Classifier Class Reference

ShapeFinder2 classifier object. More...

Inherits object.

Public Member Functions
cvb.Angle	angular_tolerance (self)
	Gets the acceptance parameter for feature gradient angles. More...
int	gradient_type (self)
	Gets the gradient type this classifier uses for feature extraction. More...
None	save (self, str file_name)
	Writes the classifier to a file. More...
List[cvb.shapefinder2.SearchResult]	search_all (self, cvb.ImagePlane plane, cvb.Rect aoi, int precision, float relative_threshold, int minimum_threshold, int coarse_locality, Optional[cvb.AngleRange] rotation_range, Optional[cvb.NumberRange] scale_range)
	Use this classifier to perform a ShapeFinder search on an image plane. More...
None	set_angular_tolerance (self, cvb.Angle angular_tolerance)
	Sets the acceptance parameter for feature gradient angles. More...
bool	use_cuda (self, int status)
	Set the flag to use CUDA if possible or to force not using CUDA. More...
cvb.Image	visualization (self, Optional[int] layer=-1)
	Creates and returns a color coded image representation of this ShapeFinder classifier. More...

Properties
	coarse_features = property
	List[cvb.Point2D]: The features the classifier uses on the coarsely granular level.
	coarse_scale = property
	int: Scale factor between the fine and the coarse feature level.
	comment = property
	str: Gets or sets the comment assigned to the classifier.
	contrast_mode = property
	int: Contrast mode for which this classifier has been created (see cvb.shapefinder2.ContrastMode).
	contrast_threshold = property
	int: Gets or sets the threshold for the gradient slope.
	feature_count = property
	int: Gets the number of features this classifier contains.
	feature_window = property
	cvb.Rect: Gets the feature window of this classifier.
	file_name = property
	str: Gets the name of the file, from which this classifier was loaded.
	fine_features = property
	List[cvb.Point2D]: The features the classifier uses on the finely granular level.
	num_layers = property
	int: Number of layers in the classifier.
	rotation = property
	cvb.AngleRange: Range of rotations for which this classifier has been generated.
	rotation_step = property
	float: Step size at which the rotations in the classifier have been generated.
	scale = property
	cvb.NumberRange: Range of scales for which this classifier has been generated.
	scale_step = property
	float: Step size at which the scales in the classifier have been generated.
	training_window = property
	cvb.Rect: Training window of the classes in the classifier relative to the anchor point.

Detailed Description

ShapeFinder2 classifier object.

Creates a classifier object loading a classifier file.

Parameters

file_name : str The file name.

Member Function Documentation

angular_tolerance()

cvb.Angle angular_tolerance

(

self

)

Gets the acceptance parameter for feature gradient angles.

Returns

cvb.Angle The angular tolerance.

gradient_type()

int gradient_type

(

self

)

Gets the gradient type this classifier uses for feature extraction.

Returns

int The gradient type (see cvb.shapefinder2.GradientType).

save()

None save	(		self,
		str	file_name
	)

Writes the classifier to a file.

Parameters

file_name : str The file name to save the classifier.

search_all()

List[cvb.shapefinder2.SearchResult] search_all	(		self,
		cvb.ImagePlane	plane,
		cvb.Rect	aoi,
		int	precision,
		float	relative_threshold,
		int	minimum_threshold,
		int	coarse_locality,
		Optional[cvb.AngleRange]	rotation_range,
		Optional[cvb.NumberRange]	scale_range
	)

Use this classifier to perform a ShapeFinder search on an image plane.

Parameters

plane : cvb.ImagePlane The image plane in which to use the classifier.

aoi : cvb.Rect The area of interest in the plane to search in.

precision : int The search mode to use (see cvb.shapefinder2.PrecisionMode).

relative_threshold : float The threshold used for eliminating less promising candidates in the edge model search step.

minimum_threshold : int The minimum gray value different that must be visible in the image before an edge location may be considered a feature.

coarse_locality : int The minimum distance between two results in the coarse layer during the edge model search.

rotation_range : Optional[cvb.AngleRange] The range of rotations to be reported.

scale_range : Optional[cvb.NumberRange] The range of scales to be reported.

Returns

List[cvb.shapefinder2.SearchResult] The list of all results that were found.

set_angular_tolerance()

None set_angular_tolerance	(		self,
		cvb.Angle	angular_tolerance
	)

Sets the acceptance parameter for feature gradient angles.

Parameters

angular_tolerance : cvb.Angle The angular tolerance.

use_cuda()

bool use_cuda	(		self,
		int	status
	)

Set the flag to use CUDA if possible or to force not using CUDA.

Parameters

status : int The CUDA status. If CudaStatus.ForceDisable CUDA won't be used in SF2Search. (see cvb.shapefinder2.CudaStatus).

Returns

bool True if status == CudaStatus.Default and a CUDA capable device is present, false otherwise.

visualization()

cvb.Image visualization	(		self,
		Optional[int]	layer = -1
	)

Creates and returns a color coded image representation of this ShapeFinder classifier.

Parameters

layer : Optional[int] The layer of the classifier to be visualized. A value of -1 superimposes all layers (default -1).

Returns

cvb.Image The image representation.