Image Segmentation Functions

Functions
cvbres_t	LabelMarker (IMG ImgMarker, cvbval_t MinLabel, cvbval_t MaxLabel, TSegmentationNorm Norm, cvbdim_t &NumLabels, IMG &ImgLabelled)
	This method labels markers in the source image ImgMarker with different integer values. More...
cvbres_t	SegmentGradient (IMG ImgIn, TSegmentationNorm Norm, TSegmentationBorder Border, IMG ImgMarker, cvbdim_t Threshold, IMG &ImgOut)
	This method performs segmentation by region growing to the least gradient direction. More...
cvbres_t	SegmentWatershed (IMG ImgIn, TWatershedNorm Norm, TWatershedSegmentation Segmentation, TSegmentationBorder Border, IMG ImgMarker, cvbdim_t Threshold, IMG &ImgOut)
	This method performs watershed segmentation with markers. More...

Detailed Description

Function Documentation

LabelMarker()

cvbres_t LabelMarker	(	IMG	ImgMarker,
		cvbval_t	MinLabel,
		cvbval_t	MaxLabel,
		TSegmentationNorm	Norm,
		cvbdim_t &	NumLabels,
		IMG &	ImgLabelled
	)

This method labels markers in the source image ImgMarker with different integer values.

Each connected set of non-zero image pixels is treated as a separate marker. All pixels belonging to the same marker are set to the same value from the interval [MinLabel, MaxLabel]. Two markers can be labelled with the same value if the number of connected components exceeds MinLabel - MaxLabel + 1.

Note

This functions works with 8-bit and 16-bit images.

Attention

Images with more than 1 plane will be split for computation in single plane images. All image planes are treated independently to each other.

Parameters

[in]	ImgMarker	Handle of input image to be labelled.
[in]	MinLabel	Minimum value of the marker label (0 < MinLabel <= MaxLabel).
[in]	MaxLabel	Maximum value of the marker label (MinLabel <= MaxLabel < 255 for 8-bit markers, and MinLabel <= MaxLabel < 65535 for 16-bit markers). values see TWatershedSegmentation.
[in]	Norm	Norm to form the mask for marker propagation. Either 4- or 8-connectivity can be used depending on the norm type. See TSegmentationNorm for possible values.
[out]	NumLabels	Number of labelled regions.
[out]	ImgLabelled	Image handle of labelled image.

Returns

0 : OK
< 0: an error occurred.

Supported platforms:

Win32
Win64

Related Topics:

TSegmentationNorm, SegmentWatershed

SegmentGradient()

cvbres_t SegmentGradient	(	IMG	ImgIn,
		TSegmentationNorm	Norm,
		TSegmentationBorder	Border,
		IMG	ImgMarker,
		cvbdim_t	Threshold,
		IMG &	ImgOut
	)

This method performs segmentation by region growing to the least gradient direction.

This function performs image segmentation by region growing with markers. Non-zero pixels of ImgMarker belong to initial image regions. Marker values propagate through the whole image in the direction of the least value of the absolute value of the image gradient.

Image segments are formed by groups of connected ImgMarker pixels with the same value.

Note

This function works only with 8-bit images.

Attention

For 3-channel image the gradient is calculated as the maximum of channel gradients.

Parameters

[in]	ImgIn	Handle of input image to be segmented.
[in]	Norm	Controls the marker propagation connectivity. See TSegmentationNorm for possible values.
[in]	Border	Border to be applied to the segments. See TSegmentationBorder for possible values. Note: The pixels between image segments are marked with the value 254. The maximum possible value (255) is used internally and can not used to mark segments.
[in]	ImgMarker	Marker image which is needed to perform gradient segmentation. If NULL is specified, the marker image is created automatically using morphological reconstruction.
[in]	Threshold	Threshold which is applied for marker creation. This value is only used if ImgMarker is NULL.
[out]	ImgOut	Image handle of segmented image.

Returns

0 : OK
< 0: an error occurred.

Supported platforms:

Win32
Win64

Related Topics:

TSegmentationNorm, TSegmentationBorder, SegmentWatershed

SegmentWatershed()

cvbres_t SegmentWatershed	(	IMG	ImgIn,
		TWatershedNorm	Norm,
		TWatershedSegmentation	Segmentation,
		TSegmentationBorder	Border,
		IMG	ImgMarker,
		cvbdim_t	Threshold,
		IMG &	ImgOut
	)

This method performs watershed segmentation with markers.

The watershed segmentation is preferable for images with local minimums, for example, gradient images. Image markers generally correspond to the local minimums and can be created, for example, manually or using morphological reconstruction.

Non-zero pixels of ImgMarker image belong to water source markers. Marker values propagate through the whole image according to the watershed algorithm. Image segments are formed by groups of connected ImgMarker pixels with the same value.

Note

This function works only with 8-bit images.

Attention

Images with more than 1 plane will be split for computation in single plane images. All image planes are treated independently to each other.

Parameters

[in]	ImgIn	Handle of input image to be segmented.
[in]	Norm	Controls the marker propagation connectivity. See TWatershedNorm for more information.
[in]	Segmentation	Specifies the algorithm of segmentation. For possible values see TWatershedSegmentation.
[in]	Border	Border to be applied to the segments. See TSegmentationBorder for possible values.
[in]	ImgMarker	Marker image which is needed to perform watershed segmentation. If NULL is specified, the marker image is created automatically using morphological reconstruction.
[in]	Threshold	Threshold which is applied for marker creation. This value is only used if ImgMarker is NULL.
[out]	ImgOut	Image handle of segmented image.

Returns

0 : OK
< 0: an error occurred.

Supported platforms:

Win32
Win64

Related Topics:

TWatershedNorm, TWatershedSegmentation, TSegmentationBorder, SegmentGradient