PartSegCore.segmentation¶

Here we describe base classes for segmentation

Base classes¶

This module contains roi_extraction algorithms

class PartSegCore.segmentation.BaseThreshold[source]¶

Bases: PartSegCore.algorithm_describe_base.AlgorithmDescribeBase, abc.ABC

classmethod get_fields()¶

This function return list of parameters needed by algorithm. It is used for generate form in User Interface

Return type: List[Union[AlgorithmProperty, str]]
Returns: list of algorithm parameters and comments

abstract classmethod get_name()¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Return type: str
Returns: name of algorithm

class PartSegCore.segmentation.BaseWatershed[source]¶

Bases: PartSegCore.algorithm_describe_base.AlgorithmDescribeBase, abc.ABC

base class for all sprawl interface

classmethod get_fields()¶

This function return list of parameters needed by algorithm. It is used for generate form in User Interface

Return type: List[Union[AlgorithmProperty, str]]
Returns: list of algorithm parameters and comments

abstract classmethod get_name()¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Return type: str
Returns: name of algorithm

classmethod sprawl(sprawl_area, core_objects, data, components_num, spacing, side_connection, operator, arguments, lower_bound, upper_bound)[source]¶

This method calculate sprawl

Parameters

sprawl_area (ndarray) – Mask area to which sprawl should be limited
core_objects (ndarray) – Starting objects for sprawl
data (ndarray) – density information
components_num (int) – numer of components in core_objects
spacing – Image spacing. Needed for sprawls which use metrics
side_connection (bool) –
operator (Callable[[Any, Any], bool]) –
arguments (dict) – dict with parameters reported by function get_fields()
lower_bound – data value lower bound
upper_bound – data value upper bound

Returns

class PartSegCore.segmentation.NoiseFilteringBase[source]¶

Bases: PartSegCore.algorithm_describe_base.AlgorithmDescribeBase, abc.ABC

Base class for noise filtering operations

classmethod get_fields()¶

This function return list of parameters needed by algorithm. It is used for generate form in User Interface

Return type: List[Union[AlgorithmProperty, str]]
Returns: list of algorithm parameters and comments

abstract classmethod get_name()¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Return type: str
Returns: name of algorithm

classmethod noise_filter(channel, spacing, arguments)[source]¶

This function need be overloaded in implementation

Parameters

channel (ndarray) – single channel ad 2d or 3d array
spacing (Iterable[float]) – image spacing
arguments (dict) – additional arguments defined by get_fields()

Return type

ndarray

Returns

channel array with removed noise

class PartSegCore.segmentation.ROIExtractionAlgorithm[source]¶

Bases: PartSegCore.algorithm_describe_base.AlgorithmDescribeBase, abc.ABC

Base class for all segmentation algorithm.

Variables

~.image (Image) – Image to process
~.channel (numpy.ndarray) – selected channel
~.segmentation (numpy.ndarray) – final segmentation
~.mask (numpy.ndarray) – mask limiting segmentation area

classmethod get_fields()¶

This function return list of parameters needed by algorithm. It is used for generate form in User Interface

Return type: List[Union[AlgorithmProperty, str]]
Returns: list of algorithm parameters and comments

abstract classmethod get_name()¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Return type: str
Returns: name of algorithm

static get_steps_num()[source]¶: Return number of algorithm steps if your algorithm report progress, else should return 0

set_mask(mask)[source]¶: Set mask which will limit segmentation area

class PartSegCore.segmentation.ROIExtractionResult(roi, parameters, additional_layers=<factory>, info_text='', roi_annotation=<factory>, alternative_representation=<factory>, file_path=None, roi_info=None, points=None)[source]¶

Bases: object

Class to store results of roi extraction process.

Variables

roi (np.ndarray) – Region of Interest represented as numpy array.
parameters (ROIExtractionProfile) – parameters of roi extraction process.
~.additional_layers (Dict[str,AdditionalLayerDescription]) – additional layers returned by algorithm. Could be previewer using napari viewer.
roi_annotation (dict) – Annotation for ROI. Currently displayed as tooltip
alternative_representation (Dict[str,np.ndarray]) – Arrays with alternative representations of ROI.
~.file_path (Optional[str]) – information on which file roi extraction was performed.
~.roi_info (ROIInfo) – ROIInfo for current roi.
~.points (Optional[np.ndarray]) – array of points.

class PartSegCore.segmentation.RestartableAlgorithm(**kwargs)[source]¶

Bases: PartSegCore.segmentation.algorithm_base.ROIExtractionAlgorithm, abc.ABC

Base class for restartable segmentation algorithm. The idea is to store two copies of algorithm parameters and base on difference check from which point restart the calculation.

Variables

~.parameters (dict) – variable for store last run parameters
~.new_parameters (dict) – variable for store parameters for next run

abstract calculation_run(report_fun)[source]¶

Restartable calculation may return None if there is no need to recalculate

Return type: Optional[ROIExtractionResult]

classmethod get_fields()¶

This function return list of parameters needed by algorithm. It is used for generate form in User Interface

Return type: List[Union[AlgorithmProperty, str]]
Returns: list of algorithm parameters and comments

abstract classmethod get_name()¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Return type: str
Returns: name of algorithm

static get_steps_num()¶: Return number of algorithm steps if your algorithm report progress, else should return 0

set_mask(mask)[source]¶: Set mask which will limit segmentation area

class PartSegCore.segmentation.StackAlgorithm[source]¶

Bases: PartSegCore.segmentation.algorithm_base.ROIExtractionAlgorithm, abc.ABC

classmethod get_fields()¶

This function return list of parameters needed by algorithm. It is used for generate form in User Interface

Return type: List[Union[AlgorithmProperty, str]]
Returns: list of algorithm parameters and comments

abstract classmethod get_name()¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Return type: str
Returns: name of algorithm

static get_steps_num()¶: Return number of algorithm steps if your algorithm report progress, else should return 0

set_mask(mask)¶: Set mask which will limit segmentation area

.watershed¶

This module contains PartSeg wrappers for function for sprawl_utils.find_split.

class PartSegCore.segmentation.watershed.BaseWatershed[source]¶

Bases: PartSegCore.algorithm_describe_base.AlgorithmDescribeBase, abc.ABC

base class for all sprawl interface

classmethod get_fields()¶

This function return list of parameters needed by algorithm. It is used for generate form in User Interface

Return type: List[Union[AlgorithmProperty, str]]
Returns: list of algorithm parameters and comments

abstract classmethod get_name()¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Return type: str
Returns: name of algorithm

classmethod sprawl(sprawl_area, core_objects, data, components_num, spacing, side_connection, operator, arguments, lower_bound, upper_bound)[source]¶

This method calculate sprawl

Parameters

sprawl_area (ndarray) – Mask area to which sprawl should be limited
core_objects (ndarray) – Starting objects for sprawl
data (ndarray) – density information
components_num (int) – numer of components in core_objects
spacing – Image spacing. Needed for sprawls which use metrics
side_connection (bool) –
operator (Callable[[Any, Any], bool]) –
arguments (dict) – dict with parameters reported by function get_fields()
lower_bound – data value lower bound
upper_bound – data value upper bound

Returns

class PartSegCore.segmentation.watershed.DistanceWatershed[source]¶

Bases: PartSegCore.segmentation.watershed.BaseWatershed

Calculate Euclidean sprawl (watershed) with respect to image spacing

classmethod get_fields()¶

This function return list of parameters needed by algorithm. It is used for generate form in User Interface

Return type: List[Union[AlgorithmProperty, str]]
Returns: list of algorithm parameters and comments

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

classmethod sprawl(sprawl_area, core_objects, data, components_num, spacing, side_connection, operator, arguments, lower_bound, upper_bound)[source]¶

This method calculate sprawl

Parameters

sprawl_area (ndarray) – Mask area to which sprawl should be limited
core_objects (ndarray) – Starting objects for sprawl
data (ndarray) – density information
components_num (int) – numer of components in core_objects
spacing – Image spacing. Needed for sprawls which use metrics
side_connection (bool) –
operator (Callable[[Any, Any], bool]) –
arguments (dict) – dict with parameters reported by function get_fields()
lower_bound – data value lower bound
upper_bound – data value upper bound

Returns

class PartSegCore.segmentation.watershed.FDTWatershed[source]¶

Bases: PartSegCore.segmentation.watershed.BaseWatershed

classmethod get_fields()¶

This function return list of parameters needed by algorithm. It is used for generate form in User Interface

Return type: List[Union[AlgorithmProperty, str]]
Returns: list of algorithm parameters and comments

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

classmethod sprawl(sprawl_area, core_objects, data, components_num, spacing, side_connection, operator, arguments, lower_bound, upper_bound)[source]¶

This method calculate sprawl

Parameters

sprawl_area (ndarray) – Mask area to which sprawl should be limited
core_objects (ndarray) – Starting objects for sprawl
data (ndarray) – density information
components_num (int) – numer of components in core_objects
spacing – Image spacing. Needed for sprawls which use metrics
side_connection (bool) –
operator (Callable[[Any, Any], bool]) –
arguments (dict) – dict with parameters reported by function get_fields()
lower_bound – data value lower bound
upper_bound – data value upper bound

Returns

PartSegCore.segmentation.watershed.FlowMethodSelection¶: alias of PartSegCore.segmentation.watershed.WatershedSelection

class PartSegCore.segmentation.watershed.MSOWatershed[source]¶

Bases: PartSegCore.segmentation.watershed.BaseWatershed

This algorithm has following parameters:

step_limits (ConstrainedIntValue)- Steep limits, Limits of Steps
reflective (bool)- Reflective

classmethod get_fields()¶

This function return list of parameters needed by algorithm. It is used for generate form in User Interface

Return type: List[Union[AlgorithmProperty, str]]
Returns: list of algorithm parameters and comments

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

classmethod sprawl(cls, sprawl_area, core_objects, data, components_num, spacing, side_connection, operator, arguments, lower_bound, upper_bound)[source]¶

This method calculate sprawl

Parameters

sprawl_area (ndarray) – Mask area to which sprawl should be limited
core_objects (ndarray) – Starting objects for sprawl
data (ndarray) – density information
components_num (int) – numer of components in core_objects
spacing – Image spacing. Needed for sprawls which use metrics
side_connection (bool) –
operator (Callable[[Any, Any], bool]) –
arguments (MSOWatershedParams) – dict with parameters reported by function get_fields()
lower_bound – data value lower bound
upper_bound – data value upper bound

Returns

class PartSegCore.segmentation.watershed.NeighType(value)[source]¶

Bases: enum.Enum

An enumeration.

class PartSegCore.segmentation.watershed.PathDistanceWatershed[source]¶

Bases: PartSegCore.segmentation.watershed.BaseWatershed

classmethod get_fields()¶

This function return list of parameters needed by algorithm. It is used for generate form in User Interface

Return type: List[Union[AlgorithmProperty, str]]
Returns: list of algorithm parameters and comments

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

classmethod sprawl(sprawl_area, core_objects, data, components_num, spacing, side_connection, operator, arguments, lower_bound, upper_bound)[source]¶

This method calculate sprawl

Parameters

sprawl_area (ndarray) – Mask area to which sprawl should be limited
core_objects (ndarray) – Starting objects for sprawl
data (ndarray) – density information
components_num (int) – numer of components in core_objects
spacing – Image spacing. Needed for sprawls which use metrics
side_connection (bool) –
operator (Callable[[Any, Any], bool]) –
arguments (dict) – dict with parameters reported by function get_fields()
lower_bound – data value lower bound
upper_bound – data value upper bound

Returns

class PartSegCore.segmentation.watershed.PathWatershed[source]¶

Bases: PartSegCore.segmentation.watershed.BaseWatershed

classmethod get_fields()¶

This function return list of parameters needed by algorithm. It is used for generate form in User Interface

Return type: List[Union[AlgorithmProperty, str]]
Returns: list of algorithm parameters and comments

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

classmethod sprawl(sprawl_area, core_objects, data, components_num, spacing, side_connection, operator, arguments, lower_bound, upper_bound)[source]¶

This method calculate sprawl

Parameters

sprawl_area (ndarray) – Mask area to which sprawl should be limited
core_objects (ndarray) – Starting objects for sprawl
data (ndarray) – density information
components_num (int) – numer of components in core_objects
spacing – Image spacing. Needed for sprawls which use metrics
side_connection (bool) –
operator (Callable[[Any, Any], bool]) –
arguments (dict) – dict with parameters reported by function get_fields()
lower_bound – data value lower bound
upper_bound – data value upper bound

Returns

pydantic model PartSegCore.segmentation.watershed.WatershedSelection[source]¶

Bases: PartSegCore.algorithm_describe_base.AlgorithmSelection

This register contains algorithms for sprawl area from core object.

Show JSON schema

{
   "title": "WatershedSelection",
   "description": "This register contains algorithms for sprawl area from core object.",
   "type": "object",
   "properties": {
      "name": {
         "title": "Name",
         "type": "string"
      },
      "values": {
         "title": "Values",
         "anyOf": [
            {
               "$ref": "#/definitions/BaseModel"
            },
            {
               "type": "object"
            }
         ]
      },
      "class_path": {
         "title": "Class Path",
         "default": "",
         "type": "string"
      }
   },
   "required": [
      "name",
      "values"
   ],
   "additionalProperties": false,
   "definitions": {
      "BaseModel": {
         "title": "BaseModel",
         "type": "object",
         "properties": {}
      }
   }
}

Config

extra: str = forbid

Fields

class_path (str)
name (str)
values (Union[pydantic.main.BaseModel, Dict[str, Any]])

Validators

check_name » name
update_class_path » class_path
update_values » values

field class_path: str = ''¶

Validated by

update_class_path

field name: str [Required]¶

Validated by

check_name

field values: Union[pydantic.main.BaseModel, Dict[str, Any]] [Required]¶

Validated by

update_values

algorithm()¶

validator check_name » name¶

classmethod construct(_fields_set=None, **values)¶

Creates a new model setting __dict__ and __fields_set__ from trusted or pre-validated data. Default values are respected, but no other validation is performed. Behaves as if Config.extra = ‘allow’ was set since it adds all passed values

Return type: Model

copy(*, validate=True, **kwargs)¶

Duplicate a model, optionally choose which fields to include, exclude and change.

Parameters

include – fields to include in new model
exclude – fields to exclude from new model, as with values this takes precedence over include
update – values to change/add in the new model. Note: the data is not validated before creating the new model: you should trust this data
deep – set to True to make a deep copy of the model

Return type

BaseModel

Returns

new model instance

dict(*, include=None, exclude=None, by_alias=False, skip_defaults=None, exclude_unset=False, exclude_defaults=False, exclude_none=False)¶

Generate a dictionary representation of the model, optionally specifying which fields to include or exclude.

Return type: DictStrAny

classmethod from_orm(obj)¶

Return type: Model

classmethod get_default()¶

json(*, include=None, exclude=None, by_alias=False, skip_defaults=None, exclude_unset=False, exclude_defaults=False, exclude_none=False, encoder=None, models_as_dict=True, **dumps_kwargs)¶

Generate a JSON representation of the model, include and exclude arguments as per dict().

encoder is an optional function to supply as default to json.dumps(), other arguments as per json.dumps().

Return type: unicode

classmethod parse_file(path, *, content_type=None, encoding='utf8', proto=None, allow_pickle=False)¶

Return type: Model

classmethod parse_obj(obj)¶

Return type: Model

classmethod parse_raw(b, *, content_type=None, encoding='utf8', proto=None, allow_pickle=False)¶

Return type: Model

classmethod register(value, replace=False, old_names=None)¶

Function for registering AlgorithmDescribeBase based algorithms :type value: TypeVar(AlgorithmType, bound= Type[AlgorithmDescribeBase]) :param value: algorithm to register :param replace: replace existing algorithm, be patient with :type old_names: Optional[List[str]] :param old_names: list of old names for registered class

Return type: TypeVar(AlgorithmType, bound= Type[AlgorithmDescribeBase])

classmethod schema(by_alias=True, ref_template='#/definitions/{model}')¶

Return type: DictStrAny

classmethod schema_json(*, by_alias=True, ref_template='#/definitions/{model}', **dumps_kwargs)¶

Return type: unicode

validator update_class_path » class_path¶

classmethod update_forward_refs(**localns)¶

Try to update ForwardRefs on fields based on this Model, globalns and localns.

Return type: None

validator update_values » values¶

classmethod validate(value)¶

Return type: Model

PartSegCore.segmentation.watershed.calculate_distances_array(spacing, neigh_type)[source]¶

Parameters

spacing – image spacing
neigh_type (NeighType) – neighbourhood type

Returns

neighbourhood array, distance array

.restartable_segmentation_algorithms¶

class PartSegCore.segmentation.restartable_segmentation_algorithms.BaseMultiScaleOpening[source]¶

Bases: PartSegCore.segmentation.restartable_segmentation_algorithms.TwoLevelThresholdBaseAlgorithm, abc.ABC

calculation_run(report_fun)[source]¶

main calculation function

Parameters: report_fun – function used to trace progress
Return type: Optional[ROIExtractionResult]

class PartSegCore.segmentation.restartable_segmentation_algorithms.BaseThresholdFlowAlgorithm[source]¶

Bases: PartSegCore.segmentation.restartable_segmentation_algorithms.TwoLevelThresholdBaseAlgorithm, abc.ABC

calculation_run(report_fun)[source]¶

main calculation function

Parameters: report_fun – function used to trace progress
Return type: Optional[ROIExtractionResult]

class PartSegCore.segmentation.restartable_segmentation_algorithms.BorderRim[source]¶

Bases: PartSegCore.segmentation.restartable_segmentation_algorithms.RestartableAlgorithm

This class wrap the PartSegCore.mask_partition_utils.BorderRim` class in segmentation algorithm interface. It allow user to check how rim look with given set of parameters

This algorithm has following parameters:

distance (ConstrainedFloatValue)- Distance
units (Units)- Units

calculation_run(_report_fun)[source]¶

Restartable calculation may return None if there is no need to recalculate

Return type: ROIExtractionResult

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.restartable_segmentation_algorithms.LowerThresholdAlgorithm(**kwargs)[source]¶

Bases: PartSegCore.segmentation.restartable_segmentation_algorithms.OneThresholdAlgorithm

Implementation of lower threshold algorithm. It has same flow like ThresholdBaseAlgorithm. The area of interest are voxels from filtered channel with value above the given threshold

This algorithm has following parameters:

channel (Channel)- Channel
noise_filtering (AlgorithmDescribeBase)- Filter
threshold (AlgorithmDescribeBase)- Threshold
minimum_size (ConstrainedIntValue)- Minimum size (px)
side_connection (bool)- Connect only sides, During calculation of connected components includes only side by side connected pixels

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

static threshold_operator(a, b, /)¶: Same as a > b.

class PartSegCore.segmentation.restartable_segmentation_algorithms.LowerThresholdFlowAlgorithm[source]¶

Bases: PartSegCore.segmentation.restartable_segmentation_algorithms.BaseThresholdFlowAlgorithm

This algorithm has following parameters:

channel (Channel)- Channel
noise_filtering (AlgorithmDescribeBase)- Filter
threshold (AlgorithmDescribeBase)- Threshold
flow_type (AlgorithmDescribeBase)- Flow type
minimum_size (ConstrainedIntValue)- Minimum core size (px)
side_connection (bool)- Connect only sides, During calculation of connected components includes only side by side connected pixels
remove_object_touching_border (bool)- Remove objects touching border, Remove objects touching border

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

static threshold_operator(a, b, /)¶: Same as a > b.

class PartSegCore.segmentation.restartable_segmentation_algorithms.LowerThresholdMultiScaleOpening[source]¶

Bases: PartSegCore.segmentation.restartable_segmentation_algorithms.BaseMultiScaleOpening

This algorithm has following parameters:

channel (Channel)- Channel
noise_filtering (AlgorithmDescribeBase)- Filter
minimum_size (ConstrainedIntValue)- Minimum size (px)
side_connection (bool)- Connect only sides, During calculation of connected components includes only side by side connected pixels
threshold (AlgorithmDescribeBase)- Threshold
mu_mid (AlgorithmDescribeBase)- Mu mid value
step_limits (ConstrainedIntValue)- Limits of Steps

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

static threshold_operator(a, b, /)¶: Same as a > b.

class PartSegCore.segmentation.restartable_segmentation_algorithms.MaskDistanceSplit(**kwargs)[source]¶

Bases: PartSegCore.segmentation.restartable_segmentation_algorithms.RestartableAlgorithm

This class wrap the PartSegCore.mask_partition_utils.SplitMaskOnPart class in segmentation algorithm interface. It allow user to check how split look with given set of parameters

This algorithm has following parameters:

num_of_parts (ConstrainedIntValue)- Number of Parts
equal_volume (bool)- Equal Volume, If split should be done in respect of parts volume of parts thickness.

calculation_run(report_fun)[source]¶

Restartable calculation may return None if there is no need to recalculate

Return type: ROIExtractionResult

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Return type: str
Returns: name of algorithm

class PartSegCore.segmentation.restartable_segmentation_algorithms.OneThresholdAlgorithm(**kwargs)[source]¶

Bases: PartSegCore.segmentation.restartable_segmentation_algorithms.ThresholdBaseAlgorithm, abc.ABC

Base class for PartSeg analysis algorithm which apply one threshold. Created for reduce code repetition.

class PartSegCore.segmentation.restartable_segmentation_algorithms.OtsuSegment[source]¶

Bases: PartSegCore.segmentation.restartable_segmentation_algorithms.RestartableAlgorithm

This algorithm has following parameters:

channel (Channel)- Channel
noise_filtering (AlgorithmDescribeBase)- Noise Removal
components (ConstrainedIntValue)- Number of Components
valley (bool)- Valley emphasis
hist_num (ConstrainedIntValue)- Number of histogram bins

calculation_run(report_fun)[source]¶: Restartable calculation may return None if there is no need to recalculate

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.restartable_segmentation_algorithms.RangeThresholdAlgorithm(**kwargs)[source]¶

Bases: PartSegCore.segmentation.restartable_segmentation_algorithms.ThresholdBaseAlgorithm

Implementation of upper threshold algorithm. It has same flow like ThresholdBaseAlgorithm. The area of interest are voxels from filtered channel with value between the lower and upper threshold

This algorithm has following parameters:

channel (Channel)- Channel
noise_filtering (AlgorithmDescribeBase)- Filter
threshold (AlgorithmDescribeBase)- Threshold
minimum_size (ConstrainedIntValue)- Minimum size (px)
side_connection (bool)- Connect only sides, During calculation of connected components includes only side by side connected pixels

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.restartable_segmentation_algorithms.RestartableAlgorithm(**kwargs)[source]¶

Bases: PartSegCore.segmentation.algorithm_base.ROIExtractionAlgorithm, abc.ABC

Base class for restartable segmentation algorithm. The idea is to store two copies of algorithm parameters and base on difference check from which point restart the calculation.

Variables

~.parameters (dict) – variable for store last run parameters
~.new_parameters (dict) – variable for store parameters for next run

abstract calculation_run(report_fun)[source]¶

Restartable calculation may return None if there is no need to recalculate

Return type: Optional[ROIExtractionResult]

set_mask(mask)[source]¶: Set mask which will limit segmentation area

class PartSegCore.segmentation.restartable_segmentation_algorithms.ThresholdBaseAlgorithm(**kwargs)[source]¶

Bases: PartSegCore.segmentation.restartable_segmentation_algorithms.RestartableAlgorithm, abc.ABC

Base class for most threshold Algorithm implemented in PartSeg analysis. Created for reduce code repetition.

calculation_run(report_fun)[source]¶

main calculation function

Parameters: report_fun (Callable[[str, int], Any]) – function used to trace progress
Return type: Optional[ROIExtractionResult]

get_additional_layers(full_segmentation=None)[source]¶

Create dict with standard additional layers.

Parameters: full_segmentation (Optional[ndarray]) – no size filtering if not self.segmentation
Return type: Dict[str, AdditionalLayerDescription]
Returns

prepare_result(roi)[source]¶

Collect data for result.

Parameters: roi (ndarray) – array with segmentation
Return type: ROIExtractionResult
Returns: algorithm result description

class PartSegCore.segmentation.restartable_segmentation_algorithms.TwoLevelThresholdBaseAlgorithm[source]¶: Bases: PartSegCore.segmentation.restartable_segmentation_algorithms.ThresholdBaseAlgorithm, abc.ABC

class PartSegCore.segmentation.restartable_segmentation_algorithms.UpperThresholdAlgorithm(**kwargs)[source]¶

Bases: PartSegCore.segmentation.restartable_segmentation_algorithms.OneThresholdAlgorithm

Implementation of upper threshold algorithm. It has same flow like ThresholdBaseAlgorithm. The area of interest are voxels from filtered channel with value below the given threshold

This algorithm has following parameters:

channel (Channel)- Channel
noise_filtering (AlgorithmDescribeBase)- Filter
threshold (AlgorithmDescribeBase)- Threshold
minimum_size (ConstrainedIntValue)- Minimum size (px)
side_connection (bool)- Connect only sides, During calculation of connected components includes only side by side connected pixels

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

static threshold_operator(a, b, /)¶: Same as a < b.

class PartSegCore.segmentation.restartable_segmentation_algorithms.UpperThresholdFlowAlgorithm[source]¶

Bases: PartSegCore.segmentation.restartable_segmentation_algorithms.BaseThresholdFlowAlgorithm

This algorithm has following parameters:

channel (Channel)- Channel
noise_filtering (AlgorithmDescribeBase)- Filter
threshold (AlgorithmDescribeBase)- Threshold
flow_type (AlgorithmDescribeBase)- Flow type
minimum_size (ConstrainedIntValue)- Minimum core size (px)
side_connection (bool)- Connect only sides, During calculation of connected components includes only side by side connected pixels
remove_object_touching_border (bool)- Remove objects touching border, Remove objects touching border

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

static threshold_operator(a, b, /)¶: Same as a < b.

class PartSegCore.segmentation.restartable_segmentation_algorithms.UpperThresholdMultiScaleOpening[source]¶

Bases: PartSegCore.segmentation.restartable_segmentation_algorithms.BaseMultiScaleOpening

This algorithm has following parameters:

channel (Channel)- Channel
noise_filtering (AlgorithmDescribeBase)- Filter
minimum_size (ConstrainedIntValue)- Minimum size (px)
side_connection (bool)- Connect only sides, During calculation of connected components includes only side by side connected pixels
threshold (AlgorithmDescribeBase)- Threshold
mu_mid (AlgorithmDescribeBase)- Mu mid value
step_limits (ConstrainedIntValue)- Limits of Steps

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

static threshold_operator(a, b, /)¶: Same as a < b.

.threshold¶

class PartSegCore.segmentation.threshold.BaseThreshold[source]¶: Bases: PartSegCore.algorithm_describe_base.AlgorithmDescribeBase, abc.ABC

class PartSegCore.segmentation.threshold.DoubleOtsu[source]¶

Bases: PartSegCore.segmentation.threshold.BaseThreshold

This algorithm has following parameters:

valley (bool)- Valley emphasis
bins (ConstrainedIntValue)- Histogram bins

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.threshold.DoubleThreshold[source]¶

Bases: PartSegCore.segmentation.threshold.BaseThreshold

This algorithm has following parameters:

core_threshold (AlgorithmDescribeBase)- Core threshold
base_threshold (AlgorithmDescribeBase)- Base threshold

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

pydantic model PartSegCore.segmentation.threshold.DoubleThresholdSelection[source]¶

Bases: PartSegCore.algorithm_describe_base.AlgorithmSelection

Show JSON schema

{
   "title": "DoubleThresholdSelection",
   "description": "Base class for algorithm selection.\nFor given algorithm there should be Register instance set __register__ class variable.",
   "type": "object",
   "properties": {
      "name": {
         "title": "Name",
         "type": "string"
      },
      "values": {
         "title": "Values",
         "anyOf": [
            {
               "$ref": "#/definitions/BaseModel"
            },
            {
               "type": "object"
            }
         ]
      },
      "class_path": {
         "title": "Class Path",
         "default": "",
         "type": "string"
      }
   },
   "required": [
      "name",
      "values"
   ],
   "additionalProperties": false,
   "definitions": {
      "BaseModel": {
         "title": "BaseModel",
         "type": "object",
         "properties": {}
      }
   }
}

Config

extra: str = forbid

Fields

class_path (str)
name (str)
values (Union[pydantic.main.BaseModel, Dict[str, Any]])

Validators

field class_path: str = ''¶

Validated by

update_class_path

field name: str [Required]¶

Validated by

check_name

field values: Union[pydantic.main.BaseModel, Dict[str, Any]] [Required]¶

Validated by

update_values

class PartSegCore.segmentation.threshold.HuangThreshold[source]¶

Bases: PartSegCore.segmentation.threshold.SitkThreshold

This algorithm has following parameters:

apply_mask (bool)- Apply mask, If apply mask before calculate threshold
bins (ConstrainedIntValue)- Histogram bins

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.threshold.IntermodesThreshold[source]¶

Bases: PartSegCore.segmentation.threshold.SitkThreshold

This algorithm has following parameters:

apply_mask (bool)- Apply mask, If apply mask before calculate threshold
bins (ConstrainedIntValue)- Histogram bins

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.threshold.IsoDataThreshold[source]¶

Bases: PartSegCore.segmentation.threshold.SitkThreshold

This algorithm has following parameters:

apply_mask (bool)- Apply mask, If apply mask before calculate threshold
bins (ConstrainedIntValue)- Histogram bins

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.threshold.KittlerIllingworthThreshold[source]¶

Bases: PartSegCore.segmentation.threshold.SitkThreshold

This algorithm has following parameters:

apply_mask (bool)- Apply mask, If apply mask before calculate threshold
bins (ConstrainedIntValue)- Histogram bins

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.threshold.LiThreshold[source]¶

Bases: PartSegCore.segmentation.threshold.SitkThreshold

This algorithm has following parameters:

apply_mask (bool)- Apply mask, If apply mask before calculate threshold
bins (ConstrainedIntValue)- Histogram bins

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.threshold.ManualThreshold[source]¶

Bases: PartSegCore.segmentation.threshold.BaseThreshold

This algorithm has following parameters:

threshold (ConstrainedFloatValue)- Threshold, Threshold values

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.threshold.MaximumDistanceCore[source]¶

Bases: PartSegCore.segmentation.threshold.BaseThreshold

This Is algorithm intended to bue used in “* threshold with watershed” algorithms. It generates array with three values:

0 - background,

1 - area to watershed,

2 - core objects to start watershed from

This algorithm is developed to make possible split of almost convex objects that are touching each other. Core objects are identified as local maxima of distance from the border.

To perform this task the following steps are performed:

Thresholding - to detect whole area of objects. This is controlled by threshold parameter.
Remove small objects - to remove small objects. This is controlled by ``minimum_size` parameter.
Small objects close - to merge small objects. As distance transform is used, it is required small holes. This steep closes holes smaller tan 10px.
Distance transform - to find distance from the border
Identify local maxima - to find core objects
Remove local maxima that are too close to the border - to avoid artifacts. This distance is controlled by minimum_border_distance parameter.
Dilate core objects - to make them bigger. For elongated objects it is possible to have multiple local maxima along longest axis of object. This step is to merge them. This distance is controlled by dilate_radius parameter.

This is algorithm that detect core objects

This algorithm has following parameters:

threshold (AlgorithmDescribeBase)- Threshold
dilate_radius (ConstrainedIntValue)- Dilate Radius, To merge small objects
minimum_size (ConstrainedIntValue)- Minimum Size, To remove small objects
minimum_border_distance (ConstrainedIntValue)- Border Radius, Minimum distance of local maxima from the border. To avoid artifacts

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.threshold.MaximumEntropyThreshold[source]¶

Bases: PartSegCore.segmentation.threshold.SitkThreshold

This algorithm has following parameters:

apply_mask (bool)- Apply mask, If apply mask before calculate threshold
bins (ConstrainedIntValue)- Histogram bins

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.threshold.MomentsThreshold[source]¶

Bases: PartSegCore.segmentation.threshold.SitkThreshold

This algorithm has following parameters:

apply_mask (bool)- Apply mask, If apply mask before calculate threshold
bins (ConstrainedIntValue)- Histogram bins

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.threshold.MultipleOtsu[source]¶

Bases: PartSegCore.segmentation.threshold.BaseThreshold

This algorithm has following parameters:

components (ConstrainedIntValue)- Number of Components
lower_component (ConstrainedIntValue)- Lower Component
upper_component (ConstrainedIntValue)- Upper Component
valley (bool)- Valley emphasis
bins (ConstrainedIntValue)- Number of histogram bins

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.threshold.MultipleOtsuThreshold[source]¶

Bases: PartSegCore.segmentation.threshold.BaseThreshold

This algorithm has following parameters:

components (ConstrainedIntValue)- Number of Components
border_component (ConstrainedIntValue)- Border Component
valley (bool)- Valley emphasis
bins (ConstrainedIntValue)- Number of histogram bins

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Return type: str
Returns: name of algorithm

class PartSegCore.segmentation.threshold.OtsuThreshold[source]¶

Bases: PartSegCore.segmentation.threshold.SitkThreshold

This algorithm has following parameters:

apply_mask (bool)- Apply mask, If apply mask before calculate threshold
bins (ConstrainedIntValue)- Histogram bins

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.threshold.RangeThreshold[source]¶

Bases: PartSegCore.segmentation.threshold.DoubleThreshold

This algorithm has following parameters:

core_threshold (AlgorithmDescribeBase)- Upper threshold
base_threshold (AlgorithmDescribeBase)- Lower threshold

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

pydantic model PartSegCore.segmentation.threshold.RangeThresholdSelection[source]¶

Bases: PartSegCore.algorithm_describe_base.AlgorithmSelection

Show JSON schema

{
   "title": "RangeThresholdSelection",
   "description": "Base class for algorithm selection.\nFor given algorithm there should be Register instance set __register__ class variable.",
   "type": "object",
   "properties": {
      "name": {
         "title": "Name",
         "type": "string"
      },
      "values": {
         "title": "Values",
         "anyOf": [
            {
               "$ref": "#/definitions/BaseModel"
            },
            {
               "type": "object"
            }
         ]
      },
      "class_path": {
         "title": "Class Path",
         "default": "",
         "type": "string"
      }
   },
   "required": [
      "name",
      "values"
   ],
   "additionalProperties": false,
   "definitions": {
      "BaseModel": {
         "title": "BaseModel",
         "type": "object",
         "properties": {}
      }
   }
}

Config

extra: str = forbid

Fields

class_path (str)
name (str)
values (Union[pydantic.main.BaseModel, Dict[str, Any]])

Validators

field class_path: str = ''¶

Validated by

update_class_path

field name: str [Required]¶

Validated by

check_name

field values: Union[pydantic.main.BaseModel, Dict[str, Any]] [Required]¶

Validated by

update_values

class PartSegCore.segmentation.threshold.RenyiEntropyThreshold[source]¶

Bases: PartSegCore.segmentation.threshold.SitkThreshold

This algorithm has following parameters:

apply_mask (bool)- Apply mask, If apply mask before calculate threshold
bins (ConstrainedIntValue)- Histogram bins

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.threshold.ShanbhagThreshold[source]¶

Bases: PartSegCore.segmentation.threshold.SitkThreshold

This algorithm has following parameters:

apply_mask (bool)- Apply mask, If apply mask before calculate threshold
bins (ConstrainedIntValue)- Histogram bins

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.threshold.SitkThreshold[source]¶: Bases: PartSegCore.segmentation.threshold.BaseThreshold, abc.ABC

pydantic model PartSegCore.segmentation.threshold.ThresholdSelection[source]¶

Bases: PartSegCore.algorithm_describe_base.AlgorithmSelection

Show JSON schema

{
   "title": "ThresholdSelection",
   "description": "Base class for algorithm selection.\nFor given algorithm there should be Register instance set __register__ class variable.",
   "type": "object",
   "properties": {
      "name": {
         "title": "Name",
         "type": "string"
      },
      "values": {
         "title": "Values",
         "anyOf": [
            {
               "$ref": "#/definitions/BaseModel"
            },
            {
               "type": "object"
            }
         ]
      },
      "class_path": {
         "title": "Class Path",
         "default": "",
         "type": "string"
      }
   },
   "required": [
      "name",
      "values"
   ],
   "additionalProperties": false,
   "definitions": {
      "BaseModel": {
         "title": "BaseModel",
         "type": "object",
         "properties": {}
      }
   }
}

Config

extra: str = forbid

Fields

class_path (str)
name (str)
values (Union[pydantic.main.BaseModel, Dict[str, Any]])

Validators

field class_path: str = ''¶

Validated by

update_class_path

field name: str [Required]¶

Validated by

check_name

field values: Union[pydantic.main.BaseModel, Dict[str, Any]] [Required]¶

Validated by

update_values

class PartSegCore.segmentation.threshold.TriangleThreshold[source]¶

Bases: PartSegCore.segmentation.threshold.SitkThreshold

This algorithm has following parameters:

apply_mask (bool)- Apply mask, If apply mask before calculate threshold
bins (ConstrainedIntValue)- Histogram bins

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

class PartSegCore.segmentation.threshold.YenThreshold[source]¶

Bases: PartSegCore.segmentation.threshold.SitkThreshold

This algorithm has following parameters:

apply_mask (bool)- Apply mask, If apply mask before calculate threshold
bins (ConstrainedIntValue)- Histogram bins

classmethod get_name()[source]¶

Algorithm name. It will be used during interface generating and in registering to proper PartSeg.PartSegCore.algorithm_describe_base.Register.

Returns: name of algorithm

PartSegCore.segmentation¶

Base classes¶

.watershed¶

.restartable_segmentation_algorithms¶

.threshold¶

Table of Contents

Previous topic

Next topic

This Page