parameter documentation

README.md

@@ -41,9 +41,9 @@ Detailed description of paremeters can be found in
 <img src="https://raw.githubusercontent.com/mjirik/teigen/master/graphics/teigen_volume_fraction_22_unconnected_n335_paraview.png" width="200">
 <img src="https://raw.githubusercontent.com/mjirik/teigen/master/graphics/teigen_volume_fraction_48_overlap4_n400_paraview.png" width="200">
 <img src="https://raw.githubusercontent.com/mjirik/teigen/master/graphics/teigen_volume_fraction_89_porosities_n400_paraview.png" width="200">
-On left image is shown Unconnected tubes generator with volume fraction 22 %. Number of elements is 335.
+On the first image is shown Unconnected tubes generator with volume fraction 22 %. Number of elements is 335.
 Middle image contain tubes with overlap. Volume fraction 48 %. Number of elements is 400.
-Right image visualize porosities with no overlap. Volume fraction 89 %, number of elements is 400.
+Bottom image visualize porosities with no overlap. Volume fraction 89 %, number of elements is 400.
 
 
 # Screenshots

teigen/measurement.py

@@ -10,7 +10,9 @@
 import skimage.measure
 
 
-def surface_measurement(volume, voxelsize, level=1.0, return_vertices_and_faces=False, **kwargs):
+def surface_measurement(volume, voxelsize, level=None, return_vertices_and_faces=False, **kwargs):
+    if level==None:
+        level = (np.max(volume) + np.min(volume)) * 0.5
     vertices, faces = skimage.measure.marching_cubes(volume, level=level, spacing=voxelsize)
     surface_area = skimage.measure.mesh_surface_area(verts=vertices, faces=faces)
 

user_manual.md

@@ -2,128 +2,66 @@
 The basic concept of the algorithm includes the definition of objects to generate, the generation of the framework of the fiber structure, the surface representation, the quantitative description, the volume representation, and finally the file storage. 
 
 The basic element used in this task is a tube. It is a cylindrical body that ends with hemispheres. The user can set parameters for object length, object radius, and parameters that affect the direction and isotropy of objects. 
-
-## Generators
-
-Generátor Unconnected Tubes (patrně přejmenovat na něco jako Independent Tubes).
-V každé iteraci je vygenerován náhodný bod, vektor, délka a poloměr.
-Náhodný bod je vždy uvnitř definované oblasti. Vektor je dán směrovými parametry. Délka a poloměr denzitní funkcí se zvolenými parametry. Dále jsou určeny kolize tohoto objektu s objekty, které byly do oblasti umístěny dříve. Pokud kolize není, dojde ke vložení objektu do oblasti a pokračuje se do další iterace. Pokud dochází ke kolizi, generátor se osmkrát pokusí zkrátit délku na 75% a pokaždé testuje kolize. V případě úspěchu dojde ke vložení objektu do oblasti, jinak se pokračuje na další iteraci. 
-
-Postup generování je ukončen, pokud je překročena hodnota volume_fraction, nebo element_number, nebo maximální dovolený počet iterací (v tisících) přesáhne hodnotu maximum_1000_iteration_number.
-
-
-# Volume and surface measurement
-
-Pro objekty bez dotyku lze povrch a objem stanovit analyticky. 
-*vzorec*
-V případě objektů s dotykem je využívána numerická metoda výpočtu povrchu a objemu založená na triangulaci tvaru okraje objektu. Na základě zvoleného resolution je válcová část objektu reprezentována pravidelným mnohoúhelníkem s počtem hran rovným resolution. Mnohoúhelník je vepsán do kružnice o zvoleném poloměru. Kuloplochy na konci objektu jsou repezentovány mnohostěnem vytvořeným z trojúhelníků. Vrcholy trojúhelníků se nacházejí na průsečíku pomyslných rovnoběžek a polendíků. Počet rovnoběžek i poledníků je roven resolution.
-
-Vzhledem k tomu, že je triangulovný objekt konvexní a je zcela vepsán do modelovaného tvaru, jeho objem a povrch je menší, než analytická hodnota. Tento nedostatek je kompenzován kompenzováním poloměru válcové a kulové plochy. Volba poloměrů je dána následujícími možnostmi.
-
-inscribed: vepsaný mnohostěn - výchozí možnost
-circumscribed: opsaný mnohostěn
-average: poloměr mnohostěnu je mezi vepsanou a opsanou varianou
-
-cylinder volume: poloměr je kompenzován na základě analytické ekvivalence objemu útvaru s podstavou pravidelného mnohoúhelníku. Stejnou hodnotou je kompenzován i poloměr kulových ploch. 
-cylinder surface: poloměr je kompenzován na základě analytické ekvivalence povrchu útvaru s podstavou pravidelného mnohoúhelníku. Stejnou hodnotou je kompenzován i poloměr kulových ploch. 
-
-cylinder volume + sphere error: poloměr kulové plochy je kompenzován na základě změřené chyby dané rozdílem objemu koule a mnohostěnu. Válcová oblast je kompenzována stejně jako u cylinder volume
-
-cylinder surface + sphere error: poloměr kulové plochy je kompenzován na základě změřené chyby dané rozdílem povrchu koule a mnohostěnu. Válcová oblast je kompenzována stejně jako u cylinder surface
-
-cylinder surface + sphere error: poloměr kulové plochy je kompenzován na základě změřené chyby dané rozdílem povrchu koule a mnohostěnu. Válcová oblast je kompenzována stejně jako u cylinder surface
-
-cylinder surface + sphere error + join error: poloměr válce je kompenzován stejně jako u cylinder surface + sphere error. Stejně tak kulová plocha. Navíc je poloměr kulové plochy kompenzován naměřenou chybou spoje mezi kulovou plochou a válcovou plochou
-
-
-cylinder volume + sphere error + join error: poloměr válce je kompenzován stejně jako u cylinder volume + sphere error. Stejně tak kulový objem. Navíc je poloměr kulového objemu kompenzován naměřenou chybou spoje mezi kulovým objemem a válcovým objemem
+Four different generators can be used.
 
 
 
 
 #Input settings
-appearance: 
-show_surface: 
-areasampling:
-areasize_mm: 
-areasize_px: 
-voxelsize_mm: 
-filepattern: 
-filepattern_series_number: 
-generator_id: 
-generators:
-- - - Cylinder generator
-      - - [element_number
-        - [uniform_radius_distribution
-        - [normal_radius_distribution
-        - [fixed_radius_distribution
-        - [radius_distribution_minimum
-        - [radius_distribution_maximum
-        - [radius_distribution_mean
-        - [radius_distribution_standard_deviation
-        - - intensity_profile_radius
-          - &id001 [0.4, 0.7, 1.0, 1.3]
-        - - intensity_profile_intensity
-          - &id002 [195, 190, 200, 30]
-        - [random_generator_seed
-  - - Gensei generator
-    - !!python/object/apply:collections.OrderedDict
-      - - [n_objects
-  - - Cylinder continues
-    - !!python/object/apply:collections.OrderedDict
-      - - [element_number
-        - [uniform_radius_distribution
-        - [normal_radius_distribution
-        - [fixed_radius_distribution
-        - [radius_distribution_minimum
-        - [radius_distribution_maximum
-        - [radius_distribution_mean
-        - [radius_distribution_standard_deviation
-        - - intensity_profile_radius
-          - *id001
-        - - intensity_profile_intensity
-          - *id002
-        - [random_generator_seed
-  - - Unconnected cylinders: This generator produces objects in shape of cylinders of known length and radius. The cylinders end with hemispheres with the same radius attached to each end of the cylinders. If the cylinder length is set to zero, only the two endpoint hemispheres are generated, which results in generating a sphere. The objects do not interfere with each other. 
-element_number: number of elements at which the generator stops generating further elements, even before reaching the expected volume fraction
-uniform_radius_distribution: all the values of radius within the given limits appear with the same probability 
-normal_radius_distribution: the values of radius will be generated using the normal (Gaussian) parametric distribution
-fixed_radius_distribution: only the mean value of radius will be used
-radius_distribution_minimum: the lower limit of the radius of cylinders that are to be generated
-radius_distribution_maximum: the upper limit of the radius of cylinders that are to be generated
-radius_distribution_mean: the mean or expectation of the normal (Gaussian) parametric distribution of the radius
-radius_distribution_standard_deviation: the standard deviation of the normal (Gaussian) parametric distribution of the radius
-length_distribution_mean: the mean or expectation of the normal (Gaussian) parametric distribution of the cylinder length; when length_distribution_mean and length_distribution_standard deviations are both set to 0, the generated objects become spheres
-length_distribution_standard_deviation: the standard deviation of the normal (Gaussian) parametric distribution of the cylinder length
-        - - intensity_profile_radius: the relative distance of pixels from the axis of each generated cylinder, where 1 denotes the radius of the cylinder, values <1 denote pixels inside the cylinder, and values >1 are pixels outside the cylinder
-        - - intensity_profile_intensity: the intensity of pixels at the given intensity_profile_radius
-orientation_anisotropic: if checked, cylinders will be generated not randomly and isotropically, but with preferential orientations 
-        - - orientation_main: Cartesian coordinates of the terminal point of an Euclidean vector showing the preferred orientation (active only when orientation_anisotropic is checked)
-orientation_variance_rad: statistical variance of the angle between the orientation_main and the vectors representing the axes of generated cylinders (active only when orientation_anisotropic is checked)
-volume_fraction: the expected volume fraction of the cylinders within the region of interest (ROI); 
-maximum_1000_iteration_number: the maximum number of thousands of iterations at which the generator stops
-random_generator_seed: a number used to initialize the pseudorandom number generator. Can be any integer between 0 and (2^32 - 1) inclusive, an array (or other sequence) of such integers, or None (the default). If seed is None (the value is set to -1), then RandomState will try to read data from /dev/urandom (or the Windows analogue) if available or seed from the clock otherwise must be convertible to 32bit unsigned integers
-last_element_can_be_smaller: 
-
-Postprocessing:
-gaussian_blur: Gaussian blur is used if this parameter is set True. 
-gaussian_filter_sigma_mm: Standard deviation for Gaussian kernel. The standard deviations of the Gaussian filter 
-add_noise 
-noise_preview
-limit_negative_intensities
-noise_random_generator_seed
-exponent
-lambda_start
-lambda_stop
-noise_amplitude
-noise_mean
-surface_measurement
-measurement_multiplier
-measurement_resolution
-output_dtype
-negative
-required_teigen_version: 0.2.14
 
+* Appearance
+     * `skip_volume_generation` - can be used if slice images are not required
+     * `surface_3d_preview` - enable 3D preview
+     * `show_aposteriori_surface` - enable 3D aposteriori surface. This surface is based on output voxel-grid information.
+       No apriori information about shape is used.
+     * `noise_preview` - show 1 slice of noise
+     * `force_rewrite` - rewrite output files
+* Output
+     * `note` - can be used for description of experiment
+     * `aposteriori_measurement` - turn on aposteriori measurement. It is based on threshold segmentation
+     of output data and marching cubes algorithm
+     * `one_row_filenam` - CSV file where each measurement is stored as one row. It can be usefull for experiment series
+* Postprocessing
+     * `gaussian_blur` - add gaussian blur to output data. Gaussian blur is used if this parameter is set True. 
+     * `gaussian_filter_sigma_mm` - control gaussian blur standard deviation parameter
+     * `add_noise` - add noise to output data
+     * `limit_negative_intensities` - output intensities under the zero value are set to zero
+     * `noise_rng_seed` - random generator seed
+     * `noise_exponent` - exponent that controls the ratio of the individual components to the wavelength
+     * `noise_lambda0` - Minimum and maximum noise wavelengths in millimeters
+     * `noise_lambda1` - Minimum and maximum noise wavelengths in millimeters
+     * `noise_std` - Standard deviation of noise intensity
+     * `noise_mean` - Noise mean intensity
+     * `intensity_profile_radius` the relative distance of pixels from the axis of each generated cylinder, where 1 denotes the radius of the cylinder, values <1 denote pixels inside the cylinder, and values >1 are pixels outside the cylinder
+     * `intensity_profile_intensity` the intensity of pixels at the given intensity_profile_radius
+     * `background_intensity` - Control the background intensity of output image.
+* Measurement
+     * `tube_shape` - Tube-like objects are generated if set True. Otherwise cylinders are generated.
+     * `polygon_radius_selection_method` - specify compensatin method
+* Batch processing - allow to run multiple configuration 
+* Generators
+     * Unconnected Tubes - generates separated tube objects. This generator produces objects in shape of cylinders of known length and radius. The cylinders end with hemispheres with the same radius attached to each end of the cylinders. If the cylinder length is set to zero, only the two endpoint hemispheres are generated, which results in generating a sphere. The objects do not interfere with each other. 
+     * Connected Tubes - generate objects with overlap
+     * Unconnected Porosity - generates tube shaped cavities with no connection
+     * Connected Porosity - generates tube shaped connected cavities
+* Generator parameters
+     * `element_number`: number of elements at which the generator stops generating further elements, even before reaching the expected volume fraction
+     * `uniform_radius_distribution`: all the values of radius within the given limits appear with the same probability 
+     * `normal_radius_distribution`: the values of radius will be generated using the normal (Gaussian) parametric distribution
+     * `fixed_radius_distribution`: only the mean value of radius will be used
+     * `radius_distribution_minimum`: the lower limit of the radius of cylinders that are to be generated
+     * `radius_distribution_maximum`: the upper limit of the radius of cylinders that are to be generated
+     * `radius_distribution_mean`: the mean or expectation of the normal (Gaussian) parametric distribution of the radius
+     * `radius_distribution_standard_deviation`: the standard deviation of the normal (Gaussian) parametric distribution of the radius
+     * `length_distribution_mean`: the mean or expectation of the normal (Gaussian) parametric distribution of the cylinder length; when length_distribution_mean and length_distribution_standard deviations are both set to 0, the generated objects become spheres
+     * `length_distribution_standard_deviation`: the standard deviation of the normal (Gaussian) parametric distribution of the cylinder length
+     * `orientation_anisotropic`: if checked, cylinders will be generated not randomly and isotropically, but with preferential orientations 
+     * `orientation_main`: Cartesian coordinates of the terminal point of an Euclidean vector showing the preferred orientation (active only when orientation_anisotropic is checked)
+     * `orientation_variance_rad`: statistical variance of the angle between the orientation_main and the vectors representing the axes of generated cylinders (active only when orientation_anisotropic is checked)
+     * `volume_fraction`: the expected volume fraction of the cylinders within the region of interest (ROI); 
+     * `maximum_1000_iteration_number`: the maximum number of thousands of iterations at which the generator stops
+     * `random_generator_seed`: a number used to initialize the pseudorandom number generator. Can be any integer between 0 and (2^32 - 1) inclusive, an array (or other sequence) of such integers, or None (the default). If seed is None (the value is set to -1), then RandomState will try to read data from /dev/urandom (or the Windows analogue) if available or seed from the clock otherwise must be convertible to 32bit unsigned integers
+     * `last_element_can_be_smaller`: 
 
 
 # Outputs: