create

create — create images in various ways

Stability Level

Stable, unless otherwise indicated

Includes

#include <vips/vips.h>

Description

These functions generate various test images. You can combine them with the arithmetic and rotate functions to build more complicated images.

The im_benchmark() operations are for testing the VIPS SMP system.

Functions

vips_black ()

int
vips_black (VipsImage **out,
            int width,
            int height,
            ...);

Optional arguments:

  • bands : output bands

Make a black unsigned char image of a specified size.

See also: vips_xyz(), vips_text(), vips_gaussnoise().

Parameters

out

output image.

[out]

width

output width

 

height

output height

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_xyz ()

int
vips_xyz (VipsImage **out,
          int width,
          int height,
          ...);

Optional arguments:

  • csize : size for third dimension

  • dsize : size for fourth dimension

  • esize : size for fifth dimension

Create a two-band uint32 image where the elements in the first band have the value of their x coordinate and elements in the second band have their y coordinate.

You can make any image where the value of a pixel is a function of its (x, y) coordinate by combining this operator with the arithmetic operators.

Set csize , dsize , esize to generate higher dimensions and add more bands. The extra dimensions are placed down the vertical axis. Use vips_grid() to change the layout.

See also: vips_grey(), vips_grid(), vips_identity().

Parameters

out

output image.

[out]

width

horizontal size

 

height

vertical size

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_grey ()

int
vips_grey (VipsImage **out,
           int width,
           int height,
           ...);

Optional arguments:

  • uchar : output a uchar image

Create a one-band float image with the left-most column zero and the right-most 1. Intermediate pixels are a linear ramp.

Set uchar to output a uchar image with the leftmost pixel 0 and the rightmost 255.

See also: vips_xyz(), vips_identity().

Parameters

out

output image.

[out]

width

image size

 

height

image size

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_gaussmat ()

int
vips_gaussmat (VipsImage **out,
               double sigma,
               double min_ampl,
               ...);

Optional arguments:

  • separable : generate a separable gaussian

  • precision : VipsPrecision for out

Creates a circularly symmetric Gaussian image of radius sigma . The size of the mask is determined by the variable min_ampl ; if for instance the value .1 is entered this means that the produced mask is clipped at values less than 10 percent of the maximum amplitude.

The program uses the following equation:

H(r) = exp( -(r * r) / (2 * sigma * sigma ) )

The generated image has odd size and its maximum value is normalised to 1.0, unless precision is VIPS_PRECISION_INTEGER.

If separable is set, only the centre horizontal is generated. This is useful for separable convolutions.

If precision is VIPS_PRECISION_INTEGER, an integer gaussian is generated. This is useful for integer convolutions.

"scale" is set to the sum of all the mask elements.

See also: vips_logmat(), vips_conv().

Parameters

out

output image.

[out]

sigma

standard deviation of mask

 

min_ampl

minimum amplitude

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_logmat ()

int
vips_logmat (VipsImage **out,
             double sigma,
             double min_ampl,
             ...);

Optional arguments:

  • separable : generate a separable mask

  • precision : VipsPrecision for out

Creates a circularly symmetric Laplacian of Gaussian mask of radius sigma . The size of the mask is determined by the variable min_ampl ; if for instance the value .1 is entered this means that the produced mask is clipped at values within 10 persent of zero, and where the change between mask elements is less than 10%.

The program uses the following equation: (from Handbook of Pattern Recognition and image processing by Young and Fu, AP 1986 pages 220-221):

H(r) = (1 / (2 * M_PI * s4)) * (2 - (r2 / s2)) * exp(-r2 / (2 * s2))

where s2 = sigma * sigma , s4 = s2 * s2, r2 = r * r.

The generated mask has odd size and its maximum value is normalised to 1.0, unless precision is VIPS_PRECISION_INTEGER.

If separable is set, only the centre horizontal is generated. This is useful for separable convolutions.

If precision is VIPS_PRECISION_INTEGER, an integer mask is generated. This is useful for integer convolutions.

"scale" is set to the sum of all the mask elements.

See also: vips_gaussmat(), vips_conv().

Parameters

out

output image.

[out]

sigma

standard deviation of mask

 

min_ampl

minimum amplitude

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_text ()

int
vips_text (VipsImage **out,
           const char *text,
           ...);

Optional arguments:

  • font : gchararray, font to render with

  • fontfile : gchararray, load this font file

  • width : gint, image should be no wider than this many pixels

  • height : gint, image should be no higher than this many pixels

  • align : VipsAlign, left/centre/right alignment

  • dpi : gint, render at this resolution

  • autofit_dpi : gint, read out auto-fitted DPI

  • spacing : gint, space lines by this in points

Draw the string text to an image. out is a one-band 8-bit unsigned char image, with 0 for no text and 255 for text. Values inbetween are used for anti-aliasing.

text is the text to render as a UTF-8 string. It can contain Pango markup, for example "<i>The</i>Guardian".

font is the font to render with, as a fontconfig name. Examples might be "sans 12" or perhaps "bitstream charter bold 10".

You can specify a font to load with fontfile . You'll need to also set the name of the font with font .

width is the number of pixels to word-wrap at. Lines of text wider than this will be broken at word bounaries. align can be used to set the alignment style for multi-line text. Note that the output image can be wider than width if there are no word breaks.

height is the maximum number of pixels high the generated text can be. This only takes effect when dpi is not set, and width is set, making a box. In this case, vips_text() will search for a dpi which will just fit the text into width and height .

You can use autofit_dpi to read out the DPI selected by auto fit.

dpi sets the resolution to render at. "sans 12" at 72 dpi draws characters approximately 12 pixels high.

spacing sets the line spacing, in points. It would typically be something like font size times 1.2.

See also: vips_xyz(), vips_text(), vips_gaussnoise().

Parameters

out

output image.

[out]

text

utf-8 text string to render

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_gaussnoise ()

int
vips_gaussnoise (VipsImage **out,
                 int width,
                 int height,
                 ...);

Optional arguments:

  • mean : mean of generated pixels

  • sigma : standard deviation of generated pixels

Make a one band float image of gaussian noise with the specified distribution. The noise distribution is created by averaging 12 random numbers with the appropriate weights.

See also: vips_black(), vips_xyz(), vips_text().

Parameters

out

output image.

[out]

width

output width

 

height

output height

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_eye ()

int
vips_eye (VipsImage **out,
          int width,
          int height,
          ...);

Optional arguments:

  • factor : maximum spatial frequency

  • uchar : output a uchar image

Create a test pattern with increasing spatial frequence in X and amplitude in Y. factor should be between 0 and 1 and determines the maximum spatial frequency.

Set uchar to output a uchar image.

See also: vips_zone().

Parameters

out

output image.

[out]

width

image size

 

height

image size

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_sines ()

int
vips_sines (VipsImage **out,
            int width,
            int height,
            ...);

Optional arguments:

  • hfreq : horizontal frequency

  • vreq : vertical frequency

  • uchar : output a uchar image

Creates a float one band image of the a sine waveform in two dimensions.

The number of horizontal and vertical spatial frequencies are determined by the variables hfreq and vfreq respectively. The function is useful for creating displayable sine waves and square waves in two dimensions.

If horfreq and verfreq are integers the resultant image is periodical and therfore the Fourier transform does not present spikes

Pixels are normally in [-1, +1], set uchar to output [0, 255].

See also: vips_grey(), vips_xyz().

Parameters

out

output image.

[out]

width

image size

 

height

image size

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_zone ()

int
vips_zone (VipsImage **out,
           int width,
           int height,
           ...);

Optional arguments:

  • uchar : output a uchar image

Create a one-band image of a zone plate.

Pixels are normally in [-1, +1], set uchar to output [0, 255].

See also: vips_eye(), vips_xyz().

Parameters

out

output image.

[out]

width

image size

 

height

image size

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_identity ()

int
vips_identity (VipsImage **out,
               ...);

Optional arguments:

  • bands : number of bands to create

  • ushort : TRUE for an unsigned short identity

  • size : number of LUT elements for a ushort image

Creates an identity lookup table, ie. one which will leave an image unchanged when applied with vips_maplut(). Each entry in the table has a value equal to its position.

Use the arithmetic operations on these tables to make LUTs representing arbitrary functions.

Normally LUTs are 8-bit. Set ushort to create a 16-bit table.

Normally 16-bit tables have 65536 entries. You can set this smaller with size .

See also: vips_xyz(), vips_maplut().

Parameters

out

output image.

[out]

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_buildlut ()

int
vips_buildlut (VipsImage *in,
               VipsImage **out,
               ...);

This operation builds a lookup table from a set of points. Intermediate values are generated by piecewise linear interpolation. The lookup table is always of type VIPS_FORMAT_DOUBLE, use vips_cast() to change it to the type you need.

For example, consider this 2 x 2 matrix of (x, y) coordinates:

0 0
255 100

We then generate:

Index Value
0 0
1 0.4
... etc. by linear interpolation
255 100

This is then written as the output image, with the left column giving the index in the image to place the value.

The (x, y) points don't need to be sorted: we do that. You can have several Ys, each becomes a band in the output LUT. You don't need to start at zero, any integer will do, including negatives.

See also: vips_identity(), vips_invertlut(), vips_cast(), vips_maplut().

[method]

Parameters

in

input matrix

 

out

output image.

[out]

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_invertlut ()

int
vips_invertlut (VipsImage *in,
                VipsImage **out,
                ...);

Optional arguments:

  • size : generate this much

Given a mask of target values and real values, generate a LUT which will map reals to targets. Handy for linearising images from measurements of a colour chart. All values in [0,1]. Piecewise linear interpolation, extrapolate head and tail to 0 and 1.

Eg. input like this:

4 3    
0.1 0.2 0.3 0.1
0.2 0.4 0.4 0.2
0.7 0.5 0.6 0.3

Means a patch with 10% reflectance produces an image with 20% in channel 1, 30% in channel 2, and 10% in channel 3, and so on.

Inputs don't need to be sorted (we do that). Generate any precision LUT, default to 256 elements.

It won't work too well for non-monotonic camera responses (we should fix this). Interpolation is simple piecewise linear; we ought to do something better really.

See also: vips_buildlut().

[method]

Parameters

in

input mask

 

out

output LUT.

[out]

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_tonelut ()

int
vips_tonelut (VipsImage **out,
              ...);

Optional arguments:

  • in_max : input range

  • out_max : output range

  • Lb : black-point [0-100]

  • Lw : white-point [0-100]

  • Ps : shadow point (eg. 0.2)

  • Pm : mid-tone point (eg. 0.5)

  • Ph : highlight point (eg. 0.8)

  • S : shadow adjustment (+/- 30)

  • M : mid-tone adjustment (+/- 30)

  • H : highlight adjustment (+/- 30)

vips_tonelut() generates a tone curve for the adjustment of image levels. It is mostly designed for adjusting the L* part of a LAB image in a way suitable for print work, but you can use it for other things too.

The curve is an unsigned 16-bit image with (in_max + 1) entries, each in the range [0, out_max ].

Lb , Lw are expressed as 0-100, as in LAB colour space. You specify the scaling for the input and output images with the in_max and out_max parameters.

Parameters

out

output image.

[out]

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_mask_ideal ()

int
vips_mask_ideal (VipsImage **out,
                 int width,
                 int height,
                 double frequency_cutoff,
                 ...);

Optional arguments:

  • nodc : don't set the DC pixel

  • reject : invert the filter sense

  • optical : coordinates in optical space

  • uchar : output a uchar image

Make an ideal high- or low-pass filter, that is, one with a sharp cutoff positioned at frequency_cutoff , where frequency_cutoff is in the range 0 - 1.

This operation creates a one-band float image of the specified size. The image has values in the range [0, 1] and is typically used for multiplying against frequency domain images to filter them. Masks are created with the DC component at (0, 0). The DC pixel always has the value 1.0.

Set nodc to not set the DC pixel.

Set optical to position the DC component in the centre of the image. This makes the mask suitable for multiplying against optical Fourier transforms. See vips_wrap().

Set reject to invert the sense of the filter. For example, low-pass becomes low-reject.

Set uchar to output an 8-bit unsigned char image rather than a float image. In this case, pixels are in the range [0 - 255].

See also: vips_mask_ideal(), vips_mask_ideal_ring(), vips_mask_ideal_band(), vips_mask_butterworth(), vips_mask_butterworth_ring(), vips_mask_butterworth_band(), vips_mask_gaussian(), vips_mask_gaussian_ring(), vips_mask_gaussian_band().

Parameters

out

output image.

[out]

width

image size

 

height

image size

 

frequency_cutoff

threshold at which filter ends

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_mask_ideal_ring ()

int
vips_mask_ideal_ring (VipsImage **out,
                      int width,
                      int height,
                      double frequency_cutoff,
                      double ringwidth,
                      ...);

Optional arguments:

  • nodc : don't set the DC pixel

  • reject : invert the filter sense

  • optical : coordinates in optical space

  • uchar : output a uchar image

Make an ideal ring-pass or ring-reject filter, that is, one with a sharp ring positioned at frequency_cutoff of width width , where frequency_cutoff and width are expressed as the range 0 - 1.

See also: vips_mask_ideal().

Parameters

out

output image.

[out]

width

image size

 

height

image size

 

frequency_cutoff

threshold at which filter ends

 

ringwidth

ring width

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_mask_ideal_band ()

int
vips_mask_ideal_band (VipsImage **out,
                      int width,
                      int height,
                      double frequency_cutoff_x,
                      double frequency_cutoff_y,
                      double radius,
                      ...);

Optional arguments:

  • nodc : don't set the DC pixel

  • reject : invert the filter sense

  • optical : coordinates in optical space

  • uchar : output a uchar image

Make an ideal band-pass or band-reject filter, that is, one with a sharp cutoff around the point frequency_cutoff_x , frequency_cutoff_y , of size radius .

See also: vips_mask_ideal().

Parameters

out

output image.

[out]

width

image size

 

height

image size

 

frequency_cutoff_x

position of band

 

frequency_cutoff_y

position of band

 

radius

size of band

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_mask_butterworth ()

int
vips_mask_butterworth (VipsImage **out,
                       int width,
                       int height,
                       double order,
                       double frequency_cutoff,
                       double amplitude_cutoff,
                       ...);

Optional arguments:

  • nodc : don't set the DC pixel

  • reject : invert the filter sense

  • optical : coordinates in optical space

  • uchar : output a uchar image

Make an butterworth high- or low-pass filter, that is, one with a variable, smooth transition positioned at frequency_cutoff , where frequency_cutoff is in range 0 - 1. The shape of the curve is controlled by order --- higher values give a sharper transition. See Gonzalez and Wintz, Digital Image Processing, 1987.

See also: vips_mask_ideal().

Parameters

out

output image.

[out]

width

image size

 

height

image size

 

order

filter order

 

frequency_cutoff

frequency threshold

 

amplitude_cutoff

amplitude threshold

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_mask_butterworth_ring ()

int
vips_mask_butterworth_ring (VipsImage **out,
                            int width,
                            int height,
                            double order,
                            double frequency_cutoff,
                            double amplitude_cutoff,
                            double ringwidth,
                            ...);

Optional arguments:

  • nodc : don't set the DC pixel

  • reject : invert the filter sense

  • optical : coordinates in optical space

  • uchar : output a uchar image

Make a butterworth ring-pass or ring-reject filter, that is, one with a variable, smooth transition positioned at frequency_cutoff of width width , where frequency_cutoff is in the range 0 - 1. The shape of the curve is controlled by order --- higher values give a sharper transition. See Gonzalez and Wintz, Digital Image Processing, 1987.

See also: vips_mask_ideal().

Parameters

out

output image.

[out]

width

image size

 

height

image size

 

order

filter order

 

frequency_cutoff

frequency threshold

 

amplitude_cutoff

amplitude threshold

 

ringwidth

ringwidth

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_mask_butterworth_band ()

int
vips_mask_butterworth_band (VipsImage **out,
                            int width,
                            int height,
                            double order,
                            double frequency_cutoff_x,
                            double frequency_cutoff_y,
                            double radius,
                            double amplitude_cutoff,
                            ...);

Optional arguments:

  • nodc : don't set the DC pixel

  • reject : invert the filter sense

  • optical : coordinates in optical space

  • uchar : output a uchar image

Make an butterworth band-pass or band-reject filter, that is, one with a variable, smooth transition positioned at frequency_cutoff_x , frequency_cutoff_y , of radius radius . The shape of the curve is controlled by order --- higher values give a sharper transition. See Gonzalez and Wintz, Digital Image Processing, 1987.

See also: vips_mask_ideal().

Parameters

out

output image.

[out]

width

image size

 

height

image size

 

order

filter order

 

frequency_cutoff_x

band position

 

frequency_cutoff_y

band position

 

radius

band radius

 

amplitude_cutoff

amplitude threshold

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_mask_gaussian ()

int
vips_mask_gaussian (VipsImage **out,
                    int width,
                    int height,
                    double frequency_cutoff,
                    double amplitude_cutoff,
                    ...);

Optional arguments:

  • nodc : don't set the DC pixel

  • reject : invert the filter sense

  • optical : coordinates in optical space

  • uchar : output a uchar image

Make a gaussian high- or low-pass filter, that is, one with a variable, smooth transition positioned at frequency_cutoff .

See also: vips_mask_ideal().

Parameters

out

output image.

[out]

width

image size

 

height

image size

 

frequency_cutoff

frequency threshold

 

amplitude_cutoff

amplitude threshold

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_mask_gaussian_ring ()

int
vips_mask_gaussian_ring (VipsImage **out,
                         int width,
                         int height,
                         double frequency_cutoff,
                         double amplitude_cutoff,
                         double ringwidth,
                         ...);

Optional arguments:

  • nodc : don't set the DC pixel

  • reject : invert the filter sense

  • optical : coordinates in optical space

  • uchar : output a uchar image

Make a gaussian ring-pass or ring-reject filter, that is, one with a variable, smooth transition positioned at frequency_cutoff of width ringwidth .

See also: vips_mask_ideal().

Parameters

out

output image.

[out]

width

image size

 

height

image size

 

frequency_cutoff

frequency threshold

 

amplitude_cutoff

amplitude threshold

 

ringwidth

ringwidth

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_mask_gaussian_band ()

int
vips_mask_gaussian_band (VipsImage **out,
                         int width,
                         int height,
                         double frequency_cutoff_x,
                         double frequency_cutoff_y,
                         double radius,
                         double amplitude_cutoff,
                         ...);

Optional arguments:

  • nodc : don't set the DC pixel

  • reject : invert the filter sense

  • optical : coordinates in optical space

  • uchar : output a uchar image

Make a gaussian band-pass or band-reject filter, that is, one with a variable, smooth transition positioned at frequency_cutoff_x , frequency_cutoff_y , of radius radius .

See also: vips_mask_ideal().

Parameters

out

output image.

[out]

width

image size

 

height

image size

 

frequency_cutoff_x

band position

 

frequency_cutoff_y

band position

 

radius

band radius

 

amplitude_cutoff

amplitude threshold

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_mask_fractal ()

int
vips_mask_fractal (VipsImage **out,
                   int width,
                   int height,
                   double fractal_dimension,
                   ...);

Optional arguments:

  • nodc : don't set the DC pixel

  • reject : invert the filter sense

  • optical : coordinates in optical space

  • uchar : output a uchar image

This operation should be used to create fractal images by filtering the power spectrum of Gaussian white noise. See vips_gaussnoise().

See also: vips_mask_ideal().

Parameters

out

output image.

[out]

width

image size

 

height

image size

 

fractal_dimension

fractal dimension

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_fractsurf ()

int
vips_fractsurf (VipsImage **out,
                int width,
                int height,
                double fractal_dimension,
                ...);

Generate an image of size width by height and fractal dimension fractal_dimension . The dimension should be between 2 and 3.

See also: vips_gaussnoise(), vips_mask_fractal().

Parameters

out

output image.

[out]

width

output width

 

height

output height

 

fractal_dimension

fractal dimension

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_worley ()

int
vips_worley (VipsImage **out,
             int width,
             int height,
             ...);

Optional arguments:

  • cell_size : gint, size of Worley cells

Create a one-band float image of Worley noise. See:

https://en.wikipedia.org/wiki/Worley_noise

Use cell_size to set the size of the cells from which the image is constructed. The default is 256 x 256.

If width and height are multiples of cell_size , the image will tessellate.

See also: vips_perlin(), vips_fractsurf(), vips_gaussnoise().

Parameters

out

output image.

[out]

width

horizontal size

 

height

vertical size

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error


vips_perlin ()

int
vips_perlin (VipsImage **out,
             int width,
             int height,
             ...);

Optional arguments:

  • cell_size : gint, size of Perlin cells

  • uchar : output a uchar image

Create a one-band float image of Perlin noise. See:

https://en.wikipedia.org/wiki/Perlin_noise

Use cell_size to set the size of the cells from which the image is constructed. The default is 256 x 256.

If width and height are multiples of cell_size , the image will tessellate.

Normally, output pixels are VIPS_FORMAT_FLOAT in the range [-1, +1]. Set uchar to output a uchar image with pixels in [0, 255].

See also: vips_worley(), vips_fractsurf(), vips_gaussnoise().

Parameters

out

output image.

[out]

width

horizontal size

 

height

vertical size

 

...

NULL-terminated list of optional named arguments

 

Returns

0 on success, -1 on error

Types and Values