Functions
cvbbool_t	ButterWorth (IMG imgIn, short highPass, double gain, cvbval_t addOffset, cvbval_t order, double cutOff, IMG &imgOut)
	Calculates a band pass filtered image. More...

cvbbool_t	FilterDilate (IMG imgIn, IMG &imgOut)
	Calculates a gray scale dilation of the input image with a 3x3 filter mask. More...

cvbbool_t	FilterEdge2x2 (IMG imgIn, IMG &imgOut)
	Creates an edge-filtered version of the input image. More...

cvbbool_t	FilterEdge3x3 (IMG imgIn, IMG &imgOut)
	Creates an edge-filtered version of the input image. More...

cvbbool_t	FilterErode (IMG imgIn, IMG &imgOut)
	Calculates a gray scale erosion of the input image with a 3x3 filter mask. More...

cvbbool_t	FilterLaplace3x3 (IMG imgIn, IMG &imgOut)
	Creates a Laplace-filtered image. More...

cvbbool_t	FilterLow2x2 (IMG imgIn, IMG &imgOut)
	Creates a blurred version of the input image. More...

cvbbool_t	FilterLow3x3 (IMG imgIn, IMG &imgOut)
	Creates a blurred version of the input image. More...

cvbbool_t	FilterLow5x5 (IMG imgIn, IMG &imgOut)
	Creates a blurred version of the input image. More...

cvbbool_t	FilterPyramid3x3 (IMG imgIn, IMG &imgOut)
	Applies a binomial low pass filter and downscales the image by a factor of 2 in width and height. More...

cvbbool_t	FilterPyramid4x4 (IMG imgIn, IMG &imgOut)
	Applies a binomial low pass filter and downscales the image by a factor of 2 in width and height. More...

cvbbool_t	FilterPyramid5x5 (IMG imgIn, IMG &imgOut)
	Applies a binomial low pass filter and downscales the image by a factor of 2 in width and height. More...

cvbbool_t	FilterSharpen (IMG imgIn, IMG &imgOut)
	Creates a sharpened version of the input image. More...

cvbbool_t	FilterUser2x2 (IMG imgIn, const TFilterDef &filterDef, IMG &imgOut)
	Creates an image by applying a 2x2 linear convolution filter to the input image. More...

cvbbool_t	FilterUser3x3 (IMG imgIn, const TFilterDef &filterDef, IMG &imgOut)
	Creates an image by applying a 3x3 linear convolution filter to the input image. More...

cvbbool_t	FilterUser5x5 (IMG imgIn, const TFilterDef &filterDef, IMG &imgOut)
	Creates an image by applying a 5x5 linear convolution filter to the input image. More...

Detailed Description

Function Documentation

◆ ButterWorth()

cvbbool_t ButterWorth	(	IMG	imgIn,
		short	highPass,
		double	gain,
		cvbval_t	addOffset,
		cvbval_t	order,
		double	cutOff,
		IMG &	imgOut
	)

Calculates a band pass filtered image.

This function uses the 1st or 2nd order Butterworth pass filter to process the input images. Butterworth filters can act as hight or low pass filters and have a fairly flat frequency response. They are not very computationally expensive, making them a good alternative to FFT-based band pass filtering.

original	Butterworth 1st order	Butterworth 2nd order

(both example images use the following parameters: highPass = TRUE, gain = 1.0, addOffset = 128, cutOff = 10)

Parameters

[in]	imgIn	Handle to the input image.
[in]	highPass	When set to `TRUE` the filter acts as a high pass filter, when set to `FALSE` it will act as a low pass filter.
[in]	gain	Magnitude of the frequency response.
[in]	addOffset	Offset to be added to the result.
[in]	order	Order of the butter-worth filter. Must be either 1 or 2.
[in]	cutOff	Cut-off parameter.
[out]	imgOut	Receives the handle of the newly created output image. Remember to dispose of the output image with `ReleaseObject` from the CVCImg.dll once it is no longer needed.

Returns

TRUE if the filter function succeeded, FALSE otherwise. Possible reasons for failure are:

imgIn does not refer to a valid image
imgIn has more than 8 bits per pixel in at least one plane
creation of the output image failed due to insufficient memory

◆ FilterDilate()

cvbbool_t FilterDilate	(	IMG	imgIn,
		IMG &	imgOut
	)

Calculates a gray scale dilation of the input image with a 3x3 filter mask.

The dilation is one of the basic morphology operations. It is based on scanning every location in the input image with a structural element (in this case simply a rectangle) and setting the corresponding pixel value in the output image to the brightest value found in the structural element.

Dilation effectively expands bright structures in the image (in this case, because a 3x3 mask is used, by 1 pixel in every direction). Dilation followed by an erosion (FilterErode) can be used for closing small holes dark pixels in bright objects) in objects.

original	dilated

Parameters

[in]	imgIn	Handle to the input image.
[out]	imgOut	Receives the handle of the newly created output image. Remember to dispose of the output image with `ReleaseObject` from the CVCImg.dll once it is no longer needed.

Returns

TRUE if the filter function succeeded, FALSE otherwise. Possible reasons for failure are:

imgIn does not refer to a valid image
imgIn has more than 8 bits per pixel in at least one plane
creation of the output image failed due to insufficient memory

See also: FilterErode

◆ FilterEdge2x2()

cvbbool_t FilterEdge2x2	(	IMG	imgIn,
		IMG &	imgOut
	)

Creates an edge-filtered version of the input image.

The output value for position (x,y) is calculated according to:

(effectively the maximum norm over the two Roberts filters). The results are clamped to the range [0...255].

original	low pass filtered

Parameters

[in]	imgIn	Handle to the input image.
[out]	imgOut	Receives the handle of the newly created output image. Remember to dispose of the output image with `ReleaseObject` from the CVCImg.dll once it is no longer needed.

Returns

TRUE if the filter function succeeded, FALSE otherwise. Possible reasons for failure are:

imgIn does not refer to a valid image
imgIn has more than 8 bits per pixel in at least one plane
creation of the output image failed due to insufficient memory

See also: FilterEdge3x3

◆ FilterEdge3x3()

cvbbool_t FilterEdge3x3	(	IMG	imgIn,
		IMG &	imgOut
	)

Creates an edge-filtered version of the input image.

The output value for position (x,y) is calculated according to:

The results are clamped to the range [0...255].

original	low pass filtered

Parameters

[in]	imgIn	Handle to the input image.
[out]	imgOut	Receives the handle of the newly created output image. Remember to dispose of the output image with `ReleaseObject` from the CVCImg.dll once it is no longer needed.

Returns

TRUE if the filter function succeeded, FALSE otherwise. Possible reasons for failure are:

imgIn does not refer to a valid image
imgIn has more than 8 bits per pixel in at least one plane
creation of the output image failed due to insufficient memory

See also: FilterEdge2x2

◆ FilterErode()

cvbbool_t FilterErode	(	IMG	imgIn,
		IMG &	imgOut
	)

Calculates a gray scale erosion of the input image with a 3x3 filter mask.

The erosion is one of the basic morphology operations. It is based on scanning every location in the input image with a structural element (in this case simply a rectangle) and setting the corresponding pixel value in the output image to the darkest value found in the structural element.

Erosion effectively expands dark structures in the image (in this case, because a 3x3 mask is used, by 1 pixel in every direction). Erosion followed by a dilation (FilterDilate) can be used for removing bright spots in otherwise dark objects.

original	dilated

Parameters

[in]	imgIn	Handle to the input image.
[out]	imgOut	Receives the handle of the newly created output image. Remember to dispose of the output image with `ReleaseObject` from the CVCImg.dll once it is no longer needed.

Returns

TRUE if the filter function succeeded, FALSE otherwise. Possible reasons for failure are:

imgIn does not refer to a valid image
imgIn has more than 8 bits per pixel in at least one plane
creation of the output image failed due to insufficient memory

See also: FilterDilate

◆ FilterLaplace3x3()

cvbbool_t FilterLaplace3x3	(	IMG	imgIn,
		IMG &	imgOut
	)

Creates a Laplace-filtered image.

The convolution kernel for the Laplace filter is

0	-1	0
-1	4	-1
0	-1	0

The results will be offset by 128 and clamped to the range [0...255].

original	Laplace-filtered

Parameters

[in]	imgIn	Handle to the input image.
[out]	imgOut	Receives the handle of the newly created output image. Remember to dispose of the output image with `ReleaseObject` from the CVCImg.dll once it is no longer needed.

Returns

TRUE if the filter function succeeded, FALSE otherwise. Possible reasons for failure are:

imgIn does not refer to a valid image
imgIn has more than 8 bits per pixel in at least one plane
creation of the output image failed due to insufficient memory

◆ FilterLow2x2()

cvbbool_t FilterLow2x2	(	IMG	imgIn,
		IMG &	imgOut
	)

Creates a blurred version of the input image.

This function blurs the image data of the input image with the convolution kernel

1/4	1/4
1/4	1/4

For a stronger blurring effect try FilterLow3x3 or FilterLow5x5.

original	low pass filtered

Parameters

[in]	imgIn	Handle to the input image.
[out]	imgOut	Receives the handle of the newly created output image. Remember to dispose of the output image with `ReleaseObject` from the CVCImg.dll once it is no longer needed.

Returns

TRUE if the filter function succeeded, FALSE otherwise. Possible reasons for failure are:

imgIn does not refer to a valid image
imgIn has more than 8 bits per pixel in at least one plane
creation of the output image failed due to insufficient memory

See also: FilterLow3x3, FilterLow5x5

◆ FilterLow3x3()

cvbbool_t FilterLow3x3	(	IMG	imgIn,
		IMG &	imgOut
	)

Creates a blurred version of the input image.

This function blurs the image data of the input image with the convolution kernel

1/16	2/16	1/16
2/16	4/16	2/16
1/16	2/16	1/16

For a weaker blurring effect try FilterLow2x2. For a stronger blurring effect try FilterLow5x5.

original	low pass filtered

Parameters

[in]	imgIn	Handle to the input image.
[out]	imgOut	Receives the handle of the newly created output image. Remember to dispose of the output image with `ReleaseObject` from the CVCImg.dll once it is no longer needed.

Returns

TRUE if the filter function succeeded, FALSE otherwise. Possible reasons for failure are:

imgIn does not refer to a valid image
imgIn has more than 8 bits per pixel in at least one plane
creation of the output image failed due to insufficient memory

See also: FilterLow2x2, FilterLow5x5

◆ FilterLow5x5()

cvbbool_t FilterLow5x5	(	IMG	imgIn,
		IMG &	imgOut
	)

Creates a blurred version of the input image.

This function blurs the image data of the input image with the convolution kernel

1	4	6	4	1
4	16	24	16	4
6	24	36	24	6
4	16	24	16	4
1	4	6	4	1

For a weaker blurring effect try FilterLow3x3 and FilterLow2x2.

original	low pass filtered

Parameters

[in]	imgIn	Handle to the input image.
[out]	imgOut	Receives the handle of the newly created output image. Remember to dispose of the output image with `ReleaseObject` from the CVCImg.dll once it is no longer needed.

Returns

TRUE if the filter function succeeded, FALSE otherwise. Possible reasons for failure are:

imgIn does not refer to a valid image
imgIn has more than 8 bits per pixel in at least one plane
creation of the output image failed due to insufficient memory

See also: FilterLow2x2, FilterLow3x3

◆ FilterPyramid3x3()

cvbbool_t FilterPyramid3x3	(	IMG	imgIn,
		IMG &	imgOut
	)

Applies a binomial low pass filter and downscales the image by a factor of 2 in width and height.

The low pass filter applied is effectively the same as implemented by the functions FilterLow3x3.

original	pyramid

Parameters

[in]	imgIn	Handle to the input image.
[out]	imgOut	Receives the handle of the newly created output image. Remember to dispose of the output image with `ReleaseObject` from the CVCImg.dll once it is no longer needed.

Returns

TRUE if the filter function succeeded, FALSE otherwise. Possible reasons for failure are:

imgIn does not refer to a valid image
imgIn has more than 8 bits per pixel in at least one plane
creation of the output image failed due to insufficient memory

See also: FilterPyramid4x4, FilterPyramid5x5

◆ FilterPyramid4x4()

cvbbool_t FilterPyramid4x4	(	IMG	imgIn,
		IMG &	imgOut
	)

Applies a binomial low pass filter and downscales the image by a factor of 2 in width and height.

The low pass filter applied is defined by the filter mask

1/64	3/64	3/64	1/64
3/64	9/64	9/64	3/64
3/64	9/64	9/64	3/64
1/64	3/64	3/64	1/64

original	pyramid

Parameters

[in]	imgIn	Handle to the input image.
[out]	imgOut	Receives the handle of the newly created output image. Remember to dispose of the output image with `ReleaseObject` from the CVCImg.dll once it is no longer needed.

Returns

TRUE if the filter function succeeded, FALSE otherwise. Possible reasons for failure are:

imgIn does not refer to a valid image
imgIn has more than 8 bits per pixel in at least one plane
creation of the output image failed due to insufficient memory

See also: FilterPyramid3x3, FilterPyramid5x5

◆ FilterPyramid5x5()

cvbbool_t FilterPyramid5x5	(	IMG	imgIn,
		IMG &	imgOut
	)

Applies a binomial low pass filter and downscales the image by a factor of 2 in width and height.

The low pass filter applied is effectively the same as implemented by the functions FilterLow5x5.

original	pyramid

Parameters

[in]	imgIn	Handle to the input image.
[out]	imgOut	Receives the handle of the newly created output image. Remember to dispose of the output image with `ReleaseObject` from the CVCImg.dll once it is no longer needed.

Returns

TRUE if the filter function succeeded, FALSE otherwise. Possible reasons for failure are:

imgIn does not refer to a valid image
imgIn has more than 8 bits per pixel in at least one plane
creation of the output image failed due to insufficient memory

See also: FilterPyramid3x3, FilterPyramid4x4

◆ FilterSharpen()

cvbbool_t FilterSharpen	(	IMG	imgIn,
		IMG &	imgOut
	)

Creates a sharpened version of the input image.

Sharpened images are effectively high pass images, in this case created with the convolution kernel

0	-1	0
-1	5	-1
0	-1	0

The results are clamped to the range [0...255]. Sharpened images may show more detail, the sharpen filter also amplifies noise (and is therefore usually not a good choice for preprocessing images that are going to be used with Minos).

original	sharpened

Parameters

[in]	imgIn	Handle to the input image.
[out]	imgOut	Receives the handle of the newly created output image. Remember to dispose of the output image with `ReleaseObject` from the CVCImg.dll once it is no longer needed.

Returns

TRUE if the filter function succeeded, FALSE otherwise. Possible reasons for failure are:

imgIn does not refer to a valid image
imgIn has more than 8 bits per pixel in at least one plane
creation of the output image failed due to insufficient memory

◆ FilterUser2x2()

cvbbool_t FilterUser2x2	(	IMG	imgIn,
		const TFilterDef &	filterDef,
		IMG &	imgOut
	)

The 2x2 convolution kernel needs to be specified as a TFilterDef structure. In this case, only the central members

c0000	c0R00
c000B	c0R0B

and Offst will be evaluated. Result values will be clamped to the range [0...255].

Attention: The Minos filter functions expect the coefficients (but not the Offst member!) in units of 1/1000 so e.g., 1500 means a coefficient value of 1.5

Parameters

[in]	imgIn	Handle to the input image.
[in]	filterDef	Filter definition. See TFilterDef.
[out]	imgOut	Receives the handle of the newly created output image. Remember to dispose of the output image with `ReleaseObject` from the CVCImg.dll once it is no longer needed.

Returns

TRUE if the filter function succeeded, FALSE otherwise. Possible reasons for failure are:

imgIn does not refer to a valid image
imgIn has more than 8 bits per pixel in at least one plane
creation of the output image failed due to insufficient memory

See also: FilterUser3x3, FilterUser5x5

◆ FilterUser3x3()

cvbbool_t FilterUser3x3	(	IMG	imgIn,
		const TFilterDef &	filterDef,
		IMG &	imgOut
	)

The 3x3 convolution kernel needs to be specified as a TFilterDef structure. In this case, only the central members

c0L0T	c000T	c0R0T
c0L00	c0000	c0R00
c0L0B	c000B	c0R0B

and Offst will be evaluated. Result values will be clamped to the range [0...255].

Attention: The Minos filter functions expect the coefficients (but not the Offst member!) in units of 1/1000 so e.g., 1500 means a coefficient value of 1.5

Parameters

[in]	imgIn	Handle to the input image.
[in]	filterDef	Filter definition. See TFilterDef.
[out]	imgOut	Receives the handle of the newly created output image. Remember to dispose of the output image with `ReleaseObject` from the CVCImg.dll once it is no longer needed.

Returns

TRUE if the filter function succeeded, FALSE otherwise. Possible reasons for failure are:

imgIn does not refer to a valid image
imgIn has more than 8 bits per pixel in at least one plane
creation of the output image failed due to insufficient memory

See also: FilterUser2x2, FilterUser5x5

◆ FilterUser5x5()

cvbbool_t FilterUser5x5	(	IMG	imgIn,
		const TFilterDef &	filterDef,
		IMG &	imgOut
	)

The 5x5 convolution kernel needs to be specified as a TFilterDef structure. In this case all coefficient members and Offst will be evaluated. Result values will be clamped to the range [0...255].

Attention: The Minos filter functions expect the coefficients (but not the Offst member!) in units of 1/1000 so e.g., 1500 means a coefficient value of 1.5

Parameters

[in]	imgIn	Handle to the input image.
[in]	filterDef	Filter definition. See TFilterDef.
[out]	imgOut	Receives the handle of the newly created output image. Remember to dispose of the output image with `ReleaseObject` from the CVCImg.dll once it is no longer needed.

Returns

TRUE if the filter function succeeded, FALSE otherwise. Possible reasons for failure are:

imgIn does not refer to a valid image
imgIn has more than 8 bits per pixel in at least one plane
creation of the output image failed due to insufficient memory

See also: FilterUser2x2, FilterUser3x3

Functions

Detailed Description

Function Documentation

◆ ButterWorth()

◆ FilterDilate()

◆ FilterEdge2x2()

◆ FilterEdge3x3()

◆ FilterErode()

◆ FilterLaplace3x3()

◆ FilterLow2x2()

◆ FilterLow3x3()

◆ FilterLow5x5()

◆ FilterPyramid3x3()

◆ FilterPyramid4x4()

◆ FilterPyramid5x5()

◆ FilterSharpen()

◆ FilterUser2x2()

◆ FilterUser3x3()

◆ FilterUser5x5()