eVision.EasyDeepLearning.EUnsupervisedSegmenterResult Class Reference

An eVision.EasyDeepLearning.EUnsupervisedSegmenterResult object represents the result of a eVision.EasyDeepLearning.EUnsupervisedSegmenter tool. More...

Public Member Functions
None	__init__ (self)
	Constructs a non-valid eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.
None	__init__ (self, eVision.EasyDeepLearning.EUnsupervisedSegmenterResult other)
	Copy constructor of an eVision.EasyDeepLearning.EUnsupervisedSegmenterResult object.
bool	IsComplete (self)
	Indicates whether the result is complete and ready to be used.
bool	IsValid (self)
	Indicates whether the result was produced by a eVision.EasyDeepLearning.EUnsupervisedSegmenter object.
bool	IsDefective (self)
	Indicates whether the result is defective/not good based on the threshold of the unsupervised segmentation tool that produced the result (see eVision.EasyDeepLearning.EUnsupervisedSegmenter.ClassificationThreshold).
bool	IsGood (self)
	Indicates whether the result is good based on the threshold of the unsupervised segmentation tool that produced the result (see eVision.EasyDeepLearning.EUnsupervisedSegmenter.ClassificationThreshold).
None	Draw (self, eVision.EDrawAdapter|None drawAdapter, float zoomX, float zoomY, float panX, float panY)
	Draws the segmentation (with transparency).
None	Draw (self, eVision.EDrawAdapter|None drawAdapter, float zoomX, float zoomY, float panX)
	Draws the segmentation (with transparency).
None	Draw (self, eVision.EDrawAdapter|None drawAdapter, float zoomX, float zoomY)
	Draws the segmentation (with transparency).
None	Draw (self, eVision.EDrawAdapter|None drawAdapter, float zoomX)
	Draws the segmentation (with transparency).
None	Draw (self, eVision.EDrawAdapter|None drawAdapter)
	Draws the segmentation (with transparency).
float	ClassificationScore (self)
	Score that is used for classification of the image.
float	Error (self)
	Error of the image.
eVision.EImageBW8	SegmentationMap (self)
	Returns the segmentation map.
eVision.ERegion	Region (self)
	Returns the segmented region.

Detailed Description

An eVision.EasyDeepLearning.EUnsupervisedSegmenterResult object represents the result of a eVision.EasyDeepLearning.EUnsupervisedSegmenter tool.

Remarks:

License(s): BasicTypes

Constructor & Destructor Documentation

__init__() [1/2]

None eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.__init__

(

self

)

Constructs a non-valid eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.

__init__() [2/2]

None eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.__init__	(		self,
		eVision.EasyDeepLearning.EUnsupervisedSegmenterResult	other )

Copy constructor of an eVision.EasyDeepLearning.EUnsupervisedSegmenterResult object.

Parameters

other

Reference to the eVision.EasyDeepLearning.EUnsupervisedSegmenterResult object that should be copied

Member Function Documentation

ClassificationScore()

float eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.ClassificationScore

(

self

)

Score that is used for classification of the image.

Remarks:

The classification score is the value which is compared to the classification threshold of the eVision.EasyDeepLearning.EUnsupervisedSegmenter to decide whether the corresponding image is good or defective.

Draw() [1/5]

None eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.Draw	(		self,
		eVision.EDrawAdapter | None	drawAdapter )

Draws the segmentation (with transparency).

To indicate the importance of the defect at a given pixel, the segmentation is drawn using a gradient going from yellow (least important) to red (maximum importance) (See eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.SegmentationMap).

Parameters

drawAdapter

A pointer to an eVision.EDrawAdapter (like the eVision.EWindowsDrawAdapter).

Draw() [2/5]

None eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.Draw	(		self,
		eVision.EDrawAdapter | None	drawAdapter,
		float	zoomX )

Draws the segmentation (with transparency).

To indicate the importance of the defect at a given pixel, the segmentation is drawn using a gradient going from yellow (least important) to red (maximum importance) (See eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.SegmentationMap).

Parameters

drawAdapter	A pointer to an eVision.EDrawAdapter (like the eVision.EWindowsDrawAdapter).
zoomX	Horizontal zooming factor. A value greater than 1 means zoom in. By default, true scale is used.

Draw() [3/5]

None eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.Draw	(		self,
		eVision.EDrawAdapter | None	drawAdapter,
		float	zoomX,
		float	zoomY )

Draws the segmentation (with transparency).

To indicate the importance of the defect at a given pixel, the segmentation is drawn using a gradient going from yellow (least important) to red (maximum importance) (See eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.SegmentationMap).

Parameters

drawAdapter	A pointer to an eVision.EDrawAdapter (like the eVision.EWindowsDrawAdapter).
zoomX	Horizontal zooming factor. A value greater than 1 means zoom in. By default, true scale is used.
zoomY	Vertical zooming factor. A value greater than 1 means zoom in. If set to 0, the default value, the horizontal zooming factor is used instead, so as to provide isotropic zooming.

Draw() [4/5]

None eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.Draw	(		self,
		eVision.EDrawAdapter | None	drawAdapter,
		float	zoomX,
		float	zoomY,
		float	panX )

Draws the segmentation (with transparency).

To indicate the importance of the defect at a given pixel, the segmentation is drawn using a gradient going from yellow (least important) to red (maximum importance) (See eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.SegmentationMap).

Parameters

drawAdapter	A pointer to an eVision.EDrawAdapter (like the eVision.EWindowsDrawAdapter).
zoomX	Horizontal zooming factor. A value greater than 1 means zoom in. By default, true scale is used.
zoomY	Vertical zooming factor. A value greater than 1 means zoom in. If set to 0, the default value, the horizontal zooming factor is used instead, so as to provide isotropic zooming.
panX	Horizontal panning value expressed in pixels. By default, no panning occurs.

Draw() [5/5]

None eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.Draw	(		self,
		eVision.EDrawAdapter | None	drawAdapter,
		float	zoomX,
		float	zoomY,
		float	panX,
		float	panY )

Draws the segmentation (with transparency).

To indicate the importance of the defect at a given pixel, the segmentation is drawn using a gradient going from yellow (least important) to red (maximum importance) (See eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.SegmentationMap).

Parameters

drawAdapter	A pointer to an eVision.EDrawAdapter (like the eVision.EWindowsDrawAdapter).
zoomX	Horizontal zooming factor. A value greater than 1 means zoom in. By default, true scale is used.
zoomY	Vertical zooming factor. A value greater than 1 means zoom in. If set to 0, the default value, the horizontal zooming factor is used instead, so as to provide isotropic zooming.
panX	Horizontal panning value expressed in pixels. By default, no panning occurs.
panY	Vertical panning value expressed in pixels. By default, no panning occurs.

Error()

float eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.Error

(

self

)

Error of the image.

Remarks:

The error is the quantity that is minimized on good images during training.

IsComplete()

bool eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.IsComplete

(

self

)

Indicates whether the result is complete and ready to be used.

IsDefective()

bool eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.IsDefective

(

self

)

Indicates whether the result is defective/not good based on the threshold of the unsupervised segmentation tool that produced the result (see eVision.EasyDeepLearning.EUnsupervisedSegmenter.ClassificationThreshold).

IsGood()

bool eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.IsGood

(

self

)

Indicates whether the result is good based on the threshold of the unsupervised segmentation tool that produced the result (see eVision.EasyDeepLearning.EUnsupervisedSegmenter.ClassificationThreshold).

IsValid()

bool eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.IsValid

(

self

)

Indicates whether the result was produced by a eVision.EasyDeepLearning.EUnsupervisedSegmenter object.

A default constructed eVision.EasyDeepLearning.EUnsupervisedSegmenterResult is not valid.

Region()

eVision.ERegion eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.Region

(

self

)

Returns the segmented region.

The segmented region corresponds to the pixels that have a value strictly higher than 0 in the segmentation map (see eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.SegmentationMap).

SegmentationMap()

eVision.EImageBW8 eVision.EasyDeepLearning.EUnsupervisedSegmenterResult.SegmentationMap

(

self

)

Returns the segmentation map.

The segmentation map is a grayscale image where all defective pixels have a value strictly higher than 0. The value of a pixel is proportional to the importance of the defect at that position.