Single image analysis (darsia.single_image_analysis)#

Post-processing tools for physical data.

Submodules#

darsia.single_image_analysis.contouranalysis module#

Module containing analysis tools for segmented images.

This includes measuring lengths of contours, weighted sums (generalized mass analysis).

class ContourAnalysis(verbosity=False)[source]#

Bases: object

Contour analysis object.

contours()[source]#

Determine contour of loaded labeled image.

Returns:

list of contours, where each contour is given as an

array of pixels.

Return type:

list[np.ndarray]

fingers(direction=array([0., -1.]))[source]#

Determine local extrema of the contour, where the extremality is defined by a direction.

Parameters:

  • contours (np.ndarray | None) – contours to analyze. If None, contours are determined from the mask; default is None.

  • direction (np.ndarray) – direction vector with orientation

Returns:

pixels of peaks. array: pixels of valleys.

Return type:

array

length()[source]#

Determine length of loaded labeled image.

Returns:

length of the interface between values of interest and others.

Output in metric units.

Return type:

float

load(img, mask, roi=None, fill_holes=True)[source]#

Read labeled image and restrict to values of interest.

Parameters:

  • img (Image) – labeled image.

  • roi (array, optional) – set of points defining a box.

  • values_of_interest (int, list of int, optional) – label values of interest.

  • fill_holes (bool) – flag controlling whether holes in labels are filles.

load_labels(img, roi=None, values_of_interest=None, fill_holes=True)[source]#

Read labeled image and restrict to values of interest.

Parameters:

  • img (Image) – labeled image.

  • roi (array, optional) – set of points defining a box.

  • values_of_interest (int, list of int, optional) – label values of interest.

  • fill_holes (bool) – flag controlling whether holes in labels are filles.

number_peaks()[source]#

Determine number of peaks.

Returns:

number of peaks.

Return type:

int

plot_finger_peaks(img, peaks_pixels, roi=None, contours=None, path=None, show=True, **kwargs)[source]#

Plot peaks on top of the provided image.

Parameters:

  • img (darsia.Image) – image to plot on.

  • peaks_pixels (np.ndarray) – pixels of peaks.

  • contours (list[np.ndarray], optional) – contours to plot; if None, no contours are plotted; default is None.

  • roi (darsia.CoordinateArray | None) – region of interest. If provided, peaks are translated to the top left corner of the ROI; default is None.

  • path (Path, optional) – path to save the plot; if None, no saving is performed.

  • show (bool) – flag controlling whether the plot is shown; default is True.

  • **kwargs – additional keyword arguments for plotting. - color (str): color for the peaks; default is “r”. - size (int): size for the peaks; default is 20.

verbosity#

Vebosity flag.

class ContourEvolutionAnalysis(verbosity=False)[source]#

Bases: object

add(peaks, valleys, time=None)[source]#
find_paths(reset=True)[source]#

Find paths in the peaks and valleys.

plot(img=None, roi=None)[source]#
plot_paths(img=None, roi=None, path=None, show=False)[source]#
class PathUnit(time, peak, position)#

Bases: tuple

count(value, /)#

Return number of occurrences of value.

index(value, start=0, stop=9223372036854775807, /)#

Return first index of value.

Raises ValueError if the value is not present.

peak#

Alias for field number 1

position#

Alias for field number 2

time#

Alias for field number 0

contour_length(img, roi=None, values_of_interest=None, fill_holes=True, verbosity=False)[source]#

Calculation of the contour length of a segmented region.

Parameters:

  • img (darsia.Image) – segmented image with boolean or integer values.

  • roi (np.ndarray) – set of points, for which a bounding box defines a ROI.

  • values_of_interest (int or list of int) – only active if integer-valued image provided; defines the values of interest, i.e., which part of the image is treated as active.

  • fill_holes (bool) – flag controlling whether holes in the determined mask are filled before the contour length is computed; if not, holes are treated as contour; default is True.

  • verbosity (bool) – flag controlling whether intermediate results are plotted; default is False.

Returns:

contour length in metric units based on the coordinate system of the

input image.

Return type:

float