Utilities

Lower-level building blocks used across DarSIA. Most users will not need these directly, but they are public and stable.

Grids and finite volumes

Grid(shape[, voxel_size])	Tensor grid.
generate_grid(image)	Build a Grid matching an image's voxel count and voxel size.
FVDivergence(grid)	Finite volume divergence operator.
FVMass(grid[, mode, lumping])	Finite volume mass matrix.
FVFullFaceReconstruction(grid)	Reconstruct full face fluxes from their normal and tangential components.
FVTangentialFaceReconstruction(grid)	Projection of normal fluxes on grid onto tangential components.
cell_to_face_average(grid, cell_qty, mode)	Project scalar cell quantity to scalar face quantity via averaging.
face_to_cell(grid, flat_flux[, pt])	Reconstruct the vector fluxes on the cells from normal fluxes on the faces.

Points, voxels and coordinates

Coordinate([input_array])	Cartesian coordinate.
CoordinateArray([input_array])	Container for collection of Coordinate.
Voxel(input_array[, matrix_indexing])	Voxel coordinate.
VoxelArray(input_array[, matrix_indexing])	Container for collection of Voxel.
VoxelCenter(input_array[, matrix_indexing])	Voxel center coordinate.
VoxelCenterArray(input_array[, matrix_indexing])	Container for collection of VoxelCenter.
make_coordinate(pts)	Quick-access constructor for Coordinate or CoordinateArray.
make_voxel(pts[, matrix_indexing])	Quick-access constructor for Voxel or VoxelArray.
make_voxel_center(pts[, matrix_indexing])	Quick-access constructor for VoxelCenter or VoxelCenterArray.
to_coordinate(self[, coordinatesystem])	Conversion of point to Coordinate.
to_voxel(self[, coordinatesystem])	Conversion of point to Voxel.
to_voxel_center(self[, coordinatesystem])	Conversion of point to VoxelCenter.

Derivatives

forward_diff(img, axis[, dim, h])	Forward difference of image matrix in direction of axis.
backward_diff(img, axis[, dim, h])	Backward difference of image matrix in direction of axis.
laplace(img[, axis, dim, h, diffusion_coeff])	Laplace operator with homogeneous boundary conditions.

Interpolation

polynomial_interpolation(measurements, ...)	Determine a voxeled spatial map from measurements through polynomial interpolation.
interpolate_to_image(data, image[, method])	Interpolate data to image.
interpolate_to_image_from_csv(csv_file, key, ...)	Interpolate data from CSV to image.
interpolate_measurements_2d(measurements, ...)	Build a voxelised spatial map from scattered measurements by RBF interpolation.
illumination_interpolation(measurements, ...)	Determine a voxeled spatial map from measurements through polynomial interpolation.

Kernels

BaseKernel()	Abstract base class for kernel.
GaussianKernel([gamma])	Gaussian kernel.
LinearKernel([a])	Linear kernel.

Linear solvers

Solver([maxiter, tol, dim, mass_coeff, ...])	Abstract base class for solvers for the problem: mass_coeff * x - diffusion_coeff * laplace(x) = rhs
CG([maxiter, tol, dim, mass_coeff, ...])	Conjugate gradient solver based on scipy's cg for the problem: mass_coeff * x - diffusion_coeff * laplace(x) = rhs
Jacobi([maxiter, tol, dim, mass_coeff, ...])	Jacobi solver for the problem: mass_coeff * x - diffusion_coeff * laplace(x) = rhs
MG([depth, smoother_iterations, maxiter, ...])	Muiltilvel solver for the problem: mass_coeff * x - diffusion_coeff * laplace(x) = rhs.

Approximation and acceleration

AndersonAcceleration([dimension, depth, restart])	Anderson acceleration for fixed-point methods.
ApproximationSpace()	Abstract class for approximation spaces.
LinearApproximation(space, dim[, domain])	Linear combination of any approximation space.
PolynomialApproximationSpace(degree)	Global polynomial approximation space in 2d.
RadialPolynomialApproximationSpace(degree[, ...])	Global radial polynomial approximation space in 2d.

Detection

detect_color(img, color[, tolerance])	Detects a color in an image.
detect_value(img, value[, tolerance])	Detects a value in an image.
detect_closest_point(points, target)	Detects the point closest to a target.
orthogonal_colors(color)	Returns the orthogonal colors to the input color.
monochromatic_concentration_analysis(img, color)	Performs monochromatic concentration analysis on an image.

Data types and miscellany

convert_dtype(img, dtype)	Convert image to the specified dtype.
Format(*values)	Enumeration of formats for data.
FeatureDetection()	Class containing two class methods, once detecting features, and matching features.
ConvergenceStatus(*values)	Enum to represent the convergence status of an iterative solver.
extract_characteristic_data(signal[, mask, ...])	Extract representative colours from an image for given patches.
hsv_spectrum(img, roi[, bins])	Plot histograms for all HSV components present in a ROI of an image.
bounding_box(voxels[, padding, max_size])	Determine bounding box for a set of given coordinates.
bounding_box_inverse(bounding_box)	Inverse of bounding_box().
perimeter(box)	Returns the perimeter of a box.
random_patches(mask, width, num_patches)	Utility to extract random patches from a mask.
sort_quad(pts)	Sorts the four points of a quadrilateral in a clockwise fashion.
full_like(image, fill_value[, mode, dtype])	Analog of np.full_like but for darsia.Image objects.
ones_like(image[, mode, dtype])	Analog of np.ones_like but for darsia.Image objects.
zeros_like(image[, mode, dtype])	Analog of np.zeros_like but for darsia.Image objects.