_bis) – additional information about the analysis peformed. See [fmt] for details.
mask_im (Mask_im) – image acquisition information. See [fmt] for details.
tab_mass (list[Tab_mass]) – information about the detectors and masses. See [fmt] for details.
cal_cond (Cal_cond) – information about the conditions for data acquisition. See [fmt] for details.
poly_list (Poly_list) – header for element information. See [fmt] for details.
polyatomique (list[Polyatomique]) – sequence of element information containers. See [fmt] for details.
mask_nano (Mask_nano) – data acquisition information. See [fmt] for details.
tab_Bfield_nano (tab_Bfield_nano) – information about the magnetic field. See [fmt] for details.
anal_param_nano (Anal_param_nano) – parameters for the analysis. See [fmt] for details.
anal_param_nano_bis (Anal_param_nano_bis) – additional parameters for the analysis. See [fmt] for details.
_anal_param (bytes) – unparsed value
_setup_soft (bytes) – unparsed value
header_image (Header_image) – information regarding images. See [fmt] for details.
data (np.ndarray) – data tensor with shape (n, p, w, h), where n is the number of frames, p is the number of detectors, w is the width, and h is the height

References

fmt(1,2,3,4,5,6,7,8,9,10,11,12): Alain Morgand. “Format File N50 V7”, https://www.dropbox.com/sh/gyss2uvv5ggu2vl/AACkLTMRfKjZj3PEt3H_PxY_a/manual?preview=Format+File+N50+V7.doc

openformat.mims.process_image_data(filename, alignment_detector, padding=8, load_mims_kwargs=None, remove_hot_pixels_kwargs=None, infer_translations_kwargs=None)¶

Process a raw .im file, remove hot pixels, and align the frames of each image. Returns a dictionary of cleaned aligned data and relevent information.

Note

This function returns a dictionary with keys listed in the Returns section.

Parameters

filename (str) – path to the MIMS file with .im extension
alignment_detector (str) – the element or element combination to use for frame alignment e.g. ‘12C 14N’ or ‘32Se’
padding (int) – number of border pixels to allow image shifts without data loss
load_mims_kwargs (dict, optional) – dictionary with keywords passed to the load_mims() call used to load the raw data
remove_hot_pixel_kwargs (dict, optional) – dictionary with keywords passed to the remove_hot_pixels() call used to remove hot pixels from the dataset
infer_translations_kwargs (dict, optional) – dictionary with keywords passed to the infer_translations() call used to calculate image alignment shifts

Returns

total_lookup (dict) – cleaned, aligned detector data keyed by detector name
detector_names (list) – sequence of detector names
cleaned (list) – sequence of detector data with hot pixels replaced by median-filtered pixels
translations (np.array) – sequence of horizontal and vertical translations to align different scans
aligned (list) – sequence of aligned detector data
image_shape (tuple) – width and height of the image in pixels
image_size (tuple) – width and height of the image in micrometers
pixel_size (float) – size of each pixel in micrometers
alignment_detector_index (int) – index of detector used for aligning different scans
se_detector_index (int) – index of the detector collecting secondary electron data

openformat.mims.remove_hot_pixels(data, num_outliers=3, kernel_size=3, outlier_factor=1.999, min_count=0.5, max_repeat_fraction=0.1, num_nonzero=None)¶

Remove hot pixels due to correlated noise.

Parameters

data (np.ndarray) – data tensor with shape (n, p, w, h), where n is the number of frames, p is the number of detectors, w is the width, and h is the height
num_outliers (int) – number of times a pixel needs to be labelled as an outlier in different detectors to be considered correlated noise
kernel_size (int) – size of the kernel used for median filtering
outlier_factor (float) – factor multiplying the median-filtered image to define the threshold that serves to flag a pixel as an outlier
min_count (float) – minimum count used to determine the threshold if the median-filtered image is smaller
max_repeat_fraction (float) – maximum fraction of frames in which a pixel is allowed to be a hot pixel. If a pixel is a hot pixel candidate in more than max_repeat_fraction of all frames, it is not considered a hot pixel.
num_nonzeros (int) – number of detectors in which a hot pixel candidate must be non-zero. Default is all detectors.

Returns

cleaned – cleaned data tensor with the same shape as data with hot pixels replaced by median-filtered pixels

Return type

np.ndarray

openformat.mims.structures module¶

This module contains structures to load MIMS data. See 1 for general information and 2 for a specific reference.

1: Utrecht University. “nanosims-wiki”, http://nanosims.geo.uu.nl/nanosims-wiki/
2: Alain Morgand. “Format File N50 V7”, https://www.dropbox.com/sh/gyss2uvv5ggu2vl/AACkLTMRfKjZj3PEt3H_PxY_a/manual?preview=Format+File+N50+V7.doc

class openformat.mims.structures.Anal_param_nano¶